By: Kalyan Kolachala
Last updated : September 10, 2026 8:27 am
The future is unlikely to be a choice between doctors and AI, but rather a closer integration of clinical expertise and computational intelligence
For much of modern medicine, the central question has been: what treatment works best for most people with a particular disease? Clinical guidelines, drugs and treatment protocols have largely been built around population level evidence. This approach has transformed healthcare, but it also rests on an important reality: patients with the same diagnosis can have very different biology, risk profiles and responses to treatment.
Precision medicine is changing that equation. The focus is gradually moving from treating the average patient to understanding the characteristics of the individual. Artificial intelligence is becoming an important part of this transition because it can analyse large and diverse datasets and identify patterns that may be difficult to detect through conventional methods.
The significance of AI here is not simply speed or automation. Its greater potential lies in connecting information across genetics, imaging, clinical history, laboratory results and other sources to support decisions that are more specific to the patient.
Cancer is becoming a test case for hyper-personalisation
Cancer perhaps illustrates this shift most clearly. A tumour is not simply defined by the organ in which it develops. Tumours can carry different genetic mutations, molecular characteristics and immune signatures, even when they occur in the same organ.
Precision oncology already uses molecular profiling to identify mutations that may make a tumour more responsive to particular treatments. AI can extend this process by helping analyse genomic information, identify patterns across tumour characteristics and support the prediction of treatment response.
The next step is moving towards hyper-personalised treatment, where therapeutic decisions are increasingly shaped by the unique molecular characteristics of an individual patient's tumour.
Personalised cancer vaccines offer a striking example. In these approaches, sequencing can identify mutations unique to a patient's tumour and determine which may produce neoantigens capable of triggering an immune response. AI can assist in predicting which of these neoantigens are most likely to be recognised by the immune system.
Recent developments in personalised mRNA vaccines have brought this concept closer to clinical reality. In August 2026, a Phase III trial of a personalised mRNA cancer vaccine developed by Moderna and Merck reported a reduction in the risk of melanoma recurrence when used alongside immunotherapy. The vaccine is designed around mutations specific to each patient's tumour rather than being administered as a standard treatment across a patient population.
However, the progress has not been uniform. A separate Phase II trial of an individualised mRNA vaccine in colorectal cancer was terminated in August 2026 after an independent review found that it was unlikely to improve survival. That contrast is important. Precision medicine is advancing through evidence, validation and failure, not simply through technological breakthroughs.
Personalised medicine has an older Indian parallel
The idea of tailoring healthcare to the individual is not entirely new. Long before genomic sequencing and machine learning, traditional Indian systems of medicine developed frameworks that placed considerable emphasis on individual constitution.
In Ayurveda, the concept of Prakriti describes an individual's constitution, traditionally understood through combinations of Vata, Pitta and Kapha. Ayurvedic approaches use this classification as one factor in determining aspects of diet, lifestyle and treatment.
The comparison with modern precision medicine should not be overstated. Their scientific foundations, methods of evidence and clinical frameworks are fundamentally different. Yet there is an interesting conceptual parallel: both challenge the assumption that the same intervention should necessarily be appropriate for every individual.
The difference today is the scale of measurable information available to modern medicine. Genomic sequencing, electronic health records, imaging, biomarkers and continuous monitoring can provide biological and clinical signals that were previously unavailable. AI can potentially bring these signals together in ways that were not previously possible.
For India, this creates an interesting opportunity to examine how the longstanding emphasis on individual variation intersects with modern data driven healthcare, while keeping the comparison grounded in evidence.
From diagnosis to continuous prediction
AI's role in precision medicine also extends beyond selecting treatments. It can potentially change when healthcare interventions occur.
Conventional healthcare often responds after symptoms appear or disease has progressed sufficiently to become clinically apparent. Predictive models can examine multiple risk factors and identify patients who may require earlier investigation or closer monitoring.
In cardiology, oncology, diabetes and neurological disorders, researchers are exploring models that combine clinical histories, imaging, laboratory measurements and other data to estimate individual risk. Wearable devices and remote monitoring could add another dimension by providing information about how a patient's condition changes over time.
This could eventually lead to a more adaptive model of healthcare, where treatment is not viewed as a fixed decision made at one point in time. Instead, new information could continually inform assessments of risk and treatment response.
The harder problem is turning intelligence into outcomes
The biggest challenge for AI in precision medicine is no longer whether algorithms can identify patterns. It is whether those patterns can reliably improve clinical outcomes.
Healthcare data remains fragmented, inconsistent and often incomplete. An algorithm trained on one population may not perform equally well in another. Bias in training data can translate into unequal outcomes, while models can lose accuracy as patient populations and clinical practices change.
There is also a practical question around time and accessibility. A treatment tailored to an individual's tumour may be scientifically compelling, but it becomes less useful if producing that treatment takes longer than the patient's condition allows or makes it inaccessible to large sections of the population.
This is why interoperability, data quality, clinical validation and governance will matter as much as advances in AI itself.
The future is personalised, but not algorithm led
Precision medicine should not mean replacing clinical judgement with machine generated recommendations. An algorithm can identify a pattern, estimate a risk or suggest a potential treatment pathway, but clinicians still need to interpret that information alongside symptoms, patient preferences, comorbidities and circumstances that may not be fully captured in a dataset.
The future is therefore unlikely to be a choice between doctors and AI. It is more likely to involve a closer relationship between clinical expertise and computational analysis.
The larger shift is philosophical as much as technological. Medicine is moving away from asking only what works for the majority and towards asking why patients differ, how those differences can be measured and how treatment can respond to them.
AI may provide the analytical infrastructure for that transition. Personalised cancer therapies and mRNA vaccines show what is becoming technically possible, while concepts such as Prakriti offer a reminder that recognising individual differences in health has deeper roots than modern technology.
The real measure of precision medicine, however, will not be how personalised a treatment sounds. It will be whether that personalisation consistently leads to better decisions, better outcomes and care that is genuinely appropriate for the individual patient.
About Author
Kalyan Kolachala is the Managing Director for SymphonyAI Group India, bringing over 30 years of extensive experience in enterprise and SaaS product development across leading technology organizations. An alumnus of the prestigious Indian Institute of Technology, Kharagpur, Kalyan has demonstrated expertise in AI/ML, generative AI, SaaS, cloud technologies, and scalable product architecture.
*The author’s views are his own and do not necessarily represent those of the publisher.