Dr Chandru further noted the importane of leverging the data available with Ayushman Bharat Digital Mission.
Published Aug 17, 2024 | 10:37 AM ⚊ Updated Aug 17, 2024 | 10:37 AM
Dr Vijay Chandru (left) at Dakshin Healthcare Summit. (South First)
If Artificial Intelligence (AI) is everywhere, how is healthcare utilising it? During the inaugural Dakshin Healthcare Summit in Hyderabad, Dr Vijay Chandru, founder of Strand Life Sciences, highlighted the opportunities AI brings to the healthcare industry.
Speaking on the panel discussion “Transforming Healthcare: New-age Technology, AI & Robotics,” organized by South First and TV9 Network, he pointed out that AI has great potential in healthcare, particularly in screening applications.
However, Dr Chandru cautioned, “Whether we can go as far as to say that we can build diagnostics is not entirely clear, to declare something as a diagnostic test, it would need to meet very high standards of specificity and sensitivity. You want to be extremely careful about declaring false positives.”
He added, “However, from a screening point of view, it is definitely promising, for example, in oral cancer screening.”
He continued by noting that, while AI could provide early indications in certain screenings, further investigation into molecular pathways and a well-defined clinical pathway are crucial to avoid unnecessary stress on patients, due to potential false positives.
Another significant positive, he added, is the Ayushman Bharat Digital Mission, which involves collecting large amounts of data and creating digital lockers. “The ability to leverage that data is important,” Dr Chandru added.
Speaking of the other side of the coin, Dr Vijay Chandru pointed out significant challenges associated with certain types of AI, particularly generative AI, due to its tendency for hallucinations.
Generative AI models learn the patterns and structure of their input training data and then generate new data that has similar characteristics.
He further cautioned, “If you start using generative AI, there’s a risk of creating hallucinated data, which can eventually overshadow the high-quality data we have. Therefore, it’s crucial to be very careful with technologies like generative AI, even though much of the excitement today revolves around it.”
Highlighting another challenge, Dr Chandru noted that these are often “black box” technologies.
Explaining further he added, “When you run a neural network, particularly a deep neural network, it produces an output, but how do you actually know what it’s saying? And how do you explain the conclusions it arrives at? Unless you can tie this into an understanding of biological pathways and mechanisms, it won’t meet any regulatory standard. So, I believe it’s essential to balance the pros and cons.”
Dr Chandru said that clinical decision support systems (CDSS)—that analyze data to help healthcare providers make decisions and improve patient care are evolving to a point where even primary healthcare settings could benefit from advanced AI.
These systems could significantly aid community health workers, such as ASHA workers, in defining care pathways for patients. “The primary healthcare community may need retraining to leverage these new technologies effectively.”
Citing the example of COVID, Dr Chandru noted, “Point-of-care testing, for instance, will need to become simpler to use. During COVID, RT-PCR tests were challenging to access in rural areas. However, portable technologies, such as COVID self-tests, now make it possible for community health workers to administer tests easily.”
He also said advancements in molecular biology, particularly genomics, are making significant strides. Point-of-need testing—where tests are performed at the location where care is required—represents a promising development. “This shift will be driven not only by AI but also by bioengineering and nanotechnology.”
Emphasising the importance of technology to make healthcare affordable, Dr Chandru said, ” Empowering the primary healthcare sector to utilize these advancements is crucial for achieving more equitable healthcare outcomes.”
“The convergence of tech and bioengineering will enhance the accessibility of universal healthcare without requiring massive investments to match Western health systems.”
Dr Chandru said that an innovative training program for endoscopy was developed at the Indian Institute of Science. “This program features a haptic box that provides force feedback, simulating the real experience of inserting an endoscope. Trainees practice within a virtual reality environment that immerses them in the procedure, allowing them to visualize working inside a patient’s throat.”
“The startup Mimic is behind this technology, combining virtual reality with specialized devices to enhance training. In addition, there are new developments in VR and robotics for training younger residents. For example, my daughter, a pediatric urologist, is experimenting with these technologies at CHOP to train first-year residents,” he said.
“These advancements present a significant opportunity to further develop and refine training tools for robotic surgeries, enhancing the skills and readiness of medical professionals,” Dr Chandru concluded.
(Edited by Sumavarsha Kandula.)
