Paging Dr Bot: AI meets telemedicine

Dr Neha Mahindrakar

S.S Institute of Medical Sciences and Research Centre

KEYWORDS – AI, Artificial intelligence, Telemedicine, Telehealth, Digital health

The telehealth industry has ushered in a new face of healthcare delivery, characterized by patient centric, affordable and accessible health services. The integration of Artificial Intelligence (AI) into telemedicine furthermore, has set the stage to revolutionise the industry. My first experience with this advancement was at a pediatric clinic in Miami where I spent time as a student observer. We employed an AI based translation tool frequently for consultations with native non-English speakers. AI based translation tools are increasingly being used to support effective communication. These tools provide a real time, accurate translations during consultations, allowing patients to better understand symptoms, diagnosis and management plans. By bridging the language gaps, AI translation services provide more inclusive healthcare, especially in diverse communities.

Other than translation tools, the most impressive role of AI in telemedicine is for diagnostics. AI specifically employs Machine learning (ML) and deep learning (DL) tools that analyse massive medical datasets to recognise patterns and assist healthcare providers in making accurate diagnostic decisions. Ada Health is one such AI based health assistant that helps patients assess their symptoms by asking questions and analysing responses. The app generates a list of causes based on the patient’s responses and also provides guidance for next steps. A study in the Journal of Medical Internet Research found that Ada’s diagnostic suggestions were consistent with the findings of physicians in 69% of cases, which is a significant statistic in the era of AI diagnostics (1). Babylon Health in UK and Buoy Health in the United States offer similar AI driven triaging tools which guide patients with their symptoms.

Aidoc and Zebra Medical Vision are both companies from Israel that use AI to interpret medical imaging remotely. Their systems assist radiologists by marking urgent findings. This is particularly helpful in tele-radiology, where a specialist may not always be readily available on site. These platforms are now being used globally and have shown to reduce time to diagnosis significantly.

In the mental health domain, AI tools are coming up as a significant role in accessible care as well. Woebot, an AI powered chatbot, uses Cognitive Behavioural Therapy (CBT) to help users manage conditions like anxiety and depression. In one study conducted in the United States, patients using Woebot reported improvement in work related impairment and emotional wellbeing (2).

Cera Care, is a home care provider in the UK which uses AI for remote monitoring of elderly patients and predicting potential health issues. Through the use of wearables, Cera’s AI system analyses data from both medical records and real time vital sign monitoring to provide personalized care. Cera has contributed to a 70% reduction in hospitalizations for Cera patients (3).

Double edged sword of Innovation

Despite its numerous benefits, AI in telemedicine has brought to light the paradox of inequality – a scenario where technological advancements widen the healthcare gap. People who could benefit the most from this advancement, such as the elderly, socioeconomically disadvantaged or those in remote regions, are often the least equipped to access it. This is attributed to factors like lower digital literacy, lack of access to smartphones or poor internet connectivity. A study by Lam K found that the elderly patients and racial minorities in the United States had significantly lower rates of telehealth usage during the pandemic, even though these groups faced higher risks of COVID-19 complications (4). Similarly, in rural India, challenges persist due to economic factors where usually only one family member owns a smartphone, making medical consultations for everyone difficult, especially for women and the elderly.

Misdiagnosis due misinterpretation of data or algorithmic biases can lead to incorrect diagnoses. Moreover, inability of AI to conduct physical exams and lack of rapport can all negatively impact patient outcomes.  A well-documented example of reliance on AI for medical decisions gone wrong involves IBM’s Watson for Oncology. This was promoted as a tool to assist doctors in selecting cancer treatments. In 2018, Watson came under scrutiny when internal documents revealed that the system was frequently suggesting incorrect or unsafe treatment options (5). The case highlighted the possibility of clinical oversight by AI in medicine and the importance of using AI as a supplementation rather than replacement in medical diagnosis.

Issues regarding data privacy remain a major concern with telehealth. Weak enforcement of privacy laws, especially in countries outside of Europe and U.S, and potential risk of data breaches make patients wary of sharing sensitive information online, particularly concerning mental health and reproductive issues. 

Insurance mechanisms have struggled to keep up with the telehealth industry. Inconsistent policies regarding reimbursement and standardised fees have discouraged both patients and providers from using telehealth platforms. In remote areas, telehealth services paradoxically cost patients more than in patient consultations.   

The Path Forward

The balance between innovation and humanity is fragile. AI in telemedicine is revolutionizing healthcare and yet, it brings with it some challenges. To overcome these shortcomings, the healthcare, insurance and the tech industries need to come together. As we navigate this digital frontier, the questions remain: does AI assist in delivering more efficient care, or does it widen the gap between technology and human touch which defines true healing?

References

1. Fraser H, Crossland D, Bacher I, Ranney M, Madsen T, Hilliard R. Comparison of diagnostic and triage accuracy of Ada Health and WebMD symptom checkers, ChatGPT, and physicians for patients in an emergency department: clinical data analysis study. JMIR Mhealth Uhealth. 2023 Oct 3;11:e49995. doi:10.2196/49995. PMID: 37788063; PMCID: PMC10582809.
2.  Durden E, Pirner MC, Rapoport SJ, Williams A, Robinson A, Forman-Hoffman VL. Changes in stress, burnout and resilience associated with an 8-week intervention with relational agent “Woebot”. Internet Interv. 2023;33:100637. doi:10.1016/j.invent.2023.100637.
3. Lomas N. UK in-home healthcare provider Cera raises $150M to expand its AI platform [Internet]. TechCrunch. 2025 Jan 12 [cited 2025 May 12]. Available from: https://techcrunch.com/2025/01/12/uk-in-home-healthcare-provider-cera-raises-150m-to-expand-its-ai-platform/
4.  Lam K, Lu AD, Shi Y, Covinsky KE. Assessing telemedicine unreadiness among older adults in the United States during the COVID-19 pandemic. JAMA Intern Med. 2020 Oct 1;180(10):1389–91. doi:10.1001/jamainternmed.2020.2671. PMID: 32744593; PMCID: PMC7400189.
5.  Dolfing H. Case study: IBM Watson for Oncology failure [Internet]. Henrico Dolfing. 2024 Dec [cited 2025 May 12]. Available from: https://www.henricodolfing.com/2024/12/case-study-ibm-watson-for-oncology-failure.html