When grief meets expertise, it can become something the world did not know it needed. Dr. Geetha Manjunath, an AI and supercomputing researcher with three decades of experience, lost two cousins to late-detected breast cancer — one of whom had received a clear mammogram — and chose to answer that loss with the only tools she had mastered. In 2016, she co-founded NIRAMAI Health Analytix, whose thermal imaging and machine learning platform has since screened over 400,000 women across India, finding more than 2,000 cancers without radiation or physical compression. Her story asks a quiet but endu
IISc Researcher Turns Personal Tragedy Into AI-Powered Breast Cancer Detection Tool
Two family members died from breast cancer after late detection, including one case where mammography failed to identify the disease.
Sometimes innovation begins with a problem that refuses to go away
Mark
What made you decide to leave a secure position at Xerox to start a company?
Mimi
It wasn't a business decision at first. It was a moral one. When you lose people you love to something that might have been caught earlier, you can't unsee that. The technical skills were already there—I just had to point them at the right problem.
Mark
Why thermal imaging specifically? There are other imaging technologies.
Mimi
Because it's non-invasive and it's accessible. A mammogram requires expensive equipment and trained technicians in major hospitals. Thermal imaging is portable. It doesn't hurt. It doesn't use radiation. If you want to reach women in smaller cities or villages, you need something that can travel.
Mark
How do you know the AI is actually detecting cancer and not just finding patterns?
Mimi
That's the hard part. We validated against clinical outcomes. We worked with medical institutions. The algorithm learns what cancerous tissue looks like in thermal data, but ultimately a radiologist still reviews the findings. The AI is a tool that helps catch what human eyes might miss, not a replacement for medical judgment.
Mark
400,000 women screened—does that number feel like success to you?
Mimi
It feels like a beginning. In India, millions of women have no access to any screening at all. We've detected over 2,000 cancers that might have been found too late. But there are still far too many women we haven't reached.
Mark
Do you ever think about your cousins when you're working?
Mimi
Every day. Not in a way that paralyzes me, but in a way that clarifies what matters. They're why this work exists.
El Pulso
- Breast cancer kills quietly in India — often because screening tools are painful, inaccessible, or, as in Manjunath's own family, capable of missing the disease entirely.
- The loss of two cousins in six months transformed an abstract public health problem into a personal reckoning that a career in corporate AI research could no longer sidestep.
- NIRAMAI's Thermalytix technology exploits a biological truth — that cancerous tissue generates distinct heat signatures — turning thermal cameras and machine learning into a radiation-free, compression-free screening alternative.
- Portability is the technology's quiet revolution: where mammography machines cannot travel, a thermal camera and a laptop can, reaching smaller cities and rural clinics historically left outside the screening net.
- With 400,000 women screened and 2,000 cancers detected across 250+ locations, the numbers have moved from clinical trial to documented impact — and the expansion continues.
When grief meets expertise, it can become something the world did not know it needed. Dr. Geetha Manjunath, an AI and supercomputing researcher with three decades of experience, lost two cousins to late-detected breast cancer — one of whom had received a clear mammogram — and chose to answer that loss with the only tools she had mastered. In 2016, she co-founded NIRAMAI Health Analytix, whose thermal imaging and machine learning platform has since screened over 400,000 women across India, finding more than 2,000 cancers without radiation or physical compression. Her story asks a quiet but enduring question: what is technical knowledge truly for, and when does it finally find its purpose?
Dr. Geetha Manjunath did not intend to enter medicine. She was a supercomputing and AI researcher whose career had taken her from C-DAC's early Indian supercomputing projects through seventeen years at Hewlett Packard Labs and then to a director role at Xerox Research. Her world was data, algorithms, and computing systems. Then, within six months, two of her cousins died from breast cancer. In one case, a mammogram had found nothing — and by the time the disease was discovered, it was too late. The second loss followed shortly after. The grief was not abstract. It was family. And it posed a question she could not set aside: could three decades of technical expertise be turned toward something that might have prevented this?
The scientific foundation was already there. Cancerous tissue has higher metabolic activity and disrupted blood flow, producing heat patterns that differ measurably from healthy tissue. A thermal camera could capture those patterns — no radiation, no compression, none of the barriers that make conventional mammography inaccessible or uncomfortable for many women. The harder problem was interpretation: thermal images are complex and noisy. That is precisely where Manjunath's background in machine learning became decisive. She and her team built algorithms capable of reading those patterns and flagging abnormalities for medical review. Early validation with a Manipal medical institution confirmed the approach was sound.
In 2016, she left corporate research and co-founded NIRAMAI Health Analytix with Dr. Nidhi Mathur. Their platform, Thermalytix, pairs thermal imaging with AI to screen for breast cancer in a way that is non-invasive and, crucially, portable. A thermal camera and a laptop can reach rural health centers and smaller clinics that a mammography machine never could. Since its founding, NIRAMAI has screened more than 400,000 women across 250 locations in India and detected over 2,000 cancers — women who received early diagnoses, who had options, who had time.
Manjunath's path is not a straightforward story of innovation. It is a slower, more human one: of expertise accumulated over decades without knowing where it would ultimately matter, waiting for a loss too close to ignore and a question too important not to answer.
Dr. Geetha Manjunath did not set out to solve breast cancer. She was a supercomputing researcher, then an AI scientist at corporate labs, working on problems that had nothing to do with medicine. But in the span of six months, two of her cousins died from breast cancer. In one case, a mammogram—the standard screening tool—had missed the disease entirely. By the time it was found, there was no time left. The second diagnosis came months later, and the outcome was the same. For Manjunath, the losses were not abstract. They were family. And they posed a question she could not ignore: could the technical skills she had spent three decades building be turned toward something that might have saved them?
Manjunath grew up in Bengaluru and studied computer science at UVCE before earning her master's degree and PhD from the Indian Institute of Science, where she graduated as a gold medallist. Her career had been built at the frontier of computing. In the 1990s, she was part of the team at C-DAC that developed India's early commercial supercomputing systems. She spent seventeen years at Hewlett Packard Labs as a principal research scientist, then moved to Xerox Research in India as a lab director, leading work in data analytics. She had the credentials and the experience. What she lacked was a reason to redirect them—until her family's tragedy gave her one.
The insight came from a simple observation: cancerous tissue behaves differently from healthy tissue. It has higher metabolic activity and altered blood flow patterns, which create measurable differences in heat. A thermal camera could capture these patterns without radiation, without compression, without the limitations that sometimes allow dangerous cancers to hide. The challenge was interpretation. Thermal images are noisy and complex. But that is precisely where Manjunath's decades in artificial intelligence and machine learning became essential. She and her team developed algorithms to analyze thermal patterns, to identify abnormalities that might warrant further medical evaluation. Early clinical work with a medical institution in Manipal validated the approach.
In 2016, Manjunath left corporate research. She co-founded NIRAMAI Health Analytix with Dr. Nidhi Mathur, a former colleague. The name stands for Non-Invasive Risk Assessment with Machine Intelligence. Their technology, called Thermalytix, combines thermal imaging with AI to screen for breast cancer. Unlike mammography, it involves no radiation and no physical compression—a significant advantage for women who find conventional screening painful or inaccessible. More importantly, the equipment is portable. A thermal camera and a laptop can go places that a mammography machine cannot. They can reach clinics in smaller cities, rural health centers, places where access to conventional screening has always been limited.
Since its founding, NIRAMAI has screened more than 400,000 women across over 250 locations throughout India. The technology has detected more than 2,000 cancers. These are not hypothetical numbers. They represent women who received early diagnoses, who had options, who had time. The company continues to expand, pushing the technology into regions where breast cancer screening has historically been out of reach. Manjunath's story is not a simple narrative of innovation born from tragedy. It is something more textured: a reminder that education and technical mastery do not always find their purpose in the moment they are acquired. Sometimes they wait. Sometimes they wait for a problem that refuses to be ignored, for a loss that demands to be answered. For Manjunath, decades of work in supercomputing and artificial intelligence found their meaning not in the labs where they began, but in the question her family's deaths forced her to ask.
Citas Notables
Could technology help detect breast cancer earlier and make screening more accessible?— Dr. Geetha Manjunath's guiding question after her family's losses
Quieres la nota completa?
Lee el original en The Times of India ↗
Technology