From disease surveillance to aiding diagnoses — how AI tools are revolutionising Indian healthcare

Advancements in AI-enabled tech solutions have led to personalised diagnosis & treatment, smarter decision making in Indian healthcare industry, but the country still has a long way to go.

Illustration by Soham Sen, ThePrint

New Delhi: Dr Sai Praveen Haranath, a pulmonologist and critical-care specialist, often deals with patients who are in severe distress.

But since February this year, he has been able to quickly diagnose and treat them with the help of a new tool — a chatbot — that is available to all Indian doctors on the Apollo 24/7 platform.

Apollo Hospitals — a leading private healthcare chain in India — launched Clinical Intelligence Engine (CIE), a clinical decision support tool, that uses artificial intelligence (AI) and machine learning to learn from millions of clinical data points from Apollo Hospitals’ knowledge base and real-world clinical data.

Speaking to ThePrint, Haranath, who works at Apollo Hospitals in Hyderabad, said that the CIE helps doctors and patients in three significant ways. 

“First, it ensures that doctors don’t have to type every symptom and health issue of a patient as the engine already has the information available. Second, it ensures we don’t miss anything important, such as a possible diagnostic test, and third, it improves the patient’s experience by asking them specific and relevant questions,” he explained.

According to an Apollo spokesperson, if not for CIE, the doctors would have to read for 17 hours a day to keep up with the new drug discoveries and treatments that are emerging every day. “With CIE, the hospital group aims to provide this information to the doctors when they need it during a patient consultation,” he added.

This chatbot is one of many examples of how AI is transforming the Indian healthcare ecosystem and has come as a shot in the arm for doctors. 

Earlier this month, researchers from Great Learning, a leading e-tech company, published a scientific paper in the journal IEEE Xplore showcasing how AI can help detect pneumonia accurately and efficiently in chest X-ray images.

Using pre-trained AI algorithms, the researchers achieved the rapid classification of images into categories such as “lung opacity/normal”, enabling early detection and helping medical professionals make faster decisions. 

This work, which was also done for detecting tuberculosis and Covid in the US, has shown how AI can analyse medical images, detect abnormalities, aid diagnosis, and improve accuracy and efficiency in diagnostics.

The growing influence of AI is not just restricted to diagnosis but has also found space in spheres such as disease surveillance, in-vitro fertilisation, and drug deliveries, among others.

According to Joydeep Ghosh, partner, industry leader – Life Science & Health Care, Deloitte Touche Tohmatsu India LLP, AI can revolutionise healthcare delivery in India by enhancing efficiency, accuracy, and accessibility to ultimately improve patient outcomes.

By leveraging the capabilities of AI, Ghosh said, India can improve healthcare accessibility, enhance diagnostic accuracy, optimise treatment outcomes, and drive overall efficiency in healthcare delivery.

“These advancements have the potential to positively impact the lives of millions of people by ensuring better healthcare services and outcomes,” he added. 

Echoing Ghosh, Chaitanya Raju, executive director, HealthPlix Technologies, which is helping doctors digitise their services, highlighted how AI was enhancing the effectiveness of national healthcare delivery by enabling hospitals and doctors to treat more patients without compromising on treatment quality.

But like any advancement, the use of AI also comes with its share of challenges, especially concerns around whether enough caution is being maintained in using AI tools in settings where they are being used.

Given the fact that AI tools are currently being used in both the public and the private sectors, while some government entities have also collaborated with private players to leverage AI tools, “More awareness needs to be spread in medical and nursing colleges on AI and other enabling technologies so that doctors and health staff are attuned to adopt the technology in regular work,” said an official in National Informatics Centre (NIC), under the Union Ministry of Electronics & Information Technology.

Role in disease surveillance 

If AI solutions are aiding doctors with clinical decisions, they are not far behind when it comes to disease surveillance systems either.

Since 2022, the National Centre for Disease Control (NCDC) under the Union Health Ministry has been partnering with the Wadhwani Institute of Artificial Intelligence — India’s first AI research institute set up in 2018 by US-based philanthropist brothers Romesh and Sunil Wadhwani — that was inaugurated by Prime Minister Narendra Modi.

One of the main objectives of the Integrated Disease Surveillance Programme, under the NCDC, is to “develop and maintain an Information Communication Technology — for collection, collation, compilation, analysis, and dissemination of real-time data”. 

This work has reached a new level with Wadhwani AI providing tools to automate the process of disease surveillance.

“Our AI-powered outbreak monitoring solution automates the existing disease surveillance system in India, assisting the government in preventing disease outbreaks,” Shekar Sivasubramanian, CEO, Wadhwani AI, told ThePrint. 

Previously, this task involved manual monitoring of over 1,00,000 newspapers and digital media outlets.

“Now, our AI model scans large volumes of digital media, in English and several Indian languages, to track events indicating an upcoming outbreak. It maintains a comprehensive database of these events, generating alerts for public health officials to take proactive measures and prevent epidemics and pandemics,” he explained. 

The AI research centre has also developed a Clinical Decision Support System (CDSS), which consists of a suite of AI-powered tools that can assist doctors and other healthcare staff in diagnosing and supporting 1.5-2 lakh consultations on eSanjeevani — the telemedicine services run by the health ministry across the country— per day.

Remote consultations and more

There are also government agencies working to develop tools for India’s various problems in the healthcare sector. 

Sources in the NIC told ThePrint that they are working with nearly 1,100 hospitals across the country in areas of telemedicine and remote monitoring.

“AI-powered technologies enable remote consultations, monitoring of vital signs, and early detection of health issues, expanding access to healthcare, especially in remote areas,” said a source.

For instance, during the Covid-19 pandemic, NIC introduced Voice Support Services at the Lady Hardinge Medical College in Delhi over the telephone to help citizens reach doctors through a virtual out-patient department (OPD). 

This was done in collaboration with the college and the AI-powered service ensured that patients could just call and join the OPD queue without the hassle of confusing and complex processes like registering and taking an appointment online. 

At the All India Institute of Medical Sciences (AIIMS), New Delhi, which was declared an AI Centre of Excellence in healthcare by the government this year, work is already underway to develop tools for areas such as clinical aid to doctors and patient grievance redressal systems to enhance the patient experience, AIIMS authorities said.

Joydeep Ghosh also highlighted the extensive reach of AI. He cited examples such as identifying diabetic retinopathy, cardiac risk prediction using AI models, chatbot-based symptom assessment where chatbots interact with patients and provide initial assessments and recommendations, and predictive analytics for hospital resource management where AI is being widely used. 

Meanwhile, Raju from HealthPlix Technologies, which mostly works with private doctors working in standalone clinics, highlighted that their AI-led health tech tool, such as Electronic Medical Records, helps clinicians treat patients with less cognitive strain by giving them crucial and pertinent information at the point of treatment.

AI’s use in IVF 

According to Nitiz Murdia, managing director and co-founder of Indira IVF, within the realm of assisted reproductive technology (ART) too, emerging technologies have brought significant developments over the years and reshaped the way fertility treatments are done.

Embryo selection, which is a critical phase in procedures, such as in-vitro fertilisation (IVF), has been transformed by AI and machine learning, he said.

He explained that earlier, embryologists had to visually assess embryos for viability and the likelihood of clinical pregnancy, but now AI-powered software, trained on thousands of embryo images, can evaluate embryos with more speed and precision, eliminating human bias and providing standardised grading.

“Thousands of embryo images are used to train the algorithm using machine learning — enhancing its ability to assess quality, genetic factors, and implantation potential,” Murdia said. 

This advanced technology revolutionises embryo selection, ensuring more consistent and accurate evaluations for successful IVF outcomes, he added. 

“Further, genetic screening techniques, such as preimplantation genetic testing, further enhance embryo selection by detecting potential genetic anomalies that could lead to miscarriages,” Murdia highlighted. 

AI and drug discoveries

AI-powered technologies are also benefitting the area of drug discovery.

Mahesh Bhalgat, chief operating officer with Bengaluru-based Syngene International — a Biocon subsidiary that focuses on integrated research, development, and manufacturing services and serves global pharmaceutical and biotechnology companies — said that AI is making significant contributions to drug discovery and development by using its ability to analyse huge amount of data, find patterns, and generate important insights.

“Discovering a drug is a complex process that involves finding a biological entity, which can be targeted to alleviate the disease, and designing a molecule that can modulate the target with the right pharmacological activity and safety profile,” he said.

But Syn.AITM, Syngene’s proprietary platform, is designed to expedite data-driven drug discovery. 

According to Bhalgat, this platform integrates and learns from diverse chemical, biological and clinical data to provide better and faster outcomes and helps identify effective targets for a disease area and reduces the risk of later-stage failure for efficacy or safety reasons.

“It optimises properties and assesses toxicity to design molecules with the optimal balance of properties required for a successful drug,” he added. 

Challenges ahead

Though AI has the potential to transform the Indian healthcare industry by improving efficiency, accuracy, and accessibility, there are still many barriers and risks that need to be overcome to realise its full potential.

According to Pradeep Kumar Jain, chief customer officer, HealthEM.AI — a data platform company that uses data science and a vertical-first approach to enhance patient outcomes and optimise the cost of care — despite the undeniable benefits, the adoption of AI in the Indian healthcare industry is lagging. 

“It is essential to change this situation. All stakeholders must play their part, including the government, providers, startups, health-tech multinationals, and healthcare professionals.,” he said. 

Pointing to the global situation, Jain said that global health-tech companies are utilising India’s AI talent pool to develop solutions for global markets. 

“The government should incentivise these companies to make India a hub for AI innovation in healthcare, addressing local challenges and adapting solutions for global markets,” he added.

Then there are concerns that whether enough caution is being maintained in using AI tools in settings where they are being used.

In May this year, the World Health Organization issued a statement calling “for caution to be exercised in using AI-generated large language model (LLM) tools to protect and promote human well-being, human safety, and autonomy, and preserve public health”.

LLMs include some of the most rapidly expanding platforms such as ChatGPT, Bard, Bert and many others that imitate understanding, processing, and producing human communication, the statement said.

“It is imperative that the risks be examined carefully when using LLMs to improve access to health information, as a decision-support tool, or even to enhance diagnostic capacity in under-resourced settings to protect people’s health and reduce inequity,” WHO cautioned. 

( This is an updated version of the article. Joydeep Ghosh’s designation has been corrected.)

(Edited by Richa Mishra)