Utilizing Artificial Intelligence in Health sector

According to Niti Aayog, Health care in India is expected to grow to USD 280 billion by 2020 at a CAGR of upwards of 16%, from the current ~USD 100 billion. In a similar vein, Centre for Internet society report states, “AI could potentially add USD 957 billion (or 15% of current gross value added) to the Indian economy by 2035”. India is gradually moving towards innovations in AI and healthcare as can be seen in various state governments providing support to AI startups. For instance, the Karnataka government is endeavoring to raise 2,000 crore by 2020 in various AI organisations. Furthermore, the Karnataka government has also formulated a startup policy and Karnataka Information Technology Venture Capital Fund that can fund AI startups. According to another report by Transparency Market Research (TMR) “global healthcare automation market is growing at a CAGR of 8.8% and will touch $58.98 billion by the end of 2025, up from $28.31 billion in 2016” (Livemint, 2018).

The major challenges in India’s health sector are uneven ratios of doctors (see the chart-1 below), shortage of qualified doctors, non-uniform accessibility to healthcare, affordability and reactive approach to healthcare (patients approach a hospital only when the disease nears an adverse scenario). When it comes to uneven ratio of doctor is to patient, according to Indian Journal of Public Health, “India had 4.8 practicing doctors per 10,000 population”. Estimates indicate that healthcare is expected to grow to 6.9 per 10,000 people. However, according to World Health Organisation (WHO) this ratio must be 1:1000. AI can effectively meet the challenges of unequal ratios, lack of skilled doctors, catering to rural areas for a high-quality healthcare, training doctors and nurses to tackle complicated procedures.

AI and healthcare in different states of India

AI in healthcare includes a wide variety of machine learning knowledge in comprehending and acting on the administrative and healthcare functions. The AI is also deployed for training and research purposes aiding a holistic achievement of goals. Machine learning can be used to merge an individual’s omic which includes genomic, proteomic and metabolic data assisting in predicting the chances of developing a disease. In such cases, preventive therapy can be undertaken by acting on time.

The technological innovation of AI is already in use in different states in resolving healthcare crisis, benefitting in diagnosing a disease, monitoring chronic conditions, assisting in robotic surgery, and drug discovery. Following Health care startups are majoring in AI:

1. Niramai- uses AI for pain-free breast cancer screening. The startup was founded in 2016, Bangalore.

2. MUrgency- A mobile application aiding people in connecting them with medical emergency responses with qualified medical, safety, rescue and assistance professionals. The startup is based in Mumbai.

3. Advancells- The startup is based in Bangalore providing stem-cell therapy also known as regenerative therapy significantly useful for organ transplantation.

4. Portea- The startup assists patients by visiting them at their residences including a wide range of services from nurses, doctors, physiotherapists, and technicians. The startup is based in Bangalore.

5. Addresshealth- This startup in Bangalore provides primary pediatric services to school children where they are screened for screening, vision, dental health, and anthropometry.

6. LiveHealth- A startup based in Pune functions as a management information system in collecting samples, managing patient records, diagnosing them and generating reports.

Source: Healthcare Budget in BRICS countries

Tackling Healthcare Challenges

The healthcare in India suffers from two major challenges firstly, “shortage of qualified healthcare professionals and services like qualified doctors, nurses,technicians and infrastructure: as evidenced in 0.76 doctors and 2.09 nurses per 1,000 population” (Niti Aayog:24) and secondly, “Non-uniform accessibility to healthcare across the country with physical access continuing to be the major barrier to both preventive and curative health services, and glaring disparity between rural and urban India”(ibid). The Tata memorial hospital, one of the leading cancer hospitals in India reveals in its data that less than 23% of the new patients were geographically based in Maharashtra, with a whopping 21.7% of patients traveling from the states of UP, Bihar, Jharkhand and West Bengal to TMH. The time taken to travel more than 1800 km in availing cancer treatment speaks volumes about the deplorable health conditions in India. Similarly, a news report by Indiaspend in 2016 reveals that 48 percent overnight trips made by Indians from rural areas (25% from urban areas) are for medical purposes telling a tale about failing public health system in India. The most probable reason for such overnight trips is lack of qualitative healthcare available in their home states. The table below tells that maximum trips are made by people from West Bengal, Assam, Jharkhand, Goa and Bihar.

Source: National Sample Survey Office

These complications in terms of harsh realities like accessing, spending, investing on health care explicitly shows that healthcare in India is a multi-layered concept. The multiple layers in the healthcare only speaks volumes about the diversity in India and how it is inclined to disrupt with emerging technologies at multiple levels. The use of AI alongwith robotics and Internet of Medical Things (IoMT) could potentially be the new nervous system for healthcare, presenting solutions to address healthcare problems and helping government in meeting its objectives. The use of AI in diagnosing early stage cancer and providing affordable treatment can aid in early detection and treatment of more than 1 million new cases of cancer every year (Niti Aayog: 28).

Arenas for Implementing AI in healthcare

1. Diagnosis– According to estimates by PwC, 80 percent of health data is unstructured making it invisible to current systems. The firms like IBM and Google have deployed solutions to manage this heavy data. For instance, IBM’s cognitive technology processes and analyses data. According to this report by Livemint, “Watson can review and store far more medical information – every medical journal, symptom, and case study of treatment and response around the world – exponentially faster than any human. And it doesn’t just store data, it’s capable of finding meaning in it. Unlike humans, its decisions are all evidence-based and free of cognitive biases or overconfidence, enabling rapid analysis and vastly reducing – even eliminating – misdiagnosis (PwC)”.

2. Monitoring of Chronic conditions– Serious conditions like diabetes, cholesterol, fertility and cardiac health are managed by regular monitoring and lifestyle changes. The devices of machine learning and AI can generate data on patient’s bodily parameters which can be further combined with lifestyle determinants such as food habits, exercise etc.

3. AI assisted robotic surgery– Robots can assist human effort by guiding surgeon’s instrument during a procedure by saving time and avoiding complications.

4. Image Analysis- Pathological evaluations like malaria depend amply on image analysis. Similarly, finding out abnormalities in MRI scan is done manually by radiologists. In these spheres, AI can effectively screen the image for an accurate diagnosis.

5. Fitness wearables– As we have seen with fitbit and other watches such as that of Apple and Xiaomi, these applications provide an insight on the individual’s health on daily basis. AI can manage by encrypting this data and sharing it with doctors or relevant people with a personalised vision on fitness goals.

6. Drug Discovery- AI has potential to help researchers create drugs. One example in this regard is that of Atomwise which uses deep learning process to reduce the time taken to discover new drugs. IBM has also initiated the methods to discover drug use via its Watson AI. “The platform allows researchers to generate new hypotheses with the help of dynamic visualizations, evidence-backed predictions and natural language processing trained in the life sciences domain. It is used by pharmaceutical companies, medical device companies and academic institutions to assist with new drug target identification and drug repurposing,” IBM explains on its website (Livemint, 2018).

To conclude, AI in healthcare has an immense potential to not only eradicate the obstacles but also at the same time introduce some revolutionary changes.

This piece is written by Manisha Chachra. Manisha is Associate Researcher at Govern.