Empowering Early Detection: How Carenostics’ AI Bolsters Clinical Decision Support. Kanishka Rao.

In an earlier episode of Digital Health Interviews, host Alex Koshykov sits down with Kanishka Rao, Co-Founder and COO of Carenostics, to discuss how artificial intelligence is reshaping early diagnosis and why the future of chronic disease care must be proactive, not reactive.

With a background in health policy, data, and business operations, Kanishka offers a fresh perspective on AI’s role in modern care delivery. But for him, the mission is also deeply personal: his grandfather’s sudden death from undetected kidney failure led him to ask a simple but powerful question—“What if we’d caught it sooner?”

“80% of people with chronic kidney disease don’t know they have it,” he says. “If we can find them earlier, we can completely change their health trajectory.”

A Family Vision Rooted in Real-World Impact

Kanishka co-founded Carenostics with his father, a prolific AI scientist with over 70 patents in healthcare. Their goal: use machine learning to detect chronic diseases like CKD, asthma, and heart failure long before clinical symptoms appear.

They knew that the data already existed—in lab results, medications, diagnosis codes, and clinical notes—but it wasn’t being used effectively. Carenostics was born to change that.

“Doctors want to help,” Kanishka says. “But they’re overwhelmed. They don’t have time to manually sift through every possible risk factor. Our job is to bring the signal to the surface—at the right time, in the right way.”

From NSF Grant to Real-World Deployments

Carenostics’ early traction came from winning a National Science Foundation (NSF) grant that enabled the team to build a working prototype. That led to collaborations with academic medical centers like the University of Pennsylvania and Hackensack Meridian Health, where they validated their model on real-world data. The result? A predictive algorithm that achieved 95% accuracy (AUC) in identifying undiagnosed chronic kidney disease.

But the company’s go-to-market strategy wasn’t limited to providers.

Instead, Carenostics pursued strategic partnerships with pharmaceutical companies, including Bayer, which had a shared interest in detecting diseases earlier. By demonstrating how early diagnosis could expand treatment access and improve outcomes, Carenostics unlocked a commercial model where patients, providers, and pharma all benefit.

“We wanted to build something sustainable,” says Kanishka. “That meant not just building tech that worked, but finding stakeholders who wanted it to work.”

What’s Under the Hood: AI That Clinicians Can Trust

Unlike many AI startups chasing the generative trend, Carenostics focuses on clinical-grade, reproducible machine learning models. Their platform integrates with existing EHR systems and uses a mix of structured and unstructured data—labs, meds, vitals, problem lists—to flag high-risk but undiagnosed patients.

“Generative AI is great for many things, but we don’t use it for prediction,” Kanishka explains. “The stakes are too high, and the risk of hallucinations is real. We focus on validated models that providers can rely on.”

The system is designed to fit into existing clinical workflows, not replace them. Alerts are sent to physicians or care managers when a patient meets specific thresholds, enabling timely follow-up and testing without disrupting the care team’s routine.

Tackling the Incentive Problem in U.S. Healthcare

One of the interview’s central themes is the misalignment of financial incentives in healthcare, particularly in the U.S. While value-based care is growing, most reimbursement still rewards treatment over prevention.

Carenostics addresses this head-on by building flexible partnerships. In some cases, health systems pay directly. In others, pharma companies subsidize deployment in exchange for anonymized, aggregated insights about patient populations.

“The biggest challenge isn’t the tech—it’s the economics,” Kanishka says. “But if you bring the right partners together, you can create a model where everyone wins.”

This pragmatic view of the market has helped the company scale more quickly than many clinical AI startups, which often get bogged down in pilot purgatory.

Scaling the Platform: What’s Next for Carenostics

Now, the team is expanding beyond nephrology and pulmonology into cardiology, pediatrics, and other chronic disease areas. Their goal is to generalize the platform while maintaining the precision that made their early models successful.

Operationally, Kanishka is focused on building a leadership team that can scale with the company. As Carenostics prepares for its Series A, he’s transitioning from founder-led sales to scalable growth operations, refining go-to-market strategies, and strengthening enterprise delivery infrastructure.

“We’re not just building a product. We’re building a system of trust between patients, providers, and data.”

Looking Ahead: Data-Driven, Clinically Rooted, Patient-Centric

Kanishka envisions a healthcare future where clinical decisions are routinely augmented by AI, but always grounded in human judgment. He doesn’t believe that algorithms will replace doctors, but he’s convinced they can empower them.

“There’s so much data already in the system. Our job is to make it useful—to help doctors do what they already want to do, just better and faster.”

Final Thoughts: Building With, Not Just For, Clinicians

What advice would Kanishka give to other digital health founders?

“Build with your users, not just for them. Sit in on clinics. Listen to care coordinators. Understand how things actually work before you try to fix them.”

And above all, don’t chase the trend. Chase the need.

“AI is just the tool,” he says. “The real innovation is building something that actually makes a difference in real people’s lives.