Demystifying AI in healthcare in India

With the help of AI, healthcare organisations can utilise algorithms for better clinical decisions and enhance the quality of the patient experiences they provide. Image: Shutterstock

Artificial Intelligence (AI) is transforming how healthcare is delivered in India. Health organisations have assembled vast data sets in various areas, including clinical trial data, health records, and images. AI technologies are ideal for examining these data and uncovering the patterns and insights humans cannot notice easily. With the help of AI, healthcare organisations can utilise algorithms for better clinical decisions and enhance the quality of the patient experiences they provide. AI as a tool is divided into different subsets, including deep learning and machine learning. While machine learning deals with developing algorithms for effective prediction and decision-making, deep learning is a subset of machine learning focusing on artificial neural networks that analyse vast amounts of data.

The origins of AI can be traced back to the 1950s when it was not fully conceptualised. Gradually, it gained prominence in the life sciences. The early 1970s witnessed a proliferation of research as AI launched itself in healthcare and was applied to biomedical problems. Eventually, AI found its mark in the clinical domain in the 1980s, where it is enthusiastically embraced today.

Source: Datascience.com

AI in Healthcare: The Global Picture

Over several years, AI has been making waves in global healthcare with emerging technologies, from blockchain to AI, demonstrating significant potential to alter how healthcare is delivered for good. The global AI picture entails a vast range of applications, including those in pharmaceuticals, medical technology, drug discovery, diagnostics, hospital operations, health insurance operations, clinical decision-making, and robotic surgery. A deeper dive suggests that the used cases span across deploying chat-bots for customer interactions, population health monitoring, care routing and care services, data-driven diagnosis, image-based diagnosis, clinical decision support, compliance monitoring, maintaining medical records, fraud prevention, capacity planning, personnel management, quality assurance, and training. The top global healthcare companies that lead in applying AI include GE Healthcare, Remedy Health, Oncora Medical, Google Health/ DeepMind, and Augmedix.

According to the Globe Newswire publication 2023, the value of AI in the global healthcare market was around $11 billion in 2021, which is anticipated to steeply increase to $188 billion by 2030 at a compounded annual growth rate of 37 percent between 2022 and 2030.
 

AI in Healthcare in India

The AI expenditure in India increased by over 109 percent in 2018, amounting to $665 million, and is anticipated to reach $11.78 billion by 2025, going on to contribute $1 trillion to India's economy by 2035.

The critical challenges of Indian healthcare are a low doctor-patient ratio, skewed distribution of expertise, not being affordable, inadequately trained staff, and delayed detection and diagnostic errors. As a powerful tool, AI can address many of these challenges. That technology in healthcare can make execution better and faster is a time-tested fact. In a developing country like India, the significant issues that AI and digital technology are trying to solve include diagnostics, making up for the lack of skilled resources and human capital, affordability, and accessibility.

Also read: Value-based care: A value proposition for patients, providers, and payors

Several startups from India attest to the power of AI technology. These include:

• HealthifyMe: It is a health-related digitally based wellness platform in Bengaluru. It adopts Ria, an AI-powered virtual app that helps users and answers their questions about fitness, nutrition, and health in ten languages.
• Columbia Asia Hospitals: It is one of the leading hospitals in Bengaluru, specialising in critical care medicine and bariatric surgery. It employs AI to automate processes, allowing doctors to record each detail the physician and patient communicate, providing visibility into trends. Furthermore, predictive analysis helps in the early detection of any ailments and aids in treating life-threatening conditions.
• PharmEasy: A healthcare startup in India, it offers an AI-based application that connects users with pharmacies. It uses a smartphone-based application that facilitates the seamless delivery of medicines. This application uses machine learning and tools for big data analysis.
• Aindra: It is an AI-powered medical technology healthcare startup. It employs an AI platform, Astra, that helps detect critical illnesses such as cancer. The company has developed a point-of-care detection system for cervical cancer that facilitates affordable and faster detection.
• Apollo Hospitals: One of the leading hospital chains across India, it has launched the first-of-its-kind AI-driven preventive health profile program called ProHealth. It uses a predictive AI algorithm that captures a patient's health status and predicts potential risks.
• Staqu: It is a Gurugram-based startup that uses an AI-based thermal camera to identify a person with a body temperature above 37 degrees Centigrade. The camera identifies multiple suspects parallelly, within a range of 100 meters.

However, as the healthcare domain explores the application of AI, misconceptions and misbeliefs have also increased. Understanding these misconceptions is particularly relevant in healthcare. In due course, two branches of AI developed in healthcare—physical AI and virtual AI. While physical AI involves using robots and devices to assist patients in delivering quality healthcare, virtual AI is characterised by deep learning, where the algorithms are created through repetition and experience.

Also read: Can HealthifyMe burn calories in a profitable way?

What is True AI in Healthcare?

AI assists in clinical research and process improvement, including quality control and increasing prediction accuracy through learning patterns. However, there ought to be more clarity about what AI is. Where can it be used, and where it cannot? The most common possibility of errors in AI is understanding the difference between false positives and false negatives.

What AI Entails?

• Machine Learning (ML): ML relates to the ability of a computer system to enhance performance with increased exposure to relevant information. The system discovers patterns and attempts to make estimates. Healthcare companies that leverage this technology include SigTuple and Niramai.
• Robotics: Robotics refers to creating efficient and intelligent robots that facilitate the execution of routine tasks without further interventions. Bengaluru-based Alpha Care, a healthcare startup, employs robotics to simplify patient services and streamline operational processes.
• Image Recognition: This consists of a computer's ability to recognise patterns in medical imaging for effective and timely diagnosis and treatment. Artelus, a healthcare startup, uses AI-powered image recognition for effective and early diagnosis.
• Speech Recognition: Various healthcare organisations are embracing speech recognition technology to provide easy-to-access transcripts of medical records. Examples include Augnito, a company that seeks to convert human voice to written text, further empowering healthcare practitioners to minimise errors and save time.
• Natural Language Processing (NLP): NLP can help scan clinical data and identify the right disease. Ahmedabad-based healthcare startup Maruti Techlabs employs AI-driven NLP for text processing to gauge the patient's healthcare needs.
• Rule-based Expert Systems: These are widely employed in healthcare as effective clinical decision support systems.
  

Where AI is Lagging Currently

A few domains have yet to involve significant involvement of AI. Some of them are mentioned below.

• Telemedicine: It enables audio or video appointments between a patient and the concerned healthcare practitioner. For instance, monitoring a patient's condition post-surgery through telemedicine does not involve any AI.
• Direct Patient Consultations: Direct patient consultations are typically done by the physical presence of the doctors or required medical practitioners. For instance, in hospitals, the doctors do the rounds. Patient dealing is still doctor-dependent in both out-patient and in-patient departments.

Conclusion

While AI has been a boon in healthcare, some of the apparent roadblocks to more widespread healthcare adoption still include blind spots in data access and collection, privacy issues, data misuse, and regulatory ambiguity. India has begun its journey in AI-based healthcare and is growing rapidly, yet there is a long way to go. The idealistic goals of universal healthcare delivery and affordability that can be achieved by AI-based healthcare, as of now, is a distant reality. Yet hope is on the horizon, as more and more innovations are coming up in AI-based healthcare services and may soon convert this dream into reality.

[This article has been reproduced with permission from the Indian School of Business, India]