AI in India | Chapter 5: The Rise of Artificial Intelligence in the Indian Healthcare Ecosystem

Source: Forseemed

Over the years, numerable technological advancements have transformed people’s lives and disrupted industries worldwide. In Healthcare, technological innovations such as wearables, chatbots, and mobile apps, have helped people be in charge of their health and made them more health-savvy. From X-rays to electronic medical records, technology has driven success in Healthcare in many ways. Although there are various technological advancements that happened in Healthcare over a period of time, the coming of Artificial Intelligence (AI) has led to innovations that have a profound impact on the masses.

Even though AI has been around as an academic and scientific discipline since the 1950s, AI has emerged as a rapidly evolving data science technology and has seen widespread acceptance in many fields over the last few years. It has the potential to radically transform industries around the world, especially Healthcare. By utilizing advanced machine learning (ML) models, algorithms, and computing power, AI is transforming the strategic as well as the operational landscape of businesses in various industries. AI models combine information (or data points) from various sources, analyze the data, and deliver data-driven insights — all in real-time. For instance, imagine being able to collect and analyze patient’s data on their visit to the hospital, medication prescribed, lab tests and procedures performed, along with their family medical history, as well as data outside of the healthcare ecosystem — such as purchases made digitally, social media, census records, internet search activity logs that contains potential health-related information, and you’ll get a sense of potential AI-based capabilities holds in transforming patient care and diagnoses.

1.1 Artificial Intelligence Landscape in India

Being the world’s fastest-growing business hub with over 1.39 billion citizens generating their digital footprints every second, India’s Artificial Intelligence moment is truly here and now. The Indian AI market is entering a new phase in 2021, and the narrative is shifting from asking whether AI is viable to declaring that AI is now a requirement for most organizations trying to compete globally. With this change in business mindset, the revenue from the AI industry in India is growing at a healthy annual rate over the last few years. According to The Hitchhikers Guide to Artificial Intelligence 2019–20, the revenue generated in 2019 by India’s Artificial Intelligence industry grew by 80% compared to last year, adding a whopping 415 Million USD in revenues alone. In addition to this, the ongoing COVID-19 pandemic has accelerated digital uptake and helped bridge the adoption gap; these numbers are expected to grow even further in the coming years.

The Indian Artificial Intelligence market is valued at 6.4 Billion USD as of July — August 2020, and it is expected to flourish and act as a dominant economic growth driver in the coming years. It has been adapting to shifting narrative swiftly and is driving vital R&D initiatives of AI-integrated technologies across several industries and segments, including Healthcare.

As private investments pouring in and the government unlocking the AI’s real potential by providing proper governance with the help of its policy thinktank NitiAayog, more and more organizations across various sectors started leveraging AI to enhance their existing processes to yield better outputs. According to a report published by Accenture, AI can add 957 Billion USD to India’s economy in 2035. That’s 15% of India’s current gross value. The constantly evolving AI functions are surging rapidly to penetrate all industry sectors, which should contribute to India’s overall GDP growth.

1.2 Impact of Integrating AI in India (Across Industry/Sectors)

AI utilities in India are maturing at varying rates across industries and segments. Nonetheless, the average pace of growth in India is very high. Moreover, the growth pace for the year 2020–2021 would be lower than the average pace of growth for the period 2021–2025 because of the recessionary effects caused by the unfortunate pandemic. According to a report, the CAGR for the entire Indian AI market is 13.8%, taking the entire Indian AI market’s valuation to 11.4 Billion USD by 2025.

Talking about AI penetration and its consequent impact on various industries, the IT services segment is the most significant adapter of AI capabilities. Hence, it is one of the most significant segments in terms of the overall market share at 41.4% and 2625 Million USD in the market value. The Technology segment, which includes software and hardware technologies, follows the IT services with 23.3% and 1488 Million USD in Indian AI market value. Following the two is the BFSI sector having a market share of 9.6%, equivalent to 615.3 Million USD in market value.

These three segments drive India’s maximum growth of AI capabilities and constitute nearly 3/4th of the total Indian AI market share. However, there are other segments that contribute to the AI market in India. These segments have small but significant and steadily growing market size and share. These include Engineering, Industrials, Automation at 6%, eCommerce & Retail at 5%, Energy & Metals at 2.3%, Telecom at 2.2%, Automotive at 2.1%, Education and Public Research at 2%, Healthcare at 1.9% of AI market share among others.

1.3 State of Healthcare in India

Affordability and accessibility to quality healthcare have emerged as one of the major concerns worldwide, especially in developing nations like India. Indian Healthcare System faces several challenges including inadequate access, affordability, low insurance penetration, and chronic disease burden. In order to facilitate the world’s second-largest population with quality healthcare and matching infrastructure, India is yet to raise its total spending more than five percent of the GDP, as compared to the global average of over ten percent. As out-turn, the Indian Healthcare system has numerous shortcomings.

In India, a significant portion of the population is underserved in health care. With a distressing doctor-to-patient ratio of 1:10 (10 times lower than the recommended 1:1000 by WHO), around 122 women die out of every 100,000 live birth, every minute a newborn dies out of 25 million childbirth every year, nearly 46 per cent of all maternal deaths and 40 per cent of neonatal deaths happen during labor or the first 24 hours after birth — all because of inadequate healthcare infrastructure. Not only does India sit at the bottom on healthcare spending among the major economies such as the USA, China, UK, Germany, Canada, etc., it also has the lowest number of beds available at its public hospitals. According to data from National Health Profile–2019, India has around 7,13,986 total government hospital beds, which amounts to 0.55 beds per 1000 population. It is worrisome to note that all stats mentioned above are just the tip of the iceberg. On top of it all, the second wave of ongoing Covid-19 just shattered the entire Healthcare system — exposing its shortcomings dreadfully.

Despite various shortcomings, things are finally starting to change for good. With the inclusion of advanced technologies, the Indian Healthcare sector has a lot going for it on several fronts. With the growing population and healthcare cost, allocating a higher budget to national healthcare expenditure can help in solving many issues related to infrastructure building and strengthening the existing ones. And the government’s aim to increase healthcare spending to three per cent of the GDP by 2022 is a move in the right direction.

According to India Brand Equity Foundation, the healthcare market in India is expected to reach 372 Billion USD by the year 2022 — driven by rising per capita income, health awareness, lifestyle, and increasing access to insurance.

Moreover, the Government Of India is making some commendable steps ensuring improvement of overall healthcare in the country. For example, intending to lower the prices of medical products, such as stents and implants, a patient has to pay; the Indian government has launched the ‘Make in India’ initiative to encourage domestic manufactures to produce medical devices that were usually imported. Additionally, the availability of telecom bandwidth is making medical expertise reach underserved rural markets through telemedicine and tele-consulting programs delivered over mobile phones. Moreover, earlier this year, the Parliament passed the National Commission for Allied, Healthcare Professions Bill 2021 to regulate and standardize the education and practice of allied and healthcare professionals to match global healthcare standards.

With the inclusion of technological advancements disrupting the healthcare industry, it is imperative to design and develop technology that takes into account Indian local constraints, among them affordability. There are many local and behavioral challenges in the Indian Healthcare sector, but the cost is still a key driver. For it to succeed and make a difference at scale, new technology has to be priced for the country and developed to tackle its constraints. The good news is that this is what AI inspires. If implemented correctly, AI has the potential to improve, by leaps and bounds, the outcomes of precision medicine and the quality of healthcare.

With its vast pool of human talent coupled with huge unstructured medical data and population diversity, India is perfectly positioned to lead AI-based innovations that can stand the challenges of robustness and performance at scale, two principal metrics of success in an AI system. In this article, we are set out to explore:

• Impact of AI on Indian Healthcare ecosystem and its potential use cases,
• How AI-powered technologies help fight COVID-19 pandemic,
• Major players to leverage AI capabilities in the Indian Healthcare landscape, and
• Challenges for AI adoption in the Indian Healthcare system.

2. AI in Indian Healthcare System

The adoption of AI is reshaping the Indian healthcare market significantly. Recognizing AI’s potential to transform economies in the foreseeable future and the need for India to strategies approach, NITI Aayog, in its 2018 report, put Healthcare on the top priority in the list of verticals that need an AI push. As per ResearchandMarkets.com’s report “Artificial Intelligence (AI) in Healthcare Market in India (2018–2023)”, the applications of artificial intelligence in the healthcare space are expected to register an explosive CAGR of 40% through 2021 and reach 6.6 Billion USD this year.

AI-empowered healthcare services like automated analysis of medical tests, early detection by screening X-rays and CT scans by cross-checking with millions of recorded data, predictive healthcare diagnosis, and wearable sensor-based medical devices are already revolutionizing medical treatment processes in the country.

2.1 How AI Is Changing Healthcare in India?

With huge resources of amorphous unmedical data and population diversity, India is uniquely placed for enterprises and institutions globally to develop scalable AI-powered solutions which can be easily implemented in the rest of the developing and emerging economies. Each of the systems falling for the Indian Healthcare system is a unique opportunity for AI-based capabilities. Simply put, solving for India means solving for 40% or more of the world.

According to NITI Aayog, the policy think tank of India, AI adoption for healthcare applications is expected to rise exponentially in the coming years. It also suggests that the advancement in AI-based technologies and interest from innovators attracting external investments will facilitate India to solve some of its ever-lasting challenges in providing quality healthcare services to a major section of its population. Furthermore, AI combined with the Internet of Things — Medical Devices (IoT-MD) and robotic automation processes could act as the new nervous system for the healthcare ecosystem, presenting solutions to address healthcare issues and help the government meet its mission-critical objectives.

Riding on advanced technology, the Indian Healthcare industry is expected to reach the 372 Billion USD mark by 2022, according to NITI Aayog CEO Amitab Kant.

Considering the potential impact AI can have over the masses, and the prioritization of AI-based solutions for the Healthcare ecosystem by NITI Aayog has created an incentive for greater collaboration between government, technology companies, and traditional healthcare providers. For example, the Maharashtra state government, in partnership with NITI Aayog and Wadhwani AI group, has launched the International Centre for Transformational Artificial Intelligence (ICTAI) primarily to focus on rural healthcare. Similarly, NITI Aayog is collaborating with Microsoft and the medical technology startup Forus Health to develop a pilot for the early detection of eye-related diseases.

2.2 Potential Use Cases For AI in Healthcare in India

2.2.1. Cancer Screening

The incidence of cancer cases is on the rise in India, and the AI-inclusive Healthcare ecosystem has the potential to affect several facets of cancer care, making it more accessible, affordable, and accurate. Unfortunately, India sees more than 1 million new cancer cases every year, and due to inadequate healthcare infrastructure, affordability issues, and social stigma, 70 percent of these cases consult doctors in the terminal stage.

For an annual incidence of more than 1 million new cancer diagnoses every year, the Indian Healthcare ecosystem lacks adequate human resources and quality pathology services essential for cancer care. AI and machine learning solutions could provide immense scope for targeted large-scale interventions in cancer screening and treatment procedures. It could also help oncologists in the early detection and management of cancer cells — essential for building a robust cancer care system in the country. Inclusion of AI-based capabilities in cancer screening and treatment can help in improving:

• accuracy and speed,
• surgical outcomes,
• surgical learnings, and
• cancer research.

2.2.2. Tuberculosis (TB) Diagnosis

Tuberculosis (TB) has remained one of the world’s most infectious diseases, with an estimated 10 million cases around the globe in 2019, and over a quarter of them were in India, according to Global Tuberculosis Report 2020.

As per India’s annual TB report, 79,144 casualties occurred due to TB in 2019 alone and remained to be one of India’s biggest public health challenges.

TB is an onerous disease to fight against. Physicians say that the bacteria responsible for causing TB can stay dormant for an indefinite period and has the ability to activate at any time, which can cause the patient to manifest symptoms. Launching the TB Free India Campaign at ‘Delhi End TB Summit, Prime Minister Narendra Modi said India is implementing a national strategic plan to eliminate TB by 2025, five years ahead of the global target of 2030. But with the lack of health awareness, affordability, & accessibility to quality infrastructure along with stereotypical social stigma, this has always been an uphill battle.

Consequently, AI-based capabilities can potentially play a role in detecting and eliminating TB through the cascade of care. The Government of India has already been collaborating with various AI-based startups in integrating AI for TB-related healthcare programs. Inclusion of AI can help in:

• Reducing the chances of human error and turnaround times in gaining results.
• Reading chest X-rays and CT scans for finding or predicting the presence of TB tissues in a patient.
• Efficacy of anti-TB drugs.
• Once diagnosed, the patient can be tracked using certain behavioral patterns and contextual data and could be provided with a differentiated care system based on their specific needs.
• Improving existing systems in locating the potential hotspot based on patient multitudinous data points.

2.2.3. Diabetic Retinopathy Screening

India is set to become the diabetic capital of the world. The number of cases of diabetes is expected to grow from 40.6 million in 2006 to 79.4 million by 2030.

Diabetic retinopathy (DR), a morbid microvascular complication that arises due to 2/3rd of all type II and all type I diabetics, is one of the leading causes of preventable blindness. With over 400 million people diagnosed with diabetes worldwide, India is currently home to over 72 million cases, with an estimated DR prevalence of around 18 percent.

The clinical course of DR has a long asymptomatic stage. People with diabetes may not experience visual symptoms, so they may not visit the ophthalmologist for regular retinal screening. Their regular point of care is usually the endocrinology/diabetology or the physician clinic. Hence, the importance of screening is well established for early detection, as timely intervention can reduce visual impairment due to DR. This is exactly where an AI-based system has the potential to fit in. Inclusion of AI can help in:

• Early detection and referral recommendations at a primary care level.
• Processing a large amount of imaged-based data with accuracy, consistency, and that too in a short time.
• Screening a large number of people with diabetes in rural and remote areas based on patient medical history.

2.2.4. Chronic Obstructive Pulmonary Disease Diagnosis and Management

Chronic Obstructive Pulmonary Disease (COPD) is a common chronic respiratory disease caused by exposure to harmful gases and particulate matter. Although several international guidelines for the diagnosis and management of COPD are available, there are many gaps in the recognition and management of COPD in India due to the scant healthcare infrastructure available in the country.

For a long time, the medical community has depended on patient history and clinical symptoms for disease diagnosis, which often prevents early detection, and advancing the disease adds to the medical bill through frequent hospitalizations. Moreover, with pollution on the rise, especially in the urban cities, COPD has begun a high health burden on the Indian Healthcare system. However, with the recent development of diagnostic devices based on internet-of-things medical devices integrated with AI, the Healthcare system is now equipped with AI-based solutions that are helping in the effective diagnosis and management of COPD. Some of the application areas are:

• AI-based early warning systems can help patients in the identification of triggers, symptoms, trends, and other personalized insights related to COPD.
• AI-aided inhalers can help in monitoring the correctness of drug delivery techniques.
• AI-aided lung imaging can assist pulmonologists in the visualization of structural and functional parameters of lungs.

2.3 AI and COVID-19 in India

Years before the Covid-19 outbreak, Bill Gates, in a 2015 TED talk titled “The next outbreak? We’re not ready”, predicted that the next big disaster that would kill more than 30 million people would not be a nuclear war but an infectious virus and stressed on the need for the word to be well-equipped to tackle the situation.

And yet, here we are today.

While the rest of the world has started heading towards normalcy, India is far from anything but normal. In such dreadful times, where the second wave of coronavirus collapsed the heavily burdened Indian Healthcare system, AI and its applications have been a boon in dealing with COVID-19 situations. According to a global study by PwC India, India has seen a sudden surge in the adoption & implementation of AI-based solutions. These solutions help track the pandemic, screen COVID-19 cases, containment of coronavirus, contact tracing, enforce quarantine and social distancing, monitor suspects, treatment, remote monitoring of COVID-19 patients, and are also being used for vaccine and drug development.

2.3.1 Modelling the Pandemic

Artificial Intelligence and Machine Learning-based predictive models have emerged as one of the powerful weapons in the fight against COVID-19 in the country. Key stakeholders and decision-makers now rely on computer-based simulations to understand how the pandemic situation will evolve over time. Companies such as TCS and KPMG India have developed AI/ML models which can predict the severity of the disease and identify at-risk populations across the country. For instance, in collaboration with Pune-based Prayas Health Group, TCS has developed “Digital twin”, a virtual computerized model of a physical system that takes real-world data as input and predicts the system’s future evolution to forecast the spread of COVID-19 in urban districts.

2.3.2. AI for Pandemic Preparedness

2.3.2.1 Conversational AI-enabled MyGov

MyGov, the world’s largest citizen management platform, facilitates two-way communication between the Indian government and its citizens. In the fight against Covid-19, MyGov, JioHaptik Technologies Limited, and the WhatsApp team collaborated to develop AI-enabled ‘MyGov Corona Helpdesk’ — a conversational AI platform, to improve the government’s outreach and engagement. The conversational AI platform or chatbot engages citizens in a user-friendly manner, allowing them to ask questions, clear their doubts, share government advisories, advice of medical experts, stories of COVID19 survivors, and provide real-time updates related to Covid-19.

Source: MyGov

2.3.2.2 IBM Watson Assistant

In collaboration with The Indian Council of Medical Research (ICMR), tech giant IBM has implemented the Watson Assitance, an AI-enabled query answering model, on its portal. The Watson Assistant responds to the queries on COVID-19 raised by front-line workers and data entry operators across testing and diagnostic facilities in India. The Watson Assistant works 24*7 and responds both in English and Hindi.

2.3.2.3 Thermal Cameras and tools for screening

Agrex.ai, an AI-based startup in Gurugram, has developed a thermal sensor-based detection system capable of scanning a large number of people from a distance up to twenty meters. These thermal cameras can scan temperature within a fraction of seconds, eliminating the need to stop and scan each person individually, enabling users to examine 80–100 people in one minute, ensuring early detection of Covid-19 cases, and preventing the spread of the virus.

In the fight against the pandemic and to control the surge in the covid-19 cases due to large gatherings, the Government of India also leveraged this technology by installing thermal cameras at different parts of the Kumbh location, organized earlier this year, to monitor people’s body temperature.

Source: The Conversation

2.3.2.4 Contact Tracing and Investigation

Contact tracing is one of the key pillars in the control of infectious diseases, including COVID-19. The government has launched Aarogya Setu App, an Indian COVID–19 “contact tracing, syndromic mapping, and self-assessment” digital service, primarily a mobile app developed by the National Informatics Centre under the Ministry of Electronics and Information Technology. This app is available in 11 Indian languages and uses location & Bluetooth to track the user’s movement. An alert is generated whenever a user has been within six feet of a COVID-19 patient by cross-referencing the pan-India government database of COVID-19 patients.

Along with Aarogya Setu, other apps have also been launched, such as Sandhane and SAHYOG, to trace COVID-19 suspects in rural &remote areas and to help community workers in carrying out door-to-door surveys, public awareness campaigns, contact tracing, and delivery of essential items, respectively.

Source: Business Standard

2.3.2.5 Enforcing Quarantine and Social Distancing

AI-powered Geofencing technology is used for enforcing strict vigilance on COVID-19 suspects in quarantine. It uses location-based services like GPS to set up a virtual boundary around the quarantine centre. The mobile app is installed in the COVID-19 suspect’s mobile device, and this app uses GPS to trigger an alert whenever the suspect’s mobile device enters or exits the virtual boundary.

Many Indian states have launched their geofencing technology-based mobile apps to monitor the Covid-19 spread due to incoming travelers from different states or countries. Mobile applications like MahaKavach by Maharastra Government, Quarantine Monitor by Tamil Nadu Government, Corona Watch by Karnataka Government, Corona Mukt Himachal by Himachal Government, SMC COVID-19 Tracker App by Gujrat Government, etc. are examples of AI-powered geofencing technology-based solutions currently used for enforcing quarantine and social distancing measures in the country.

2.3.2.6 Treatment and Remote Monitoring of Patients

With the combination of robotic process automation and AI, healthcare professionals and patients can now monitor their health from a distance and automatically send data to remote health centres.

For monitoring COVID-19 patients, Indian states are exploring the use of remote monitoring systems. These remote monitoring systems enable remote monitoring of a patient’s vital parameters like heart rate, SPO2 — blood oxygen level, pulse rate, body temperature, respiration rate, etc. Examples of operational remote monitoring systems would include LiFi (Light Fidelity) technology in Ahmedabad (Gujrat), 311 mobile app in Indore (Madhya Pradesh), Monal 2020 in Uttarakhand, Milagrow Humanoid ELF in AIIMS, New Delhi, to name a few. Moreover, the Kerala Government has initiated the use of robots’ KARMI-Bot’ and ‘Nightingale19’. These robots serve medicines and food to the COVID-19 patients, collect trash used by the patients, enable video calls between patients and doctors or relatives and perform disinfection of the isolation ward.

2.3.2.7 Sero Survey Platform

Indian IT startup Thalamus Irwine has developed an AI and IoT-based solution named ‘Garuda’, which claims to complete a serosurvey with one crore samples of COVID-19 cases within a week. This technology could be helpful in the identification of the vulnerable groups, communities, and geographical pockets where least or no immunity has been developed against COVID-19. In addition, these AI-empowered capabilities could also be used in vaccine prioritization and putting brakes on the spread of the COVID-19 infection.

2.4 Top AI-based Organization in Healthcare (India)

India has encountered several brilliant minds in the sphere of artificial intelligence technology catering to the Healthcare industry. Here are the top five Indian startups leveraging the advantages of AI in the field of Healthcare:

2.4.1. NIRAMAI

NIRAMAI stands for “Non-Invasive Risk Assessment with Machine Intelligence,” is a Banglore-based deep-tech startup that has developed a novel breast cancer screening solution that uses ML models over thermography images to detect breast cancer at an earlier stage than traditional self-examination methods. The core of the NIRAMAI solution is Thermalytix, a computer-aided diagnostic engine that Artificial Intelligence powers. The solution uses a high-resolution thermal sensing device and a cloud-hosted analytics solution for analyzing thermal images. Along with Thermalytix, NIRAMAI’s cancer screening tool, SMILE (Software with Machine Intelligence for Life Enhancement), is a web interface for the NIRAMAI certified technician to upload demography information about the patient along with her thermal images.

Source: NIRAMAI

AI & ML-based models, along with big data analytics tools, are leveraged to analyze the patient’s breast health condition, automatically generating a report listing certain unique parameters and recommending a breast health score. This method of screening can detect tumors five times smaller than what a clinical exam can detect. It is a low-cost, accurate, automated, portable cancer screening tool that a simple clinician can operate. Unlike mammography, this imaging method is radiation-free, non-touch, not painful, and works for women of all ages.

2.4.2. Qure.ai

Qure.ai, based in Mumbai and founded in 2016 by Prashant Warier, uses deep learning algorithms in its healthcare products. The team believes in using AI & ML-based models to hangle the easier medical task so that the medical practitioners can focus on cases that truly matter and are intricate. Qure’s products use large medical datasets to develop its deep learning algorithms for medical imagining.

One of Qure’s AI solutions, qXR, uses insights from processing 3.5 million chest X-rays to detect tuberculosis symptoms early. Now the four-year-old Indian company is putting the same data to use in the fight against Covid-19.

Source: Qure.ai

2.4.3. SigTuple

SigTuple, a Bangalore-based startup founded in 2015, is building cloud-based machine learning solutions to detect abnormalities and patterns in medical data to aid diagnosis. It combines artificial intelligence, robotics, and data science to build smart screening solutions, to make healthcare accurate, accessible, and affordable. The company is building an artificial intelligence platform called ‘Manthana’ to help identify visual data efficiently. This AI platform helps them in the following five screening processes of the Healthcare industry:

• Analysis of peripheral blood smears
• Urine microscopy
• Semen screening
• Fundus screening
• OCT scans and chest X-rays

2.4.4. OncoStem

Founded in 2011, OncoStem uses machine learning models to help physicians formulate a customized treatment plan by understanding cancer tumor biology. OncoStems uses prognostic tests that assess aggressiveness to identify the unique characteristics of cancer recurrence risk. As a result, patients receive the most effective customized cancer therapy designed specifically for them — thus, creating personalized cancer patients’ care.

tumorsTheir flagship product CanAssist Breast provides information about the risk of recurrence of early-stage, hormone-receptor-positive breast cancer patients. It digitizes historical patient medical records and feeds into an AI algorithm that analyzes the data to generate actionable knowledge for doctors. They are also currently working on similar ML-based models for other types of cancer, including oral and ovaries cancer.

2.4.5. Artelus

Artelus stands for “artificial learning system,” is a Bengaluru-based deep learning startup focusing on creating and predicting Early Warning Systems to prevent fatal diseases and ailments.

Its flagship product, Diabetic Retinopathy Screening (DRISTi), is an AI-based solution designed to detect diabetic retinopathy (DR) by using deep learning algorithms during the eye check-up screening process instantaneously. It captures the patient’s retina image, analyses it, and creates a report out of it in less than three minutes. It is also working on creating early diagnostic AI solutions that are accurate, affordable, and easily accessible for screening, diagnosing, and preventing diseases like TB, breast cancer, and lung cancer. Moreover, Artelus AI models are now also being used to reduce the burden on the healthcare workforce by reading chest X-rays for Pneumonia and Covid-19.

Source: Artelus

2.5 Challenges and Barriers for AI in the Indian Healthcare System

The unstructured data sets, interoperability issues, lack of open medical data sets, and inadequate analytics solutions that could work with big data are some challenges for AI-driven healthcare. While AI could potentially transform the delivery of patient care, much needs to be done before AI would prove to be a game-changer in the complex Indian Healthcare landscape involving numerous stakeholders, competing priorities, entrenched incentive systems, and institutional cultures.

In resource-poor settings like India, there is a shortage of high-quality data sets related to diseases and conditions prevalent in these settings, making it difficult to train AI algorithms to identify risk factors or diagnose diseases. Furthermore, in many healthcare settings in India, health records are hand-written in local languages, making the process even more challenging. Also, because the Indian population is so diverse, there is a possibility that datasets may have cultural biases like caste, sexuality, etc. Additionally, healthcare data remains fragmented in India, spreading across various organizations, including hospitals, clinics, pharmacies, testing laboratories, etc. Therefore, input data sets used to train AI algorithms must be derived from a large and diverse population, which raises complex issues around data use, privacy, and security.

This section will explore the challenges involved in using AI-based capabilities in healthcare solutions in India.

2.5.1. Infrastructure

While the Government of India has increased spending in the Healthcare industry, the amount of public funding it invests in healthcare is small compared to other emerging economies. Moreover, the government’s investment in health-related AI in India is limited, and research is under-funded and explored. The unavailability of digital infrastructure necessary for AI to take off in India is a further constraint. Most healthcare organizations lack the data infrastructure needed to collect, store, and compute large datasets require to train algorithms optimally — i.e., test them for bias, adjust the model, and continually monitor and evaluate field outcomes. Cloud-computing infrastructure is mainly available in servers outside India, leading many Indian startups to establish themselves outside India. Further, many types of equipment that are used in healthcare for diagnostic or therapeutic purposes are imported from countries outside India, which raises the issue of software compatibility for the adoption of AI-driven healthcare.

2.5.2. Liability and Accountability

In India, in cases of medical negligence, medical professionals are liable to be taken to court. However, it is unclear how the accountability and liability will be determined if the doctor takes the wrong decision due to the glitch in the AI-based system. Effective governance is needed to establish laws related to liability and accountability for AI in Healthcare and develop guidance defining the boundaries of the Healthcare system where AI would not be allowed to take over.

2.5.3. Cost

Implementation of AI in Healthcare requires a substantial initial investment, which is one of the major concerns for a low resource setting like India. Investment by the government of India in the domains like AI and research is growing but yet remains limited.

2.5.4. Trust Issues

One of the issues with adopting AI in Healthcare in India is the acceptability of the results achieved from AI algorithms. The decisions made by doctors based on AI solutions must be explainable, especially in the Indian scenario where the doctor-patient relationship is given complete trust. Further, there is still a lack of understanding about AI, its application areas, and its benefits among the general population and the medical and healthcare professionals.

2.5.5. Training Issues

Lack of AI-trained professionals is another challenge to use AI in Healthcare in India. The readily available workforce does not have the necessary skills to use the AI systems effectively. Careful handling of sensitive health information, protection against data theft, and use of AI systems effectively with optimal results require a specially trained workforce, which is currently a big challenge for India to adopt AI in the healthcare sector.

2.5.6. Inequality Concerns

Large data sets are fundamental in developing AI algorithms but must represent every section of the population to ensure all can benefit. AI models are trained using these large data sets that inevitably reflect the past. If the training data sets contain inherent biases from past human decisions, these biases get amplified by AI-based algorithms. Due to the paramountcy of cultural prejudice related to ethnicity, socioeconomic status, and gender in Indian society, datasets used to develop AI algorithms are likely to aggravate these gaps and inequalities. By analyzing skewed data, these algorithms can generate discriminatory and unfair results that may reflect biases towards a particular gender, race, caste, and religion.

2.5.7. Inadequate Framework and Regulatory Weaknesses

Currently, India does not have any regulatory structure or framework to guide the development of AI solutions and ensure their quality, privacy, and security. This is one of the biggest challenges for India to adopt AI-based healthcare on a large scale.

2.5.8. Data Protection and Privacy

At the core of AI-driven healthcare lies data, and privacy of information is one of the biggest roadblocks for the adoption of AI in Healthcare. Healthcare data is highly sensitive, and data breaches can have detrimental consequences on patients’ autonomy, dignity, and even access to work. Cybersecurity is also a major concern. Much confidential health information available online across the cloud computing environment poses a risk of data security. For example, in 2016, the hacking of a Mumbai-based diagnostic laboratory database led to the leaking of medical records (including HIV status reports) of more than 35,000 patients. This database held the records of patients across India, and many may still be unaware that their details have been exposed.

3. The Future of AI in the Indian Healthcare System

India’s healthcare challenges are unique, complex, and growing. These challenges combined with AI-based capabilities mean India’s approach towards AI strategy must cater to its diverse population and needs. The way forward for India in empowering the AI-enabled Healthcare ecosystem is for its government to foster public-private partnerships in the domain of AI and Health.

The following steps may be a good beginning in that direction.

1. Enact and effectively enforce laws and legislation related to AI in Healthcare.
2. Invest in digital infrastructure development necessary in accelerating AI adoption.
3. Frame policies addressing issues related to ethics, privacy, and confidentiality the AI-driven healthcare.
4. Upskilling the medical workforce, including doctors, nurses, and medical practitioners, for the AI age, so that they can carefully handle sensitive health information, protect data against theft and use AI systems effectively.
5. Incentivising core and applied research in AI which can provide the basis for its commercialization and utilization.

4. Conclusion

AI can unquestionably bring new efficiencies in the Indian Healthcare ecosystem and drastically improve the quality of patient care in the country. And with the recent AI adoption surge leading to AI-based research, investments, and real business applications in India due to pandemic, AI-derived business is expected to grow exponentially, and its value is projected to reach 3.9 Trillion USD in 2022. As a result, many big conglomerate and Indian startups have already started leveraging AI to develop healthcare solutions catering to local needs — ensuring accessibility and affordability to all and thus promoting much-needed healthcare awareness.

However, the technological possibility that AI promises cannot be equated to its adoption. Lack of adequate infrastructure, low funding, weak regulations, presence of a large, diverse, and unregulated private sector, and deeply embedded socio-cultural practices are some concerning factors that cannot be addressed by AI solutions alone.

Adequate digital infrastructure, effectively enforced laws and legislation, training programs to upskill medical practitioners, framing policies incentivizing core and applied research are pre-dominant areas where government intervention will play a crucial role for AI in Healthcare. Finally, concerns around privacy, misuse, and accountability issues are only slowly being understood and require much more far-reaching consideration before AI can deliver safe and fair healthcare solutions in India.