by Heidi Happonen

Three first-year graduate students from Northeastern’s College of Professional Studies walked into a healthcare innovation Care-AI-thon at Lahey Clinic this spring prepared to compete. They just didn’t expect the competition to be entirely made up of professional teams and seasoned medical researchers. Or that they’d have to scrap 48 hours of work with 45 minutes left on the clock.

They won anyway.

Ruthvik Nath Bandari, Om Manojkumar Patel, and Yash Mahavir Jain — all enrolled in CPS’s Master of Science in Applied Artificial Intelligence program — are the AI Innovation Award winners for exceptional clinical insight and patient-centric design in Lahey’s inaugural Care-AI-thon. The event is a human-centered approach to a traditional hackathon, bringing together both clinical and patient perspectives to accelerate technology development and implementation planning.

As the only student team among five competing groups, they went up against three industry companies and a team from the UMass Chan-Lahey Medical School campus that included faculty. The challenge: Identify individuals who qualify for CT lung screening, which can detect cancer early. Their solution: An end-to-end automated system designed to identify lung cancer patients who qualify for potentially life-saving CT scans for lung cancer screenings and proactively reach out to connect them with care.

The team calls themselves BreatheWell. They call themselves something else, too. But that comes later.

A Problem That Needed Solving

Lahey’s Care-AI-thon posed two challenges: one focused on the clinical perspective – using electronic medical record (EMR) data to identify patients eligible for CT lung screening – and another centered on improving patient outreach and awareness among those who qualify for this potentially life-saving screening.

BreatheWell chose Track B, the patient-focused challenge, and was tasked with addressing a gap outlined by Dr. Carla Lamb, director of both Lahey’s interventional pulmonary medicine and incidental lung nodule programs as well as member of the Lahey CT lung screening program: many patients who qualify for low-dose CT lung cancer screening are often unaware of their eligibility.

The data lives inside the clinic’s own electronic health records. Lahey’s patient data contains key indicators, such as age and smoking history, that make someone a candidate for a CT lung screening. Converting that data into proactive patient outreach requires a level of automation not currently in place.

Knowing that, BreatheWell designed a system that pulls patient records, filters them against clinical eligibility guidelines, and then launches a staged outreach sequence — text or message in week one, a follow-up in week two, another in week three. If a patient still hasn’t responded, the system escalates to a phone call. And if that doesn’t land, it flags the case for a nurse to follow up in person.

“It’s a communications platform where you’re helping to connect at-risk patients with healthcare solutions that could save their lives,” Patel explained.

For Richard Meiklejohn, co-director of the Innovation Hub at Lahey Clinic, the Care-AI-thon was built around a deliberate question: not just what AI can do, but what it should do. And why.

“We’re very aware that AI can come up with things very quickly,” Meiklejohn said. “We just naturally, being a clinic and healthcare setting, wanted to be very thoughtful towards that.”

For Lahey, the answer started with lung cancer screening, a domain where the clinic already has deep roots. Lahey established the first accredited LDCT lung cancer screening program in the United States, and that legacy gave the Care-AI-thon a natural and consequential starting point.

“Lung cancer is the leading cause of cancer deaths. And the screening rates are really low,” Meiklejohn said.

Despite the existence of a proven, potentially life-saving protocol for CT lung screening, many patients are not aware they qualify and the gap between who qualifies and who gets screened remains enormous.

The Care-AI-thon’s two tracks were designed to attract different kinds of problem-solvers, and they did. The established commercial companies gravitated toward the clinical workflow side. The student team came at it differently.

“We got to see both ends of the spectrum. We got to think about it not just from the clinical end, but also the patient lens,” Meiklejohn said. “Because I think that’s really important.”

The Pivot That Changed Everything

The team almost didn’t get there.

In the weeks leading up to the Care-AI-thon, BreatheWell burned through late nights and final exams. They found out about the competition in April, right in the thick of end-of-semester deadlines. They spent the 48 hours before the event building and refining a solution, connected virtually through Discord, staying awake through the night to have something ready.

What they built was technically impressive. It just wasn’t what Lahey needed.

“We got to know in the discussion phase that what we were building, they don’t want that,” said Jain. “They want something entirely different.”

Bandari described the moment simply: “We just froze for a second. Our brains went blank.”

They had 45 minutes left.

“We were representing Northeastern,” Bandari said. “Everyone had put their trust in us. We didn’t want to break that trust because it was the first time we were representing Northeastern at something like this.”

A Lahey mentor assigned to provide guidance stepped in to encourage them. And then BreatheWell did what engineers do under pressure: they broke the problem down and started rebuilding.

“Dr. Sarju (Ganatra), director of cardio-oncology at Lahey and co-director of the Innovation Hub, gave us a phrase that stuck in my mind,” said Patel. “If you are given a problem for 60 minutes, spend 55 minutes understanding the problem. Because if you can’t understand it well, you can’t get to the solution.”

“If you are given a problem for 60 minutes, spend 55 minutes understanding the problem. Because if you can’t understand it well, you can’t get to the solution.”

Om Manojkumar Patel, quoting Dr. Sarju (Ganatra) of Lahey Clinic

What they built in those final 45 minutes won the competition’s Innovation Award.

For Meiklejohn, the students’ performance, including the last-minute pivot, illustrated something essential about what makes a hackathon valuable in the first place.

What distinguished the students, he said, wasn’t just the solution they delivered, it was the way they entered the room.

“The students come into this open-minded and were thoughtfully engaged as the medical team shared the challenges they were being asked to address,” Meiklejohn said. “That’s the bit where you see the energy and enthusiasm. The students are studying the latest in AI right now and bringing that along with their energy to the problem.”

The pivot BreatheWell executed in those final 45 minutes, he noted, was exactly the kind of adaptive thinking the Care-AI-thon was designed to surface: the willingness to scrap a plan and rebuild around a real problem, even under pressure.

The Path That Led Here

None of this came out of nowhere.

All three students came to Northeastern’s Applied AI program with both technical depth and real-world experience. Jain, who graduated in 2023 and spent two years as a full-stack developer while running his own AI consultancy, had been fielding requests from business owners wanting to automate and integrate AI into their operations. Healthcare, he said, became his primary focus. His computer vision coursework at CPS centered on brain tumor detection and 3D segmentation.

Patel brings four years of software development experience spanning MERN stack, Python, Django, AWS, and GCP. He is a contributor to one of the world’s most active open-source UI projects and is currently conducting research in AI applications for aviation safety. His previous coursework focused on using computer vision to detect drowsiness and sleep disorders. Work that, like his teammates’, lives at the intersection of AI and human health outcomes.

Bandari came to Boston having already founded a student AI club in India, trained two humanoid robots, and served as a machine learning intern. He pivoted from a background in cybersecurity to healthcare AI after arriving in Boston, a city, he noted, where the healthcare industry’s appetite for AI innovation is unmistakable. He is also the co-author of a research paper on diabetic foot ulcer detection, selected for the SPIE Medical Imaging Conference through a collaboration with a Harvard faculty member.

The three found each other through Northeastern and began competing together in hackathons, accumulating three wins before Lahey. Professor Rominder Singh, a CPS faculty member in regulatory sciences started guiding them and Uwe Hohgrawe, the college’s Associate Dean of Global Learner Access, Strategic Partnerships, and AI introduced the opportunity at Lahey to them.

“Being a campus on our doorstep, there’s a lot of potential there. We’re very open-minded. That’s where we see this hope for this relationship really blossoming with Northeastern.”

Richard Meiklejohn, Co-director of the Innovation Hub at Lahey Clinic

For Meiklejohn, the Care-AI-thon wasn’t a standalone event. It was a first step in a longer conversation between Lahey and institutions like Northeastern.

“There’s increasing pressure on what university education leads to,” he said. “The gap between academia and industry is being compressed and pressurized all the time. We hope through these types of initiatives that we can create avenues for collaboration and creation so students get the experience they need to then go forward in their careers.”

Lahey’s inaugural Care-AI-thon was deliberately scaled: roughly 60 attendees, a focused problem, and a carefully curated mix of commercial companies, academic researchers, and students. Meiklejohn described the format as an intentional choice.

“We pulled the plaster off, we got it done,” he said. “And I think we learned a lot.”

The proximity to Northeastern, whose Innovation campus sits close to Lahey’s Burlington location, makes that ongoing relationship a natural one.

“Being a campus on our doorstep, there’s a lot of potential there,” Meiklejohn said. “We’re very open-minded. That’s where we see this hope for this relationship really blossoming with Northeastern.”

What Comes Next

The win hasn’t slowed them down. At the time of this writing, BreatheWell had already registered for an AI hackathon at MIT, just days after their Lahey win. They have a year left in their graduate program, and all the momentum of a team that has figured out how to function under pressure.

When asked what advice they’d give to other CPS students considering opportunities like this, the answers came quickly.

“Accept the challenge,” said Jain. “Solve the challenge. Learn from the challenge. Develop yourself.”

“Interact with domain experts as much as you can,” Patel added. “Healthcare is not sorted. It has problems everywhere in the world that need to be addressed. Students can contribute.”

Bandari put it plainly: “Use every opportunity Northeastern gives you. That’s what makes you stand out.”

“Use every opportunity Northeastern gives you. That’s what makes you stand out.”

Ruthvik Nath Bandari

As for the team name: BreatheWell is what they called themselves for the Lahey Care-AI-thon. Their general team name, the one they carry into every competition?

“Three Idiots,” Bandari said, with a straight face.

The name comes from the beloved Bollywood film about three engineering students dismissed as underachievers — who go on to do extraordinary things. Given what just happened in a conference room at Lahey Clinic with 45 minutes left on the clock, the irony lands exactly as intended.