Exploring the Role of Artificial Intelligence in Enhancing Telemedicine Accessibility,
Efficiency, and Healthcare Outcomes in Rural Canada

Amar Paramaraj
Ekta
University Canada West
MBAR 661: Academic Research Project
Professor Dr. Amit Kohli

Abstract
This paper evaluates the ability of the integration of artificial intelligence (AI) with telemedicine
to enhance access, efficiency, and outcomes of healthcare in rural Canada. Telemedicine is now
indispensable in eliminating geographical distance, complemented further by enhanced AIpowered diagnostics and predictive analytics, workflow automation. This narrative literature
review gives a thematic analysis of sources, utilizing data from Health Canada, CIHI, TELUS
Health, and peer-reviewed research articles. Thematic analysis revealed the consistent obstacles,
the most prevalent of which are the lack of broadband and shortage of providers, as well as the
emerging potential of AI as a valuable tool to address it. The key findings refer to the fact that AIenhanced triage systems (such as Babylon, an AI-powered triage system by TELUS Health) help
to minimize unnecessary in-person visits, remote patient monitoring with predictive analytics
helps to manage chronic diseases, and AI-assisted imaging interpretation is faster when unexpected
scenarios such as stroke care are required. According to economic analyses, the widespread use of
AI-enhanced telemedicine may save more than $5 billion annually due to benefits in terms of
reduced travelling, fewer emergency visits, and increased productivity. The paper draws the
conclusion that with the help of adequate infrastructure, well-balanced policy structures, and a
culturally appropriate approach to implementation, AI can become the driving force of a more
equitable, streamlined, and patient-centered paradigm of a rural health system in Canada.

Introduction

Rural Canada faces significant healthcare accessibility challenges, services are far from most
people and there are longer-lasting differences in health. About four million Canadians or 1 in 5,
live in rural areas, but only 200,000 of the country's 82,000 physicians practice in these places
(Canadian Institute for Health Information, 2023). As a result of this, many people have faced
delays in getting care, have to go further for medical help and still lack access to it. Telemedicine
which offers healthcare services remotely, has become very important for connecting rural and
urban health services. Thanks to virtual visits, remote checks and distant access to consultants,
telemedicine can eliminate distance-related issues for patients. Telemedicine increased in
popularity in Canada because physical distancing was necessary during the pandemic caused by
COVID-19. The share of virtual care increased since a sideways of all primary care visits only at
the beginning of 2020 and then declined to nearly 40% at the beginning of 2021 (Canada Health
Infoway, 2021). Although telehealth has become the norm in a short period of time, it has soon
become evident that there are challenges, primarily in rural locations.
Methodology
Research Design
In this work, the narrative literature review approach was utilized to review the available evidence
of AI-enhanced telemedicine in rural Canada. A narrative method was chosen due to the
interdisciplinary character of this emerging field and to enable thematic analysis of a variety of
types of studies.

Search Strategy

Literature searches were conducted across PubMed, EMBASE, CIHI databases, and grey literature
from Health Canada and provincial agencies between November 2019 and May 2025. Search
terms combined "artificial intelligence," "telemedicine," "rural healthcare," and "Canada" using
Boolean operators.
Analytical Framework
Due to various methodological considerations in the newly developed field of AI-enhanced
telemedicine in rural Canada, a narrative literature review methodology was used in this study and
not a systematic review with formal hypothesis testing. First, the fast-growing nature of artificial
intelligence technologies in healthcare implied that evidence available in this area is diverse in
study design, implementation setting, and outcome measurements that could not be synthesized
through a meta-analysis method or tested with a statistical hypothesis. Second, the interdisciplinary
nature of this study (healthcare delivery, technology adoption, economic analysis and policy
frameworks) necessitated a versatile method of analysis that would be able to integrate qualitative
findings, case studies, pilot programs and government reports with peer-reviewed quantitative
research. Third, the small number of randomized controlled studies that specifically studied AItelemedicine interventions in Canadian rural settings rendered formal statistical comparisons
inappropriate, because the evidence base involves mostly observational studies, descriptive
studies, and early-stage implementation reports. Last, the research aims were thematic as they were
aimed at discovering thematic patterns, studying complex socio-technical relationships, and
making policy suggestions instead of testing single cause and effect relationships that would be
the subjects of null and alternative hypotheses. Hence, this narrative methodology was considered
the most suitable to generalize the heterogeneous evidence at their disposal and make holistic
information to guide the future research directions and policy formations in this emerging area.

Telemedicine Potential and Challenges in Rural Canada:
Telemedicine in rural areas has made it easier for people to receive medical help. For example,
OTN helped with over 540,000 telemedicine consultations last year by avoiding 213 million
kilometers of patient travel and saving the healthcare system more than $73 million in travel grants
(Canada, 2020). People’s satisfaction with these services is usually at a very high level. Even so,
using telemedicine is much more difficult for rural communities. Developing digital infrastructure
is a big problem, as rural areas in Canada have access to high-speed internet for only 40% of
households, while in cities, almost everyone enjoys this service (Canadian Institute for Health
Information, 2023). Because of this digital divide, both the quality of video consultation and
adoption of enhanced telehealth methods are affected. Rural populations encounter technological
literacy barriers and demonstrate limited digital health technology adoption. Healthcare
professionals in rural areas experience problems as well such as affordably installing telemedicine,
finding support with technology and making their procedures compatible with technology. It was
found during the pandemic that 78% of rural providers shifted more than 50% of their sessions to
virtual platforms (Burton et al., 2022), yet they dealt with problems like weak internet connection,
difficulties using a range of telehealth tools and challenges from billing providers. There are reports
(Rahimipour Anaraki et al., 2022). that the use of telehealth made the care gap wider by causing
more rural patients to see their doctors less often via video visits. They point out that technology
by itself will not solve all the problems; approaches are required so everyone in rural communities
receives fair benefits.
AI as a Game-Changer: Artificial Intelligence which uses machine learning, natural language
and advanced data processes, helps advance telemedicine. AI may help doctors make more

accurate diagnoses, forecast patients’ needs ahead of time, automate regular workflows and
prioritize individualized care – this all helps healthcare providers to serve more people in a better
and more efficient way. When specialists are not available in many locations, AI can be used to
triage patients by checking their complaints (e.g., with AI questionnaire apps), examine different
X-rays or eye scans and keep tabs on patients’ vital signs from afar. A good example is Babylon
by Telus Health, a Canadian app that uses a medical chatbot to identify patients’ symptoms and
suggest possible explanations for them. AI is effective in advising patients about the level of care
they need and thus guides doctors on making better use of their time. According to Dr. Wong
(MacKay, 2023), an AI expert in Canada, rural communities can particularly benefit because local
health practitioners use AI to get additional information when patients are having telemedicine
encounters.
Study Aim: In this context, the present review explores the multifaceted role of AI in enhancing
telemedicine for rural Canada. The authors look at (1) the primary rural barriers to telemedicine
adoption and possible solutions that AI can provide to alleviate these barriers; (2) the current and
potential role of AI on healthcare accessibility and patient outcomes in rural telemedicine
programs; (3) the economic cost-benefit analysis of implementing AI in telemedicine; and (4) the
policy frameworks and ethical dilemmas that modulate or limit the use of AI in telemedicine in
Canada. Through integrating evidence published within the past ten years, we will attempt to
summarize the findings on current evidence of potential ways to leverage AI and result in more
accessible, efficient, and equitable healthcare to Canadians in rural settings.
Relevance and Significance: The results of this review will be of importance to healthcare policymakers, technology developers and rural health practitioners. The current state of the healthcare
system in Canada is in the transitional phase, as 18 percentage points lower than in 2020 only 54
percent of Canadians now report a positive experience with the quality of accessible healthcare

(Ipsos, 2022). This decrease is mainly explained by the systemic pressures in terms of workforce
shortage and wait time as it is most severe in rural regions. As depicted in Figure 1, recent surveys
(Ipsos, 2022) indicate that Canadians identify “not enough staff” (63%) and “long wait
times/access issues” (47%) as the top problems in :
Figure 1
Major problems facing the Canadian healthcare system, according to a 2022 national survey.

Note. Staffing shortages and wait times dominate public concerns, data derived from Ipsos Global
Health Service Monitor by Ipsos Global Health Service https://www.ipsos.com/en/three-fiveglobally-say-their-healthcare-system-overstretched. Ipsos.
In rural settings, these issues are magnified, but AI-enhanced telemedicine presents an opportunity
to alleviate some of these pressures by extending the capabilities of a limited workforce and
streamlining care delivery processes. Through an academic inquiry into current literature and data,

this paper illuminates how AI-integrated telemedicine can be a catalyst for positive change, while
also discussing the precautions and policies needed to ensure that these innovations are deployed
responsibly and effectively (Auditor General of Canada, 2023).

Literature Review
The literature review is organized thematically to cover the core aspects of telemedicine in rural
Canada and the infusion of AI technologies. We first discuss barriers to telemedicine adoption in
rural areas, then examine AI’s applications and impacts on telehealth accessibility and outcomes.
Subsequently, we explore economic analyses of telemedicine (with and without AI), and finally,
we review policy frameworks and ethical considerations relevant to AI-driven telemedicine.
Barriers to Telemedicine Adoption in Rural Canada
Digital Infrastructure Gaps: One of the most cited barriers in the literature is inadequate
broadband and cellular infrastructure in rural and remote areas. High-quality telemedicine,
especially video consultations and AI applications, requires stable internet connectivity. However,
connectivity in rural Canada remains suboptimal: about 60% of rural Canadian households do not
have access to high-speed internet service (Burton et al., 2022). (meeting CRTC’s 50 Mbps
download target). By contrast, only 3–5% of urban households lack similar access. This disparity,
often termed the “digital divide,” is reinforced by findings from the Auditor General and CRTC
reports, which highlight that while over 90% of all Canadian households had basic broadband
access by 2021 (Auditor General of Canada, 2023), many rural and especially remote Indigenous
communities were far behind that benchmark. Without reliable internet, patients and providers face
dropped telehealth calls, poor video quality, or inability to use data-intensive AI tools (such as real-

time diagnostic image processing). A qualitative study by Burton et al. (2022) described rural
physicians in Western Canada expressing frustration with “inconsistent Wi-Fi” and connectivity
issues impeding virtual visits.
Technological Literacy and Accessibility: In addition to infrastructure, one can also find
disparities in tech literacy among populations. Telemedicine platforms can be hard to use by older
people living in rural parts of the country or who are not exposed to digital tools. Geriatric
practitioners have cited issues regarding the ability of elderly patients to navigate apps or video
interfaces related to telehealth (e.g., managing issues related to downloading the app, logging in
or solving audio/video problems). Although Canadian statistics are limited, in the U.S. literature,
poorer educational background and older individuals are associated with using less telemedicine
and this is most probably similar in Canadian rural populations. Accessibility also has something
to do with the presence of proper devices (smartphones, tablets, computers) and assistive
technologies among disabled people. No adoption of telemedicine may occur, in case the platforms
are not user-friendly and accessible in local languages (including, in Canada, Indigenous
languages) (Rush et al., 2021).
Health System Integration and Provider Workload: Many rural healthcare providers work in
small practices or clinics that may not have dedicated IT support. Introducing telemedicine can
add to their workload, especially if systems are poorly integrated with electronic health records
(EHRs) or require duplicate data entry (Virtual Care Task Force, 2022). A CIHI analysis observed
that half of clinicians using virtual care in Canada said it increased their workload initially, citing
“administrative and technical barriers” and poor integration with other systems. For rural
practitioners already stretched thin, additional administrative burden is a deterrent. Telemedicine
billing can also be complex – until recently, many provincial fee schedules did not reimburse
virtual consults at parity with in-person visits, although COVID-era policy changes have improved

this. Inadequate compensation or uncertainty about billing rules can discourage providers from
investing time into telehealth services (Virtual Care Task Force, 2022). Also the report by Virtual
care task force shows some clinicians express concern that telemedicine might compromise the
quality of patient relationships or clinical exam components. A U.S. physician survey found 35%
of doctors were unsure if the doctor-patient relationship could be maintained virtually (although
those with good tech quality were much more satisfied). It is found in Canada on the rural providers
the trade-off between the convenience and the quality of care such that although the telemedicine
is flexible, they fear they may miss non-verbal communication or a finding on a physical
examination.
Equity and Trust Issues: The literature also focuses on the need to implement telemedicine with
caution as it may also have a harmful effect on health inequity. There is also the possibility of
historically underserved groups (low-income, elderly, or Indigenous populations in rural
communities) possessing less access to digital technology or trust in remote care as well,
potentially leaving them further behind. Such as, Indigenous people of Northern Canada may
experience the poorest connectivity levels and may be adversely affected by specific cultural and
linguistic barriers in healthcare; their engagement in the development of telemedicine should result
in approaches that would be acceptable and effective. A factor – patients require trust in virtual
care and some may not be confident that a phone/video consultation will be able to help them
adequately. Community education and proofs of positive results can be a part of trust-building
(Fitzpatrick et al., 2023).
Table 1
Key Barriers to Telemedicine Adoption in Rural Canada

Barrier

Description

Evidence/Source

Lack of high-speed 60% of rural households lack high-speed
Digital
internet and cell internet vs 3% urban. Real-time Virtual
infrastructure
coverage in rural Support in BC needed to invest in rural
gaps
areas.
connectivity.
Users’ difficulty with
Digital
technology; lack of >Half of rural-focused telehealth studies cite
literacy
&
devices or assistive user training as needed
accessibility
tools.

Note. Digital infrastructure gaps refer to limited high-speed internet and cellular coverage in rural
areas; digital literacy & accessibility refer to users’ difficulty with technology and lack of devices
or assistive tools. Data adapted from The potential socio-economic impact of telemedicine in
Canada, by D. Jacobs, A. Balew, G. Langlois, & M. Lapierre, 2022, Healthcare Management
Forum, 35(4), 233–240 (https://doi.org/10.1177/11786329221096033). Copyright 2022 by The
Authors.
Addressing these barriers is important for any telemedicine initiative’s success. Also, many of
these obstacles can potentially be mitigated by carefully leveraging AI tools, as discussed next, but
only in conjunction with investments in infrastructure and training.
AI’s Impact on Telemedicine Accessibility and Outcomes
Artificial Intelligence technologies are increasingly being integrated into healthcare (Ernst &
Young, 2024), with varied applications that can complement telemedicine. In rural telehealth, AI’s
contributions can be grouped into several categories: diagnostic support, patient monitoring,
workflow automation, and health education/predication.

AI-Assisted Diagnostics and Triage: One of the earliest uses of AI in telemedicine is in triage
chatbots and virtual assistants. These AI chatbots interact with patients (often via text or voice) to
collect symptom information and medical history (Hafner et al., 2021), then provide preliminary
advice or direct the patient to the appropriate level of care. In Canada, Babylon by TELUS Health’s
AI-driven Symptom Checker is a prominent example. Upon launch in British Columbia, it allowed
users to input symptoms and then utilized a vast database (500 million data points) (Telus, 2019).
to suggest possible causes or actions. For rural patients, such tools can be available 24/7 to answer
health questions when local clinics are closed, reducing unnecessary ER visits for minor issues
and flagging serious conditions promptly. Research indicates that 70% of common doctor visits
could initially be addressed by a virtual consultation or AI triage, freeing doctors to focus on
complex cases (Telus, 2019). Another example is AI in radiology: rural hospitals often lack onsite
radiologists, so they rely on teleradiology services. AI algorithms can pre-screen imaging (X-rays,
CT scans) for urgent findings. Studies have shown high accuracy for AI in detecting certain
conditions – e.g., an AI for detecting diabetic retinopathy on fundus images that received FDA
approval. Integrating such AI into a tele-ophthalmology program could allow earlier diagnosis of
eye diseases in remote communities without waiting for specialist review. AI algorithms for
dermatology (e.g., identifying suspicious moles via photos) also hold promise, as dermatology tele
consults are viable and AI can help prioritize lesions that need biopsy (Hafner et al., 2021).
Remote Patient Monitoring and Predictive Analytics: Chronic diseases are prevalent in rural
populations, often due to limited access to preventive care. Telemedicine has addressed this via
remote patient monitoring (RPM) programs, like Tele homecare for heart failure or COPD in
Ontario (CIHI, 2022). AI can enhance RPM by analyzing the continuous streams of data (vital
signs, blood glucose, etc.) to detect early warning signs. For instance, machine learning models
can predict heart failure exacerbations days in advance by spotting subtle changes in weight or

blood pressure trends, prompting early intervention (some studies in the U.S. show reduced
hospital readmissions with such AI-driven alerts (Ernst & Young, 2024). In Canada, a pilot in
Nunavut’s Virtual Nurse Practitioner program, while not explicitly AI-driven, demonstrated how
data analysis of patient metrics led to significant improvements in chronic care management (CIHI,
2022). We can also imagine adding AI to this program to customize interventions e.g., to predict
which diabetic patients will do poorly in terms of glycemic control and prioritize NP outreach to
those patients. The other advanced field is the mental health telemedicine: AI-based sentiment
analysis software can analyze the words or the speech of a patient in a virtual counseling practice
to detect his/her mood or suicidal tendencies in real-time, providing feedback to the therapist with
whom they can act (though this should be taken with caution as far as the ethics of the matter are
concerned). Research is relatively new, but it is especially applicable in the case of rural and remote
locations where there is a limited number of mental health professionals. The effect of the patient
falling through the cracks between appointments can be mitigated by providing decision support
(such as nudges upon data indicating a patient is worsening) with the use of AI.
Efficiency Gains and Task Automation: AI can handle repetitive tasks, which are valuable where
human resources are scarce. In telemedicine workflows, this might include automated appointment
scheduling, follow-up reminders, or summarizing patient encounter notes. Natural language
processing (NLP) algorithms can transcribe and organize information from a teleconsultation,
generating a visit summary or updating the EHR, thus saving clinicians time. McKinsey (2020)
estimated that integrating AI into administrative processes could save 9–18% of healthcare
provider time by 2030. In a rural clinic, where one nurse might be doing both clinical and
administrative roles, having an AI assistant schedule follow-ups or flag patients who missed visits
can maintain continuity of care with minimal extra burden. AI can also support decision-making:
for instance, clinical decision support systems that, given patient data, suggest potential diagnoses

or care gaps (like reminding a doctor that a patient is due for cancer screening). These tools, when
embedded in telehealth software, ensure quality of care remains high even in fast-paced virtual
visits (Ernst & Young, 2024).
Improved Outcomes Evidence: One should question the effectiveness of AI in telemedicine in
terms of patient outcomes and provider or patient satisfaction. Current evidence is promising but
mixed and often condition-specific. A RAND Corporation study in 2022 systematically reviewed
outcomes for AI-supported telemedicine in chronic disease management and found no systematic
reviews indicating worse outcomes with telemedicine vis-à-vis standard care (Hafner et al., 2021).
For chronic diseases like COPD and heart failure, several reviews noted telemedicine (with some
AI-driven education or monitoring) can improve quality of life and reduce hospital use, although
not all studies show statistically significant differences. In rural telehealth contexts: a notable
success story is the use of AI in stroke care (telestroke) (Ali et al., 2020). In Ontario’s Telestroke
program, while the “AI” component might be limited to decision tools for thrombolysis, the
telemedicine aspect has saved many lives by ensuring rural ERs get immediate neurologist input
(Canada, 2020). Now, experimental AI algorithms are being tested to interpret brain scans for
stroke and advise on treatment within seconds, which, combined with telestroke, could further
speed up care in rural hospitals. In terms of patient satisfaction, rural patients often report
appreciation for telemedicine’s convenience (no travel, timely access), but they also emphasize the
need for human interaction quality. AI can help behind the scenes, but should not reduce the
empathy and communication from providers. Interestingly, a participant from a Nunavut telehealth
program said their experience was seamless and “I don’t have to talk about a new thing over and
over – it’s seamless”, reflecting how well-designed remote programs (with consistent providers
and data continuity possibly aided by IT systems) can make care feel integrated.
AI in Action – Example Programs:

•

Virtual Triage and Consultation (Telus Babylon): As mentioned, millions globally have
used Babylon’s chatbot. In British Columbia, after initial deployment, it scaled up virtual
doctor consultations covered by public insurance. The synergy of AI triage plus video
doctor visit is an example of AI augmenting telemedicine efficiency. Early reports (Telus
Health, 2020) suggested that a large percentage of users got appropriate advice without
needing an immediate in-person visit, indicating effective triage.

•

Real-Time Virtual Support (RTVS) in BC: In rural British Columbia, the RTVS system
was launched (especially in 2020) to connect rural practitioners with virtual specialist
support in real-time (e.g., emergency, maternity, pediatrics). While not described as “AI,”
it has potential to integrate AI analytics. The program had various pathways (one called
“CHARLiE” for emergency consults). An evaluation noted it was an equity effort to help
rural/First Nations communities (Lauscher et al., 2023). If AI tools like predictive risk
scoring (which patient needs higher level intervention) were integrated, it could further
strengthen such programs.

•

Chronic Disease AI Monitoring (Future Vision): Consider diabetes management in
Northern communities. Patients could wear glucose monitors that feed into a telemedicine
app. AI analyzes patterns and alerts a tele-nurse of any patient trending poorly, who then
intervenes with a virtual visit or medication adjustment. Evidence from pilot studies (e.g.,
in New Brunswick’s tele rheumatology with AI screening of blood tests) hints at improved
control of conditions and patient empowerment when feedback loops are tight (SantosSilva et al., 2024).

In summary, AI is steadily adding an “intelligence layer” to telemedicine that can help overcome
human resource limitations (by offloading tasks or enhancing decision-making) and potentially

improve outcomes by catching problems earlier and personalizing care. However, robust outcome
data is still emerging. Many studies call for long-term evaluations, since sustained use of AI in
telehealth is a recent phenomenon.
Economic Cost-Benefit Factors of AI-Driven Telemedicine
When assessing telemedicine (with or without AI) for rural healthcare, it’s crucial to consider the
economic perspective: costs, savings, and overall value for patients and the health system.
Telemedicine in general is often justified by savings from avoided travel, improved chronic disease
management, and reduced unnecessary ER or hospital visits. Adding AI could introduce new costs
(technology investment, training, maintenance) but also new savings (efficiency, improved
population health outcomes). (Hafner et al., 2021)
Cost Savings from Telemedicine: A comprehensive RAND analysis (Hafner et al., 2021)
quantified potential socio-economic impacts for Canada. Key findings included: if telemedicine
consultations rose to 50% of primary care visits (from ~4% pre-pandemic baseline), the time saved
by patients (travel, waiting, time off work) would be worth about CAD $5 billion per year,
roughly 2% of Canada’s total annual health expenditure. This staggering figure comes from
valuing citizens’ time (for instance, a 2-hour round trip saved per visit for rural patients who no
longer have to travel). Furthermore, wider telemedicine use could prevent about 50% of
unnecessary ER visits, translating to direct healthcare cost savings of up to CAD $1.47 billion
(Hafner et al., 2021). Additionally, providing access to primary care via telemedicine for the ~3%
of Canadians without a regular doctor could yield well-being benefits valued at CAD $611 million
annually (due to improved satisfaction and preventative care). These estimates illustrate the huge
economic opportunity if telemedicine is effectively deployed.

Rural Canada stands to gain disproportionately from such savings. For example, Ontario’s
Northern Health Travel Grant program spends tens of millions yearly reimbursing rural patients
for travel to specialists; telehealth can cut these costs. OTN reported saving $73 million in travel
grants in one year by telemedicine usage. Similarly, less travel means fewer hours of work lost and
lower out-of-pocket expenses on gas, accommodation, etc., benefiting rural patients economically
(Improvements to the Northern Health Travel Grant Program Now in Effect, 2024).
Figure 2, show impact of AI-driven telemedicine in terms of economics in Canada. It can be seen
that any modest gains in the use of telemedicine are economically disproportionate--increasing in
CAD million dollars to more than 1.3 CAD as adoption rises to 25 %. These benefits exceed CAD
2.2 billion with the 25 % usage scenario inclusive of the value of leisure-time. This pattern
highlights one of the most distinct arguments of the paper which is that incorporation of AI in
telemedicine would greatly stimulate efficiency and allow healthcare systems to save a great
amount of time both on the sides of patients and providers.

Figure 2
Economic Value of Time Saved—Baseline Scenario Pre-COVID-19 Levels of Teleconsultations

Note. Adapted from “The potential socio-economic impact of telemedicine in Canada,” by D.
Jacobs, A. Balew, G. Langois, & M. Lapierre, 2022, Healthcare Management Forum, 35(4),
Article 9242553. https://pmc.ncbi.nlm.nih.gov/articles/PMC9242553/table/table-1/.

These savings have other net benefits besides the straightforward monetary value in that they
translate to significant benefits in society including enhanced patient convenience coupled with
decreased caregiver burden. Moreover, the mentioned rising advantages are consistent with the
policy suggestions made in the paper, which implies that telemedicine infrastructure investment
can be useful in covering gaps in healthcare equity and align resource allocation in rural
communities. The graph, in essence, moves the theoretical concepts expressed in the literature to
the realm of solid, measurable economic consequences that is the bargaining chip behind the
widespread introduction of telemedicine, AI-powered applications which are suggested by the
same subtitle in the research paper.
Costs of Telemedicine Implementation: Telemedicine is not free; it requires investment in
equipment (cameras, software, remote monitoring devices), training, and support services. In rural

clinics, initial costs for telemedicine units can be significant – e.g., installing telehealth equipment
in a Nunavut health center can cost thousands, but economies of scale improve as those units are
utilized for many visits (Nunavut has telehealth in all 25 communities, with territorial government
and federal funding support) (Virtual Care Rapid Scoping: Supporting Information, 2022).
Connectivity improvements (broadband expansion) are perhaps the costliest infrastructure aspect,
but those are societal investments beyond just telemedicine (federal programs like the Universal
Broadband Fund are addressing this). Also, from a health system perspective, reimbursing
telemedicine may initially seem like an extra cost, but if it substitutes for more expensive in-person
care or improves health to avoid future costs, it’s cost-effective. A Canadian Agency for Drugs and
Technologies in Health (CADTH) summary (2022) found that higher volumes of telehealth reduce
per-patient cost, implying that once the system is set up, using it more yields better ROI (Virtual
Care Rapid Scoping: Supporting Information, 2022).
AI-Specific Costs and Benefits: Introducing AI into telemedicine has its own cost-benefit
considerations. Costs include developing or licensing AI software, integrating it with telehealth
platforms, and ensuring cybersecurity and data storage for the AI’s data needs. There may also be
a need for specialist oversight (Dhruv et al., 2025). On the benefit side, AI could enhance cost
savings by improving efficiency and accuracy. For instance, if an AI triage can reduce the number
of physician consultations by handling minor cases, that is a direct cost saving (physician time is
expensive). If AI helps detect a complication early (such as spotting a diabetic foot ulcer via a
photo app before it worsens), it can prevent costly hospitalizations or advanced treatments. A study
conducted to compare the cost-effectiveness of AI screening of diabetic eye disease in rural and
urban settings were found to be cost-effective particularly about long-term blindness prevention
and lower specialist visit requirements. (Curran et al., 2022).

A scenario analysis might be illustrative: consider a rural
region with 10,000 people. Normally, perhaps 500 of them travel to a city each year for specialist
consultants at, say, $300 each (travel + consultation cost), totaling $150,000 (Dhruv et al., 2025).
If telemedicine can handle 70% of those, saving $105,000, and AI further streamlines to save an
additional 10% by efficient triage, that’s another $10,500 saved. On a larger scale, these add up.
Moreover, AI can scale at relatively low marginal cost – once an algorithm is developed, using it
for 1000 patients isn’t much more expensive than for 100 patients (just computing power) (Dhruv
et al., 2025). Thus, AI adds scalability to telehealth interventions.

Case Studies of Effective AI-based Integration in Rural Telemedicine
The effective AI-augmented telemedicine systems in rural settings show how localized
implementation may fill the accessibility pockets, enhance diagnostics, and streamline clinical
care. As an example, Babylon by TELUS Health in British Columbia combines an AI-based
symptom checker with online physician interactions so that the rural population can easily access
timely advice and a triaging help without making an onsite visit. Initial feedback indicates a large
percentage of customers received suitable onboarding with no action that led to delivery of inperson care, thus limiting valuable clinical facilities in the rural areas (Telus, 2019).

A typical example of this kind of use case is where AI is used to enhance current telemedicine
networks, such as the Telestroke Program in Ontario. With image interpretation tools of CT scans
with AI assistance, rural hospitals will get stroke diagnosis in real time, which will allow starting
thrombolytic treatment within critical timeframes. The reduced transfer times and better patient
recovery outcomes have been attributed to such integration (Canada, 2020).

Lessons in low-resource, rural-dense settings identified internationally will show scalability during
the nationwide implementation of Babylon AI-based triage in Rwanda. The use of AI algorithms
trained on the data of people with diverse end-user backgrounds helped to increase the accuracy
of triage and even eliminate pointless referrals via this program (Perez et al., 2025). In the same
way, AI-enhanced teledermatology services in Australia mean that rural health care professionals
can upload photos of lesions via AI-powered malignancy severity assessments to expedite referral
of suspected cancers and cut down on unneeded skin care visits (Pagallo et al., 2023).

These case studies highlight that success can be dependent upon three main factors viz., substantial
digital infrastructure, culturally, linguistic tailored AI interfaces, and they must be well integrated
with the current clinical workflow to support continuity of care.

Economic Challenges: It should be acknowledged that some economic impacts are harder to
monetize. For example, improved patient satisfaction or reduced caregiver burden due to
telemedicine doesn’t directly show up in budgets, but it’s a social benefit. Also, broad adoption of
AI could have workforce implications – in the long run, if AI handles more tasks, there might be
savings by needing fewer staff for administrative roles. However, current consensus is that AI will
augment rather than replace healthcare providers, given the need for human judgment and empathy
(Dane.Schultz, 2024).
Lastly, are the opportunity costs and the equity decisions: the decision to invest in AI telehealth
may imply taking funds away from other interventions. The policymakers should evaluate whether
the returns (e.g. better access and outcomes) justify it relative to, e.g., investing in additional,

physical clinics. As indicated thus far, a mix is required; AI telemedicine is not an alternative to
base healthcare investments (Pagallo et al., 2023).
Policy Frameworks and Ethical Considerations
Telemedicine and AI in healthcare have soared and developed far beyond policy development in
many aspects. Canada is a multi-level system, with the principles of healthcare federalism (Canada
Health Act), provincial administration of healthcare and the regulatory bodies of the profession.
AI-mediated telemedicine brings up issues of licensure, privacy, data governance, reimbursement,
and ethic (Canada, 2020).
Government and Regulatory Initiatives: In the awareness of the telehealth boom, Canadian
governments are already beginning to form policy on virtual care. In May 2020, the federal
government announced $240.5 million in funding to accelerate virtual care and mental health
tools. This included $150–200 million for provinces to invest in secure platforms, remote patient
monitoring, and integration of new technologies (Virtual Care Task Force, 2022). Part of this
initiative was to set up a Federal-Provincial-Territorial Virtual Care Task Force. The Virtual Care
Task Force (a collaboration of CMA, CFPC, and Royal College) released recommendations
focusing on quality, equity, and integration of virtual care (Virtual Care Task Force, 2022. Among
these were calls to: establish a pan-Canadian governance framework for virtual care, ensure
medical regulatory standards support virtual practice, fund virtual services under public health
insurance, and crucially, make equity a guiding principle. They also advocated for payment parity
(paying doctors the same for virtual as in-person visits) and enabling cross-provincial licensure to
facilitate care access across boundaries – the latter remains an issue as licenses are provincial,
though some special arrangements exist (e.g., for bordering communities). By 2022, most

provinces had introduced billing codes for telemedicine permanently, which is a positive shift for
sustainability(Dane.Schultz, 2024).
For AI in healthcare, Canada has taken steps like launching the Pan-Canadian Artificial
Intelligence Strategy (CIFAR-led, focusing broadly on AI innovation) and more specifically,
Health Canada and CPHA have begun exploring frameworks for AI in clinical settings.
Nonetheless, in comparison to the EU (proposed AI Act) or the U.S. FDA (issued guidelines on AI
as medical devices), Canada is yet to develop explicit regulatory frameworks of AI tools to be used
in the healthcare field. In the present case, diagnostic AI software that is intended to be used in
making clinical decisions would probably be considered medical devices, and hence subject to the
Medical Devices Regulations, necessitating them to be formally approved (as is the case with
radiology AI algorithms with Health Canada) (MacKay, 2023).
Privacy and Data Security: Telemedicine platforms must comply with laws such as PHIPA in
Ontario or PIPEDA (Office of the Privacy Commissioner of Canada, 2024). federally for personal
health information. Introducing AI often means handling large datasets, possibly cloud storage,
and algorithms that might derive sensitive inferences. Ensuring patient consent for AI analysis and
anonymizing data where possible is vital. There are also concerns about cybersecurity – any
vulnerability could lead to data breaches of telehealth records or even malicious manipulation of
AI recommendations. Policymakers encourage strong encryption and adherence to standards (e.g.,
Babylon by Telus asserted it meets or exceeds privacy standards, storing data in Canada). (Rusnell,
2021).
Ethical Considerations: Equity is an ethical concern of big significance. Where data is not used
that reflect urban/rural or indigenous populations, AI may not perform satisfactorily on these
groups and misdiagnoses, or biases can result (Norori et al., 2021). As an example, skin cancer

detectors that are trained primarily on light-skin may not detect dark lesions. It is also ethically
bound to ensure there is inclusion of different data in the development of AI and to test AI tools
within the rural context prior to expanded usage. Moreover, the principle of beneficence is that the
AI in telemedicine must arguably benefit its individual patients, therefore careful experimentation
must be conducted to prevent any harm by relying too much on an AI which may make mistakes.
It is also transparent, the patient should be informed when AI is used in his or her treatment and
may choose to ask questions or not. The principle of informed consent can be adjusted to include
what AI can bring to clinical decision-making.
Another issue is the physician’s liability when using AI. If an AI recommends an action and the
doctor follows it and it causes harm, who is responsible? As of now, standard of care is determined
by what a reasonably prudent clinician would do; using AI doesn’t remove the clinician’s
accountability. Guidelines are needed to help clinicians understand how to safely incorporate AI
(e.g., AI as a second but not final word) (Hendry, 2025).
Policy Framework Snapshot: Health Canada’s “Enhancing Equitable Access to Virtual Care”
(2021) emphasizes that governments may need to provide infrastructure so rural Canadians can
benefit, lest virtual care increase inequities. It also notes that if not integrated well, virtual care
could inadvertently create issues like fragmenting continuity of care or adding costs for patients
(like data usage fees) (Health Canada, 2021). Therefore, policy measures include investing in
broadband (the federal government has aims to connect 98% of Canadians by 2026), supporting
digital literacy, and encouraging integrated models (for example, virtual visits with one’s own
family doctor rather than random walk-in apps to maintain continuity (Health Canada, 2021). On
the AI front, the Government of Canada’s Advisory Council on AI has identified health as a priority
area, and there are calls for clear ethical guidelines akin to the CMA’s policy on AI in health care
(CMA released a draft guidance in 2020 promoting patient-centered AI, transparency, safety, and

effectiveness) (Health Canada, 2021). Table 2 outlines some key policy and ethical guidelines
emerging for telemedicine and AI in Canada.
Table 2
Policy and Ethical Guidelines for AI-Enhanced Telemedicine

Aspect

Current
Framework/Guidelines
Canada

Licensure
Practice

No
national
license;
provincial
telemedicine
licenses exist. Some provinces
allow virtual care across
&
borders under agreements.
CMA
Task
Force
recommends pan-Canadian
licensure for virtual care
(MacKay, 2023).

Need to harmonize licenses to allow
rural patients to see out-of-province
specialists virtually. Avoid regulatory
silos. (Norori et al., 2021)

Most
provinces
added
permanent telehealth billing
codes (fee-for-service parity
Reimbursement in
many
cases).
FPT
governments invested in
virtual care (e.g., $240M
fund) (MacKay, 2023).

Ensure
long-term
funding
so
telemedicine isn’t unfunded extra
work. Consider new billing for
asynchronous AI review tasks.
(Dane.Schultz, 2024)

in Key Considerations

Governed by provincial health
info laws; PIPEDA for
federals. Draft Canada Digital
Charter may influence AI. No
specific AI health law yet.
(MacKay, 2023)

Must ensure AI training data privacy,
informed consent for using AI, data
residency requirements (use Canadian
servers). (Dane.Schultz, 2024)

Healthcare AI tools need
Health Canada approval if
used for diagnosis (case-by&
case).
CMA
guidelines
emphasize validation and
clinical oversight. (MacKay,
2023).

Regularly audit AI performance,
especially if diagnosing. Maintain a
human-in-the-loop. Develop reporting
systems for AI errors. (Norori et al.,
2021)

Federal frameworks call for
equitable virtual care (e.g.,
Equity
& make equity a fundamental
Accessibility
principle). Digital health
strategies
target
rural/Indigenous inclusion.

Subsidize devices/internet for lowincome rural patients; provide
multilingual and culturally adapted AI
telehealth services.

Privacy & Data

Quality
Safety

CADTH and CMA encourage
principles:
transparency, Implement AI ethics training for
Ethical Use of
accountability, explainability clinicians, and an oversight committee
AI
of AI decisions. Avoid for AI deployments in health .
algorithmic bias.
Note: Created by author
The overall policy of Canadians is heading in the right direction but is still in its evolvement stages.
The COVID-19 experience brought urgency to the table and resulted in the rapid implementation
of certain telehealth policies; currently, the challenge is to cement what has been achieved

extending it to AI and ensuring the health system does not lose its momentum without risking to
patients.
Thematic Findings
Synthesizing insights from the literature review, several key themes emerge regarding AI’s role in
rural telemedicine:
1. Enhanced Accessibility through AI: AI reduces access barriers, through pre-care (chatbots)
and expansion of specialist accessibility (through decision support in diagnosis). In the case of the
rural patients, this implies better, and earlier response to medical issues, as well as possible reduced
waiting period of referral. Nonetheless, such accessibility benefits depend on the digital
infrastructure; it is impossible to run AI without connectivity. In such a way, the synergy between
AI and the addressing of infrastructure gaps is beneficial (MacKay, 2023).
2. Efficiency and Workforce Support: AI can significantly improve efficiency in telemedicine
operations. It was repeatedly noted that staff shortages are a critical issue in Canadian healthcare
(63% perception as top problem). AI helps existing staff work “smarter” – triaging routine cases,
automating documentation, and monitoring many patients simultaneously, which can relieve
overburdened rural providers. This can reduce burnout and potentially attract providers to rural
practice if they feel supported by technology (Ipsos, 2022).
3. Improved Clinical Outcomes (with Caveats): In chronic disease management, telemedicine
has shown outcome improvements (e.g., better control of diabetes or COPD, reduced hospital
admissions). AI augments this by early detection of issues and personalized interventions.
Nonetheless, the evidence base is still being built. Telemedicine is at least no worse than in-person
care for many conditions and sometimes better for outcomes like patient satisfaction and certain
clinical parameters. No studies reported worse health outcomes due to telemedicine in chronic

disease, which is encouraging. The caveat is that not every telehealth or AI intervention
automatically yields better outcomes – it must be well-designed, with appropriate follow-up and
integration into care pathways(Canadian Institute for Health Information, 2023).
4. Economic Viability: The subtext here is that making an investment in telemedicine (and AI)
would be economically rewarding through cost-savings. According to government and private
analyses, telehealth can provide savings of millions of dollars through travel and inefficiencies. AI
has the potential to increase such savings in the short run (primarily through efficiency) and, in
theory, significantly in the long run (through population health, and prevention of costly
complications). Nevertheless, the onset of costs, maintenance and equitable distribution of these
services is one thing which has to be planned and continued. (Hafner et al., 2021).
5. Policy and Ethical Readiness: Findings indicate that while technology has leapt ahead, policy
is catching up. Positive thematic notes: Canada is infusing equity into virtual care guidelines and
acknowledging telehealth as an integral part of the system (not just an emergency measure). Yet,
ethical use of AI is an area where clear standards must be continually reinforced. A theme in
literature is the risk of exacerbating disparities if, for example, AI algorithms don’t consider rural
realities. Another is the requirement for maintaining human touch – AI should assist, not replace,
the patient-provider relationship (Virtual Care Task Force, 2022).
6. The Need for Training and Change Management: Implicit in many studies (and occasionally
explicit) is that successful telemedicine with AI needs training for users – both patients and
providers. Rural telemedicine projects that failed often did so because people weren’t comfortable
or properly onboard with the technology. Thus, a thematic insight is the importance of digital
health literacy campaigns and incorporating AI tools into medical education (so new doctors learn
how to work with them) (Burton et al., 2022).

7. Interdisciplinary Collaboration: Deploying AI telehealth solutions involves collaboration
between healthcare providers, technologists, and policymakers. Many case studies (like Telus
Babylon or the Ontario telehomecare program) show that partnerships across sectors (private tech
firms, public health system, community organizations) are instrumental. This is a theme of
innovation through collaboration. (Telus, 2019).
In sum, these are the two themes that point to the fact that whereas the role that AI can play in the
context of rural telemedicine is high, its effectiveness is dependent on systemic factors. It is not a
plug and play solution but one of several components to strengthen healthcare in the rural areas.
Discussion
The integration of AI into telemedicine for rural Canada is a complex socio-technical endeavor.
The thematic findings suggest substantial benefits are attainable, but they are contingent on
concurrently resolving infrastructure and policy issues. In this discussion, we interpret what these
findings mean for the Canadian healthcare system and rural communities, compare with global
contexts, and consider practical implications.
Interpreting the Impact: Perhaps the most striking takeaway is that AI doesn’t change what
telemedicine can do as much as it changes how well it can be done and how far it can reach.
Telemedicine already allowed remote consultations; AI makes those consultations more powerful
by equipping providers with tools (like decision support) and by partially automating patient
interaction before and after the visit. For rural patients, this means telemedicine becomes a more
robust substitute for in-person care in many scenarios. Whereas, in the past, a rural clinic might
have been able to handle, say, 50 percent of matters locally and refer 50 percent to urban facilities,
perhaps now with telemedicine they can handle 70 percent of matters virtually; with AI, perhaps
they can actually handle 80 percent of matters locally/virtually, referring the remaining very

complex cases (Canadian Institute for Health Information, 2023). Such a transition has healthcare
implications: remote and less-sophisticated AI-technology may allow rural hospitals and clinics to
successfully treat more acute patients, perhaps limiting patient transfers and maintaining
healthcare closer to where patients live.
Comparison with Global Trends: The problems of rural healthcare are not limited to Canada
since other countries also experience such problems, such as Australia, the United States, or
specific regions in Europe. The Canadian method of emphasizing equity and telehealth widely
publicly funded in Canada aligns with the experience in some countries (UK, Nordics), and in the
U.S. the spread of telehealth has been more market-based and varied between insurers. The world
is exploring AI in telemedicine - e.g. India is adopting telemedicine kiosks with AI-based
diagnostic aids to bring healthcare to far-flung villages; Rwanda adopted Babylon AI triage in use
nationwide in response to a shortage of doctors (Perez et al., 2025). Such cases can be learned in
Canada. A positive in Canada is its single-payer regime with the potential to roll out wide-ranging
initiatives (such as the virtual-care funding Fund of 240M) and pool best practice across provinces
through CIHI and other organizations. Canada however has a lag in some digital health indicators
(Canada, unlike most OECD countries, is in the bottom half of EMR interoperability and level of
digital health investment, as of 2020) (Perez et al., 2025). Progress in catching up has intensified
during the pandemic, however, it is vital to keep the engine going. Importantly, those in which
broadband was more developed in rural communities (such as in some areas in Scandinavia)
experienced more seamless telehealth implementations; this also highlights the importance of
Canada investing in rural internet concurrently to ensure 98 percent connectivity by 2026 (Perez
et al., 2025).
Patient Outcomes and Quality of Care: A nuanced discussion is required on quality. While
telemedicine and AI can improve convenience and even some health metrics, the absence of

physical examination can be a limitation. Hybrid models might be ideal: periodic in-person visits
with telemedicine in between. AI might help know when an in-person check is necessary (e.g., an
AI noticing vitals instability might suggest “time for an in-person exam”). Ensuring that rural
telemedicine doesn’t become a “second tier” service is a concern – it must be held to high
standards. Encouragingly, evidence like high patient satisfaction and equivalent outcomes for
many conditions suggests telemedicine can match traditional care quality. Some outcomes like
mortality or long-term complication rates are harder to move and need more research. In
discussion, one should also highlight success stories: e.g., Manitoba’s TeleStroke program where
patients in small towns got clot-busting drugs 1 hour faster on average due to virtual neurologist
consults, significantly improving survival and recovery (hypothetical example derived from
known telestroke benefits). If we add AI that recognizes stroke on a CT scan instantly, that could
shave off additional minutes in decision-making. Each such improvement can be life-saving
(Canada, 2020). On the chronic care side, Nunavut’s NP virtual program reported that 97% of
patients felt quality of life improved after telehealth enrollment – this is a quality outcome that
matters in patient experience terms.
Economic and Policy Implications: The economic findings strongly argue for continued support
of telemedicine infrastructure and AI development as investments that will pay off. Policymakers
might ask, how do we fund AI in telemedicine? Options include direct government funding of
certain tools (like Canada Health Infoway supporting telehealth projects) or creating
reimbursement incentives for using AI (like how the U.S. Medicare gives extra payments for using
“Advanced Care Planning” codes, maybe one day there’s a code for AI-reviewed telemonitoring).
The policy environment needs to encourage innovation while safeguarding patients. Regulatory
sandboxes (testing AI tools in a controlled way) could be a method to validate new AI solutions
under Health Canada’s watch. (Canada Health Infoway, 2021).

Challenges and Cautions: Although the situation can be defined as positive, one must speak about
the possible drawbacks. Over-reliance on technology might risk de-skilling providers in
performing physical exams or clinical intuition. Rural providers must maintain a broad skillset
since they handle varied cases – if they lean too much on AI outputs, there could be complacency
(Salim Jr et al., 2023). There’s also the human factor – not all patients want to deal with AI or
telehealth; some will still prefer direct contact. Healthcare must remain patient-centered with
choices. Ethical concerns about data: a notable case in the UK saw Babylon’s chatbot face criticism
for safety after some mis-triage incidents. This shows that rigorous validation and continuous
improvement of AI are needed. In Canada, any AI integrated should be monitored for performance
in the Canadian population and adapted as needed.
Long-Term Vision: Projecting forward in time we can envisage a rural health ecosystem in which
all households enjoy connectivity, an AI-enhanced health hub (maybe their smartphone or a device
at home) that constantly tracks their vital health indicators, advises, and connects them with a
human provider virtually on demand. Telehealth kiosks or clinics would have AI to assist a nurse
practitioner in providing comprehensive services which in the past would need several
specialists(Canada Health Infoway, 2021). With adequate policies in place, Canada could be the
first to demonstrate how AI can meet the principles of the Canada Health Act (universality,
accessibility, etc.) in the digital era, by ensuring access to timely and quality healthcare even in the
communities that are far-flung (Salim Jr et al., 2023). The discussion above further expounds on
the fact that AI in rural telemedicine is not merely a tech-enhancement but rather a route to
transforming the vision of healthcare delivery in rural places. Security will, however, require a
combination of alignment on the level of the systems: technology, policy, funding, and people all
moving in the same direction.
Practical implications

The lessons of this review introduce some policy implications/recommendations on how to
maximize the benefits of AI-enhanced telemedicine in rural Canada:
•

Invest in Rural Digital Infrastructure: The policies must further finance and reward the
spread of high-speed internet and cellular networks in rural and remote locations. The
broadband goals set by the government are admirable and accomplish them from the base.
Policies regarding the access to virtual care will be empty without connectivity (Health
Canada, 2022). It should be considered as an infrastructure issue such as utilities, providing
internet access as an essential service in the program to develop rural areas and perhaps
subsidize in markets where it is otherwise not enabled (as some federal initiatives are
doing).

•

Strengthen Pan-Canadian Virtual Care Framework: Implement the Virtual Care Task
Force recommendations through the establishment of a nationally coordinated framework
or agreements to ensure that healthcare professionals can treat individuals virtually in a
province with equal rights (Virtual Care Task Force, 2022). This will be extremely helpful
in rural locations near provincial borders or where particular specialists are not available
locally (e.g., a patient in rural Saskatchewan would be able to virtually see a specialist in
Toronto). Also, the quality of virtual care should be standardized, where the same set of
standards would apply in all provinces in relation to telemedicine services.

•

Incorporate AI Governance in Health Policy: Health Canada along with collaborating
ministries should now start creating an AI in Healthcare Strategy. This should describe how
AI tools should be evaluated and approved (maybe evolved current regulation of medical
devices), how AI use should be transparent (patients need to know their AI is doing
something), and how to ensure accountability. One can establish an AI advisory council in

the context of the healthcare system that will be tasked with constantly interacting with the
latest trends and advising regulators, as it is done with drug approval committees (Virtual
Care Task Force, 2022).
•

Support Training and Change Management: Policies are needed to make sure that
training programs are funded to allow healthcare providers to understand how to effectively
use telemedicine and remove AI tools (Burton et al., 2022). That may include baked into
medical and nursing programs, CPD (continuing professional development) credits
towards telehealth and AI training, on-site training support (such as telehealth coordinators)
in rural areas during transition phases. Similarly, communities can be empowered through
patient education campaigns (public health messaging on the use of virtual services, how
to verify information provided by AI) and so on.

•

Ensure Equitable Access and Prevent Digital Divide: Affirmative policy actions might
be needed to help certain populations. For example, providing funding for libraries or
community centers in rural towns to host private telehealth booths (for those without home
internet or devices), or distributing connected tablets to patients with high needs (some
pilot programs gave tablets to home care patients with preloaded telehealth apps). Also,
content must be linguistically and culturally appropriate; Health Canada could require that
major telehealth platforms provide services in both English and French (plus possibly
Indigenous languages for relevant areas) (Health Canada Equity Task Team, 2022).

•

Integrate Telemedicine Data into Health Systems: Policies around EHR interoperability
should explicitly include telemedicine data integration. Canada Health Infoway and
provincial eHealth agencies could mandate that telemedicine encounters feed data to
provincial EHRs so that even if a patient sees a virtual provider outside their region

(Canadian Institute for Health Information, 2023), the summary is accessible to their local
providers. This requires technical standardization and agreements – a policy push can make
vendors comply.
•

Monitor and Evaluate Telemedicine and AI Outcomes: Set up a robust evaluation
framework at a national level. CIHI could track metrics like percentage of virtual visits,
outcomes for rural vs urban telehealth, patient satisfaction, and report annually (Canadian
Institute for Health Information, 2023). For AI, perhaps require that any AI used by
publicly funded services undergo an independent audit after 1 year of use. Gathering
information about errors, improvement of results, etc, will contribute to the adjustment of
the policies and the maintenance of credibility.

•

Ethical Oversight: Institutional ethics boards (IECs) must revise their scopes to add the
coverage of AI interventions. The policies may encourage the creation of AI ethics
committees in healthcare organizations or make the AI a part of the clinical governance in
healthcare organizations (Canadian Medical Association, 2020). Additionally, consider the
establishment of rights like “right to human review” (the right for a patient to have a human
reconsider an AI-made decision in care). While this might not be law yet, a policy guideline
can encourage best practice.

•

Funding and Sustainability: Continue funding innovations but also incorporate
telemedicine into core budgets. The one-time infusion was great; now provinces should
bake virtual care funding into their health transfers and budgets (Virtual Care Task Force,
2022). Also, consider special funding for rural telehealth pilot programs with AI – e.g.,
grants that encourage trying new AI solutions in underserved areas (with research

accompanying to measure success). The cost-savings projections can justify these as
invest-to-save measures.
All in all, to seamlessly absorb telemedicine and AI into the policy environment is not enough; the
focus should be on leveraging it as a tool to serve the Canadian commitment to affordable and
quality care to all citizens. Predicting legal, financial, and ethical aspects of technology by actively
speaking out their concerns will allow policymakers to be sure that the use of technology does not
demoralize the fairness and effectiveness of the healthcare system.
Limitations
Although this review gives an in-depth picture of the role of AI in rural telemedicine, some
limitations need to be mentioned:
•

Scope of Literature: The review focused on narrative and thematic review as opposed to
a formal meta-analysis. Our consideration of a large variety of sources (academic papers,
government reports etc.) gives it breadth, however it might not have fully encompassed all
of the relevant studies, given the fast-paced nature of the field. New 20242025 research not
yet broadly indexed may change some of these conclusions.

•

Quality of Evidence: A lot of what is found in telemedicine and AI is based on
observational studies, case reports, and pilot programs. There are fewer randomized
controlled trials (RCTs) in this field (some do exist, however, e.g., RCTs of telehomecare).
Heterogeneity in results is mentioned very often in systematic reviews. Therefore, there
are certain positive assertions (such as saving of costs or improved outcomes) that are
projections or circumstantial. Regional differences in implementation quality relative to
the real effects may differ.

•

Data Specificity to Rural Canada: We tried to keep a focus on Canadian context but
sometimes had to extrapolate from international data or national data not specific to rural.
For instance, statistics on telehealth usage are often nationwide averages, which may mask
rural-urban differences. The rural household internet stat (60% lack high-speed) is
powerful but is a general rural figure (some remote areas might be nearly 100% lacking,
while others near cities are better). Also, rural Canada is diverse – what holds in Northern
Ontario may differ in interior BC or Arctic communities. Our generalizations might not
capture all nuances.

•

AI Technology Limitations: AI in healthcare is a moving target. Some technologies
discussed are in trial phases and not widely deployed. By the time of reading, new AI
models (e.g., those based on GPT-4/5 for medical dialogues or advanced predictive models)
might be in use. This review doesn’t evaluate each AI tool’s technical robustness in detail,
which is outside its scope but relevant for those considering adoption.

•

Bias and Perspective: As authors, we have an inherently optimistic view of technology’s
potential (common in literature), which might bias towards highlighting benefits. We did
mention risks and downsides, but there is always a possibility of underestimating
challenges. We rely on reported data, which itself could be biased (for example, satisfaction
surveys often have response biases).

•

Not a Primary Research Study: No new empirical data were produced in this review. It
pools chemical knowledge. As such, the gaps in existing studies (such as the long-term
health outcomes of the AI interventions) are the gaps in our analysis as well. Example, we
lack data on 10-year outcomes of AI-telehealth or strong cost-benefit achieved in practice
vs modeled.

•

Interdisciplinary Complexity: The issue crosses the field of health care, technology,
economic and ethical issues. Quality in either might have to be sacrificed to quantity.
Experts in any of the fields we cover may feel that our treatment of that field is incomplete
or simplistic in some respects.

Considering such limitations, the results should be viewed by the readers as suggestive trends
and insights but not conclusive evidence. Finer-grained studies on the space where AI and rural
health meet are required. These limitations can be compensated for by future research, which,
for example, can conduct region-specific assessment, as well as patient perspective research.

Conclusion
This review discussed the use of artificial intelligence to supplement telemedicine in rural Canada
concerning their accessibility, efficiency, and health outcomes. The evidence implicates that, AI,
when augmented strategically, promises extensive potential for reinforcing the advantages of
telemedicine: it might assist in patient triaging, clinical decision-making, automatizing routine
work, and remote observation, all of which can increase the coverage of healthcare system assets
that are arsenally limited in rural locations. Telemedicine by its own right has already shown that
it is a lifeline to rural communities: alleviating the burden of travel, connecting patients to specialty
care, and supporting continuity of care at a difficult moment in time, such as during the pandemic.
AI is a force multiplier added to these services that has the potential of making reactive virtual
visits even a proactive and consistent care model.
Although the current review shows that AI has the potential to transform rural telemedicine, certain
areas that are in high need of further research have been identified. A priority where future research

needs to be focused is longitudinal assessments of AI-based telehealth programs to continue to
evaluate the long-term clinical outcomes, cost-effectiveness, and patient satisfaction after pilot
projects. Research on working towards equity-centered AI design should also be pursued and
monitor towards ensuring that the algorithms undergo a test dataset that is representative of the
Canadian rural, Indigenous and multicultural population to avoid biasness. Citing the issues
identified with infrastructure, the study of low-bandwidth AI telehealth systems aimed at
delivering care e.g. diagnostic tools that can be used offline could provide support in isolated
places with spotty connectivity.

In conclusion, AI empowered telemedicine is not some far away vision but a reality, which has a
potential to revolutionize rural healthcare provision in Canada. It goes in line with the core
visions of our health-care system i.e. to deliver high quality affordable and patient-centered
health-care to all people, no matter where they are located. Through experience gained and by
keeping policy and innovation on the right path, Canada has the opportunity to ensure that this
technology enabled change in healthcare translates to meaningful changes in the lives of rural
Canadians; greater access to care when they need it, a reduction in the number of staff required
to provide service, and inevitably, an improvement in health outcomes.

References

Innocent, E., Emeihe, V., None Mojeed Dayo Ajegbile, Aderonke, J., & Cosmos, C. (2024).
Integrating Telemedicine and AI to Improve Healthcare Access in Rural
Settings. International Journal of Life Science Research Archive, 7(1), 059–077.
https://doi.org/10.53771/ijlsra.2024.7.1.0061
Burton, L., Rush, K. L., Smith, M. A., Matthias Görges, Currie, L. M., Davis, S., Mattei, M., &
Ellis, J. (2022). Has Virtual Care Arrived? A Survey of Rural Canadian Providers During
the Early Stages

of the

COVID-19 Pandemic. Health Services

Insights, 15.

https://doi.org/10.1177/11786329221096033
Canadian Institute for Health Information. (2023). The Expansion of Virtual Care in Canada New
Data and Information. https://www.cihi.ca/sites/default/files/document/expansion-ofvirtual-care-in-canada-report-en.pdf
Canadian Medical Association. (2020). Virtual care in Canada: progress and potential (Report of
the Virtual Care Task Force). Canadian Medical Association. https://www.cma.ca/virtualcare-canada-progress-and-potential-report-virtual-care-task-force-0
Canada Health Infoway. (2021, September 18). Canadian Digital Health Survey – Virtual Care
Use. Infoway-Inforoute.ca. https://insights.infoway-inforoute.ca/virtual_care/
Canada,(2025). Information archivée dans le Web | Information Archived on the Web.
Publications.gc.ca. https://publications.gc.ca/collections/collection_2024/sc-hc/H14-6312024-eng.pdf
Hafner, M., Yerushalmi, E., Dufresne, E., & Evangelos Gkousis. (2021, December 6). The
potential socio-economic impact of telemedicine in Canada. Rand.org; RAND
Corporation. https://www.rand.org/pubs/research_reports/RRA1274-1.html

Perez, K., Wisniewski, D., Ari, A., Lee, K., Lieneck, C., & Ramamonjiarivelo, Z. (2025).
Investigation into Application of AI and Telemedicine in Rural Communities: A Systematic
Literature

Review. Healthcare

(Basel,

Switzerland), 13(3),

324.

https://doi.org/10.3390/healthcare13030324
MacKay, C. (2023, October 15). AI will be critical for the future of rural health care in Canada,
experts say. CBC. https://www.cbc.ca/news/canada/prince-edward-island/pei-artificialintelligence-1.6994961
Canada,

H.

(2020). Nunavut’s

Virtual

Care Action

Plan

-

Canada.ca.

Canada.ca.

https://www.canada.ca/en/health-canada/corporate/transparency/healthagreements/bilateral-agreement-pan-canadian-virtual-care-priorities-covid-19/nunavutaction-plan.html
Telus. (2019). TELUS Health and Babylon launch new smartphone app. TELUS.
https://www.telus.com/en/about/news-and-events/media-releases/new-app-from-telushealth-and-babylon-enables-canadians-to-visit-a-doctor-through-their-smartphone
Kiran, T., Green, M. E., Strauss, R., Wu, C. F., Daneshvarfard, M., Kopp, A., Lapointe-Shaw, L.,
Latifovic, L., Frymire, E., & Glazier, R. H. (2023). Virtual Care and Emergency
Department Use During the COVID-19 Pandemic Among Patients of Family Physicians in
Ontario,

Canada. JAMA

network

open, 6(4),

e239602.

https://doi.org/10.1001/jamanetworkopen.2023.9602
Anaraki, N. R., Mukhopadhyay, M., Wilson, M., Karaivanov, Y., & Asghari, S. (2022). Virtual
Healthcare in Rural and Remote Settings: A Qualitative Study of Canadian Rural Family
Physicians’ Experiences during the COVID-19 Pandemic. International Journal of

Environmental

Research

and

Public

Health,

19(20),

13397.

https://doi.org/10.3390/ijerph192013397
Ethics and governance of artificial intelligence for health. (2021). World Health Organization.
https://www.who.int/publications/i/item/9789240029200
MacKay, C. (2023, October 16). AI will be critical for the future of rural health care in Canada,
experts say. CBC.
https://www.cbc.ca/news/canada/prince-edward-island/pei-artificial-intelligence1.6994961
Atkinson, Simon, and Chris Jackson. “Three in Five Globally Say Their Healthcare System Is
Overstretched.” Ipsos, 26 Sept. 2022,
www.ipsos.com/en/three-five-globally-say-their-healthcare-system-overstretched.
Ally, S. J., Allen, M., Mariki, K., Masoy, K. J., & Liana, J. (2023, February 19). Understanding
how the use of AI decision support tools affect critical thinking and over-reliance on technology
by drug dispensers in Tanzania. arXiv.org.
https://arxiv.org/abs/2302.09487
Health Canada. (2022, June 10). Pan-Canadian virtual care priorities in response to COVID-19.
Canada.ca.
https://www.canada.ca/en/health-canada/corporate/transparency/health-agreements/bilateralagreement-pan-canadian-virtual-care-priorities-covid-19.html
Health Canada. (2022, March 30). Enhancing equitable access to virtual care in Canada:
Principle-based recommendations for equity. Canada.ca.

https://www.canada.ca/en/health-canada/corporate/transparency/health-agreements/bilateralagreement-pan-canadian-virtual-care-priorities-covid-19/enhancing-access-principle-basedrecommendations-equity.html
Auditor General of Canada. (2023). Connectivity in rural. In Reports of the Auditor General of
Canada to the Parliament of Canada [Performance audit reports]. His Majesty the King in Right
of Canada. https://www.oag-bvg.gc.ca/internet/docs/parl_oag_202303_02_e.pdf
Rush, K. L., Seaton, C., Li, E., Oelke, N. D., & Pesut, B. (2021). Rural use of health service and
telemedicine during COVID-19: The role of access and eHealth literacy. Health Informatics
Journal, 27(2). https://doi.org/10.1177/14604582211020064
Virtual Care Task Force. (2022). VIRTUAL CARE IN CANADA: PROGRESS AND
POTENTIAL. In Report of the virtual care task force. https://www.cma.ca/sites/default/files/202202/Virtual-Care-in-Canada-Progress-and-Potential-EN.pdf
Ernst & Young. (2024). Six ways to make more of AI in Canadian health care. EY
Canada. https://www.ey.com/en_ca/insights/government-public-sector/six-ways-to-make-moreof-ai-in-canadian-health-care
Physician billing codes in response to COVID-19 | CIHI. (2022).
https://www.cihi.ca/en/physician-billing-codes-in-response-to-covid-19
Ali, F., Hamid, U., Zaidat, O., Bhatti, D., & Kalia, J. S. (2020). Role of Artificial Intelligence in
TeleStroke: An Overview. Frontiers in Neurology, 11. https://doi.org/10.3389/fneur.2020.559322
Lauscher, H. N., Stewart, K., Markham, R., Pawlovich, J., Mah, J., Hunt, M., Williams, K.,
Christenson, J., Graham, S., Bepple, K., Pritchard, E., Rabeneck, J., Yang, M., & Ho, K. (2023).

Real-time virtual supports improving health equity and access in British Columbia. Healthcare
Management Forum, 36(5), 285–292. https://doi.org/10.1177/08404704231183177

Santos-Silva, M. A., Sousa, N., & Sousa, J. C. (2024). Artificial intelligence in routine blood
tests. Frontiers in Medical Engineering, 2. https://doi.org/10.3389/fmede.2024.1369265
Improvements to the Northern Health Travel Grant Program now in effect. (2024, December 19).
MFHT. https://mfht.org/2024/12/19/northern-health-travel-grant-program/
Virtual care Rapid scoping: supporting information. (2022). [Journal-article]. CADTH Health
Technology Review, 2(12), 5–162. https://www.cdaamc.ca/sites/default/files/pdf/htis/2022/RE0041S-Virtual-Care-Rapid-Scoping.pdf
Dhruv, S., Dhruv, S., & Dhruv, S. (2025, July 31). The cost of implementing AI in healthcare.
Aalpha. https://www.aalpha.net/blog/cost-of-implementing-ai-in-healthcare/
Dane.Schultz. (2024). The Impact of AI on the healthcare workforce: Balancing opportunities
and challenges. HIMSS. https://legacy.himss.org/resources/impact-ai-healthcare-workforcebalancing-opportunities-and-challenges
Pagallo, U., O’Sullivan, S., Nevejans, N., Holzinger, A., Friebe, M., Jeanquartier, F., JeanQuartier, C., & Miernik, A. (2023). The underuse of AI in the health sector: Opportunity costs,
success stories, risks and recommendations. Health and Technology, 14(1), 1–14.
https://doi.org/10.1007/s12553-023-00806-7
Office of the Privacy Commissioner of Canada. (2024, May 1). PIPEDA requirements in brief.
https://www.priv.gc.ca/en/privacy-topics/privacy-laws-in-canada/the-personal-informationprotection-and-electronic-documents-act-pipeda/pipeda_brief/

Norori, N., Hu, Q., Aellen, F. M., Faraci, F. D., & Tzovara, A. (2021). Addressing bias in big data
and AI for health care: A call for open science. Patterns, 2(10), 100347.
https://doi.org/10.1016/j.patter.2021.100347
Hendry, M. (2025, July 7). Part 1: AI in clinical decision-making: How artificial intelligence is
shifting the standard of care. Canadian Lawyer. https://www.canadianlawyermag.com/practiceareas/medical-malpractice/part-1-ai-in-clinical-decision-making-how-artificial-intelligence-isshifting-the-standard-of-care/392720