The market value of artificial intelligence (AI) in the health care industry is predicted to reach $6.6 billion by 2021. AI is becoming a big bet for preventive healthcare in India. AI-based medical technology and machine learning (ML) algorithms are set to solve the existing complexities of clinical care in 2020. The inclusion of AI in healthcare reported sluggish growth in the past few years, but with liquidity infusion from both public and private sector across economies, the growth rate of AI adoption is predicted to gain momentum.

AI-centric tools have flooded the global market. Late last year, Google’s DeepMind trained a neural network to accurately detect over 50 types of eye diseases by simply analysing 3D rental scans. Applications of IoMT (Internet of Medical Things) can be seen in smart ambulances, where sensors are placed in ambulances to track patients’ vitals and share them in real time with the healthcare ecosystem so that physicians can analyse the vitals and make the necessary preparations for treatment before the patient reaches the hospital.  

Cancer screening and treatment is another crucial area where AI provides tremendous scope for targeted large-scale interventions. India sees an incidence of more than 1 million new cases of cancer every year, and early detection and management can be crucial in an optimum cancer treatment regimen across the country. 

Ayushman Bharat and AI

The 2020 Budget has given a green flag to the use of AI/ML in the existing Ayushman Bharat- Pradhan Mantri Jan Arogya Yojana (AB-PMJAY) scheme. Allocating Rs 69,000 crore to the healthcare sector, out of which Rs 6,400 crore was given to Ayushman Bharat scheme, Union Minister for Finance, Nirmala Sitharaman said, “Using machine learning and AI in the Ayushman Bharat scheme, health authorities and the medical fraternity can target disease with an appropriately designed preventive regime.”

NITI Aayog is also pushing heavily for introducing AI into healthcare and steer the Indian medical landscape from being a reactive to preventive care model. NITI’s discussion paper on ‘AI for All’ envisages a framework within which AI tools can be rolled out seamlessly in sectors like agriculture, transport, infrastructure and healthcare. As per Ayushman Bharat’s goal of undertaking path-breaking interventions to holistically address health (covering prevention, promotion and ambulatory care), at primary, secondary and tertiary level, NITI is leading the pathway to scalability of healthcare.  

The premier think tank is collaborating with Microsoft and Forus Health to roll out the technology for early detection of diabetic retinopathy as a pilot project. 3Nethra is a portable device that can detect common eye problem. Integrating AI capabilities to this device using Microsoft’s retinal imaging APIs enables operators of 3Nethra to get AI-powered insights even when they are working at eye checkup camps in remote areas with nil or intermittent connectivity to the cloud. Collaborations like these can ensure scalability of AI products.

After preventive medicine, the second big P which the Indian healthcare landscape is chasing is precision medicine. It is a growing field that refers to dispensing the correct treatment depending on the patient’s characteristics and behaviour. It depends on interpreting vast volumes of patient data to determine the most effective treatment. Data organisation happens to be a strong suit for machine learning and AI algorithms. AI medication systems can browse through these archives to assist doctors in formulating precision medication for individual patients and expedite the treatment phase.

Challenges

There has been a string of AI-centric start-ups within the Indian market like DRiefcase which aims to digitise personal health records of a person and provide users with a single-point, easy-to-use access to medical data. For a sustained slew of start-ups to rise and grow, their entry should be encouraged by facilitative infrastructural government initiatives. The Indian market has been riddled with the following hurdles that have posed hindrance to the incorporation of AI:

• Lack of enabling data ecosystems 
• Low intensity of core research in fundamental technologies and transforming those into market applications 
• Inadequate availability of AI expertise, manpower and skilling opportunities 
• High resource cost and low awareness for adopting AI in business processes 
• Unclear privacy, security and ethical regulations 
• Unattractive Intellectual Property regime to incentivise research and adoption of AI-centric innovations

Now specifically, in the healthcare sector, there has been negligent collaborative effort between various stakeholders. While India has adopted an electronic health record (EHR) policy, sharing of data between various hospital chains still remains a work in progress, since different chains have adopted different interpretations of ‘digitising’ records. There needs to be a guiding document that erases ambiguity from these terminologies. Relevant data is also usually unavailable since there are very few robust open clinical data sets. 

Road Ahead

AI in healthcare should ideally empower the patient to track and evaluate his/her treatment process. One of the major hindrances to this could be the ‘black box’ problem - if the process showing how an algorithm reaches its recommended decision is not transparent, how does the doctor/ medical authority evaluate whether the recommendation is right? 

The principle of ART – Accountability, Responsibility and Transparency should hence be embedded in all AI applications. Guidelines must be present on how to ensure these in AI related tools wherever significant risks can be expected while performing healthcare services.

Also, if a patient is getting operated by an AI-surgeon, then it becomes crucial for mechanisms to ensure that distillation of the required information happens at different levels of expertise and authority like with the medical board who validates the procedure, the in-house doctor, the patient, his family and other ancillary entities like insurance firms.

Retrospective and prospective responsibilities and obligations need to be etched out for past or future actions emanating from AI involvement. There should be frequent auditing of AI algorithms to ensure fairness and accuracy across different groups of people based on ethnicity, gender, age, and health insurance. This stands essential given how AI applications in other fields have already shown that they can easily pick up biases.

India’s goal of achieving Universal Health Coverage (UHC) requires AI as a strong companion to ensure we deliver on patient satisfaction and isn improving the quality of life for 1.3 billion citizens.