Artificial intelligence and digital automation are rapidly redefining the delivery of diagnostic services in India, with a strong focus on improving accuracy, access and operational efficiency, experts said during a panel discussion titled “AI, Automation & Digital Diagnostics — Redefining Accuracy, Access & Efficiency.” The session was moderated by Dr. Avi Nahar, Chief Operating Officer, 1Cell.Ai.
Dr. Anirudh Kohli, CEO and Head of Department – Radiology and Imaging, Breach Candy Hospital, emphasised, "AI is not a luxury; it’s an evolutionary trend in healthcare. Over the past decade, we’ve seen technology progress from radiology information systems for scheduling and reporting, to dictation software, and now to advanced automation on diagnostic equipment—measuring fractures, labeling scans, and performing value assessments. AI has become an integral part of our daily workflow, augmenting intelligence rather than replacing doctors. While automation may reduce low-level tasks, it enhances efficiency, improves processes, and supports growth in both volume and quality of care. AI is here to stay, and its impact on healthcare is both practical and transformative."
Dr. Barnali Das, Lead consultant- Laboratory Medicine, Kokilaben Dhirubhai Ambani Hospital & Research Centre said, “Initially, we were hesitant to accept AI, machine learning, and deep learning in diagnostics. But as Einstein said, life is like riding a bicycle—to keep balance, you must keep moving.
In laboratory practice, we are riding that bicycle with two wheels: quality and cost. Quality means accuracy, precision, sensitivity, and proper validation of every new test or automation. Cost is driven by manpower, methods, and machines. The balance between these two is enabled by AI and digitisation.
We started small by implementing AI-based electronic alerts for acute kidney injury. Even minor rises in creatinine are often missed in routine practice. With machine learning models—after rigorous validation by pathologists—we achieved nearly a 30% increase in detection through AI-enabled alerts.
The key message is this: AI works only when it is validated, understood, and used responsibly. Basic AI literacy is essential, so we don’t blindly accept black-box models. With the right inputs and oversight, AI can significantly improve diagnostic outcomes.”
Anoop Arora, Founder & CEO - IT Dose Infosystems stated, “We talk a lot about AI at medical conferences—its algorithms, use cases, and outcomes—but we often miss the most important piece: data.
Without high-quality, standardised, and governed data, AI cannot deliver reliable results. Before adopting AI, we must focus on three basics: quality data, data standardisation, and strong audit and governance frameworks.
In India, thousands of labs use different test names, ranges, and systems. Without standardisation and interoperability, AI algorithms cannot work effectively. Initiatives like ABDM are pushing us in the right direction, but we need broader adoption.
If we want meaningful AI outcomes, we must first fix the inputs. AI is only as good as the data it runs on.”
Naman Garg, Director, Medionn Diagnostics Limited said, “We’ve already heard how AI is transforming radiology, and I’d like to share what we are doing on the ground.
Through PPP models, we are implementing AI-based CT calcium scoring in district hospitals in Bihar—something many people never imagined would happen in remote parts of eastern India. Ideally, this requires expensive 128-slice CT scanners, which are not feasible for government hospitals. Instead, we are using 16-slice scanners supported by AI to generate predictive cardiac risk scores.
Accuracy may not be perfect yet, but the use case is real. We cannot wait for 100% accuracy—AI adoption has to start now if we want to improve access and outcomes.”
