What if surgeons could navigate the brain with the precision of a GPS, knowing not just where a tumor sits, but exactly which functions lie in its path? That question sits at the heart of a growing field that is at the intersection of neuroscience and artificial intelligence.
Dr. Rimjhim Agarwal, co-founder and CTO of BrainSight AI, who has a PhD in computational neuroscience, and is one of India’s most compelling voices in neurotech – joins Ramakrishnan Venkatraman, Director of SOLIDWORKS and 3DEXPERIENCE Works, Dassault Systèmes India on 3DEXPERIENCE Innovation Matters to map the terrain of a field that is moving from research curiosity to clinical reality. The conversation covers surgical precision, data ethics, India’s unique data advantage, and the future of personalized brain stimulation therapy.
“Neurotech is still going to see the revolution that fintech has just seen. This is the decade when that acceleration is happening”.
– Dr. Rimjhim Agarwal
Why This Decade Belongs to Brain Health
The brain is the last great, unmapped frontier in medicine. Unlike the heart or the lungs, whose functions can be reasonably generalized across patients, the brain’s functional architecture is radically individual – the regions governing language, vision, and motor control are located differently in every person. For neurosurgeons, this is a daily clinical constraint. Removing a tumor without knowing precisely where its boundaries intersect with functional tissue risks outcomes that are, in the most literal sense, irreversible.
At the same time, an extraordinary convergence is underway. Brain-computer interfaces, wearable neural sensors, AI-driven diagnostics, and personalized stimulation therapies are all maturing simultaneously. The analogy Dr. Agarwal reaches for is fintech and the UPI revolution – a transformation that looked incremental until, suddenly, it wasn’t. Neurotech, she argues, is at exactly that pre-inflection moment. The decade ahead will determine which foundations are built, and by whom.
From Images to Intelligence: What a Brain Map Actually Does
PPerspective from Dr. Rimjhim Agarwal, Co- founder & CTO, BrainSight AI
The fundamental problem in neurosurgical planning is not image quality but interpretation. A standard MRI shows anatomy. What it does not show is function: which region governs this patient’s speech, which tract carries motor signals to their left hand, which boundary, if crossed during resection, will cost them something they cannot recover. Images are inputs. What surgeons need are decisions and no amount of resolution improvement in conventional imaging closes that gap.
BrainSight’s platform processes three MRI modalities for each patient – structural, functional, and diffusion imaging – and fuses them into a single, patient-specific functional map. Dr. Agarwal likens it to Google Maps: structural MRI gives you the road network, diffusion MRI gives you the route from A to B, and functional MRI gives you live traffic. Together, they tell a surgeon not just where the tumor is, but which neural highways run through it, and which path to take to remove the maximum amount of tissue while preserving the functions that define that patient’s life.
Getting clinicians to trust this output required a deliberate approach to adoption. Dr. Agarwal is frank about the early strategy: rather than leading with the platform’s transformative potential – which surgeons, reasonably, would have been skeptical of – BrainSight started by offering incremental, tangible value on top of existing MRI workflows. Trust was built from the ground up, one clinical validation at a time. That discipline, she argues, is what separates neurotech companies that reach the market from those that remain perpetually in development.
“In brain health, you cannot just tell clinicians that your technology will revolutionize their practice. You start with what is immediately useful on their MRI, build their understanding step by step, and earn the right to go further“.
– Dr. Rimjhim Agarwal.
Ethics at the Centre: Navigating Data Rights in the Most Sensitive Domain in Medicine
Perspective from Ramakrishnan Venkatraman, Director, SOLIDWORKS, Dassault Systèmes India
Brain data is categorically different from most other forms of health data. It does not merely describe a condition – it reveals, in extraordinary detail, how an individual thinks, processes, and functions. The ethical frameworks governing its collection, storage, sharing, and commercial use are still being written, and the gaps between jurisdictions are significant. The US has established but evolving medical AI regulations. Europe’s frameworks address healthcare data broadly but have not yet developed neuroscience-specific provisions. India is building its regulatory infrastructure in real time.
Operating across hospital networks in multiple Indian cities, this is not a future compliance question but an daily operational one. Three MRI modalities per patient generate a substantial and highly sensitive data volume. Anonymization must be maintained without breaking the identity trail that ensures a patient’s three scan types are correctly matched. HIPAA compliance, SOC 2 certification, and robust IP and IT governance are prerequisites for clinical trust, and trust is the currency that matters in this domain.
The broader point is structural: neurotech companies that treat ethics and regulation as obstacles to manage will consistently underperform those that treat them as design constraints. Compliance built into architecture from the beginning is more robust, more credible to clinical partners, and ultimately faster to market than compliance retrofitted after the fact.
“The ethical dilemmas in neuro AI are not peripheral to the technology – they sit at the center of it. How data rights, consent, and neuro privacy are handled will determine which companies earn the trust to scale“.
– Ramakrishnan Venkatraman
India’s Data Advantage: Why Diversity Is a Scientific Asset
One of the most underappreciated arguments in this conversation is demographic. India’s population diversity – across genetics, diet, climate exposure, and health profile – produces neuroimaging datasets with a variability that few countries can match. Models trained on Indian hospital data generalize reliably to other geographies, precisely because the variation within India’s patient population is broad enough to encode much of the world’s variability within it.
This matters because generalization is one of the hardest problems in medical AI. A model trained on a narrow patient cohort – even a large one – will fail on populations it has not seen. India’s diversity is, in this sense, a scientific infrastructure advantage: it compresses the validation work required to deploy globally. Combined with the sheer volume of clinical footfall in Indian hospitals, which accelerates data collection and iteration cycles, and the improving data quality standards among India’s clinical research community, the country is positioned to become a primary source of the high-integrity neuroimaging datasets that will underpin the next generation of global brain health tools.
“India’s data diversity is not just a local advantage – it is a global one. The variability within our patient population means that what we build here can generalize to the world in ways that more homogeneous datasets simply cannot“.
– Dr. Rimjhim Agarwal
The Virtual Brain: Simulation as the Foundation for Safer Therapy
The most forward-looking thread in this conversation concerns therapies that have not yet reached clinical scale, but whose foundational challenge is already well understood: non-invasive brain stimulation. Techniques that use mild electrical or magnetic signals to target specific neural regions are among the most promising frontiers in neurological treatment, but their precision depends entirely on knowing how stimulation energy will travel through a specific individual’s brain structure. Because head geometry varies significantly between patients, a stimulation protocol calibrated for one person may be ineffective or even counterproductive in another.
The solution is simulation. By building a patient-specific virtual brain twin from their own imaging data and running physics-based models of stimulation energy flow through it, clinicians can validate a therapy protocol in silico before it is ever applied to a real patient. This is a fundamentally different approach to therapeutic precision, one that treats the virtual twin as an indispensable step in the clinical workflow rather than an optional research tool. It is also a principle that extends beyond brain stimulation: any therapy whose outcome depends on individual anatomy stands to benefit from the same simulate-before-you-treat discipline.
This is where the partnership between neurotech startups and simulation platform providers becomes most consequential as a shared commitment to the idea that a therapy should be proven safe in a virtual patient before it is tested in a real one.
“You are essentially building a virtual twin of the brain. The question is how the 3DEXPERIENCE platform enables that – not just for visualization, but for simulating what will actually happen when a therapy is applied, before you apply it“.
– Ramakrishnan Venkatraman
Fail Fast, Build Trust, Think at Scale
Three principles emerge from this conversation for the researchers, startup founders, and policymakers building India’s neurotech ecosystem:
- Invest in the right data, not just any data
- Be resilient in the face of the long validation cycles that brain health demands
- Fail fast, because clinical professionals are more willing to engage with honest iteration than with overpromised technology.
The infrastructure this requires – simulation environments that can model complex biological systems, visualization tools that make 3D brain data navigable for clinicians, and clinical data management platforms that maintain integrity through every stage of a trial – is not optional. It is the precondition for the field’s ambitions to be realized safely. The organizations building that infrastructure, and the startups partnering with them to put it to clinical use, are the ones who will determine the pace at which neurotech moves from promise to practice.
“Fail fast is the key. Clinical professionals are more helpful than people think – they will validate faster if you engage them honestly. Do not be scared of the failure you are going to encounter“.
– Dr. Rimjhim Agarwal
3DEXPERIENCE Innovation Matters continues to spotlight the researchers, clinicians, and technologists building India’s most consequential healthcare innovations, from the lab to the last mile.
More Ways to Engage:
🎬 Watch on YouTube
🎧 Listen on Spotify
👥 Follow us on LinkedIn
📩 Get in Touch
Dassault Systemes India