How AI helped a founder beat rare cancer after misdiagnosis risk

At a glance:

• Conno Christou, a 35-year-old founder, was diagnosed with aggressive non-Hodgkin’s lymphoma after a routine blood clot check revealed an 11-by-11-by-8 cm tumor.
• He used AI tools like Claude to analyze medical scans and avoid unnecessary radiotherapy by identifying a rare thymus rebound phenomenon.
• Christou’s data-driven approach, including wearable tech and 12 expert opinions, led to an 85% success chemotherapy regimen and full recovery.

The diagnosis that changed everything

Conno Christou doesn’t leave things to chance. He tracks his sleep with a Whoop band, cross-references it with an Oura ring, and gets nearly 100 biomarkers checked every year. For four consecutive years, he followed protocols from longevity researchers like Peter Attia and Rhonda Patrick, optimizing supplements, circadian rhythm, and protein intake. At 35, building his second company, he was as dialed-in on health research as anyone he knew. His last checkup in 2025 was green across the board. “It was the best I’d had in years,” he says.

Then, after a workout, his arm swelled. A week passed before he saw a doctor, who found two blood clots and scheduled surgery. Pre-op exams changed everything. A doctor walked back into the room and told him the procedure wasn’t happening. “We see an 11-by-11-by-8 centimeter mass behind your sternum,” the doctor said. A biopsy confirmed an aggressive, fast-growing non-Hodgkin’s lymphoma — a rare diagnosis affecting roughly one in 420,000 people, caused by a random genetic mutation with no connection to lifestyle, diet, or stress. The tumor had only existed for about three months. In three more weeks, it would have reached stage four. “Lucky in my unluckiness,” Christou told this editor from his home in Athens. “It was only found because I went in for something else entirely.”

Seeking multiple opinions and choosing the aggressive treatment

What followed was an education in the limits of the medical system and what a determined patient can do with available tools. His first oncologist recommended the lighter of two chemotherapy regimens. Christou booked his first infusion three days out. Then, the night before, he sought a second opinion. That doctor recommended the harder regimen — continuous in-hospital infusion, cycling every three weeks across six months — citing Christou’s specific pathology. The lighter treatment carried roughly a 60% success rate; the aggressive one brought that to around 85%. Two world-class doctors. Diametrically opposite recommendations.

“As founders, we hold the wheel,” Christou says, noting the propensity of people to accept what they’re told. “You hear many things. You don’t have to follow the first advice.” He didn’t opt for just the second physician’s advice either. Over two days, he gathered 12 opinions — drawing on his professional network, reaching out to hematologists and oncologists in the US and abroad. Eleven to one voted for the harder path. He took it. The decision felt logical, not brave. He was already data-driven, and now the stakes felt existential.

Using AI as a medical tool during treatment

Over six months of treatment, Christou approached chemotherapy like building a company — a marathon of sprints, each with a finite cycle and data points. He had done a mandatory 25-month military service in Cyprus at 18 and borrowed from that experience. He wore his Whoop throughout, finding it accurate at predicting immune system lows before symptoms arrived. He kept a symptom journal using voice transcription, logging every shift, side effect, medication, and counter-medication. He narrowed focus to three variables: sleep, nutrition, and psychology. (“It moves the needle more than anything,” Christou said. “I never asked ‘why me’ — not once. That question has no useful answer.”)

He fed all data — blood results, scan data, wearable output, journal entries — into Claude. He’s far from alone in turning to chatbots for medical guidance. A March poll found a third of American adults use them for health info. Stories suggest AI delivers what the system couldn’t. Experts urge caution; Danielle Bitterman, clinical lead for data science at Mass General Brigham, told the New York Times that general-purpose chatbots are frequently wrong and “have not been thoroughly evaluated” for personalized diagnoses. Christou doesn’t disagree. “It didn’t replace the doctors,” he says, but it “helped me ask the right questions.” For a rare condition, access to a model that absorbed full medical literature was not the same as a Google search.

The ambiguous scan and AI's intervention

The model proved critical at treatment’s end. His final PET scan — used to detect active disease — came back ambiguous. His oncologist discussed a second line of therapy, potentially radiotherapy near his heart and lungs. It was alarming. Christou did his homework, reading that for this lymphoma, the false-positive rate on end-of-treatment PET scans is around 60%. “It’s 2026,” he says. “Sixty percent.” He fed all three PET scans and his MRI into Claude, which flagged a known but overlooked phenomenon: in patients under 40 recovering from this lymphoma, the thymus gland can reactivate after chemotherapy, showing up as active disease. Given his age and scan characteristics, the model put the probability at roughly 90%.

He sought three more opinions. The fourth doctor confirmed it: thymus rebound. No active disease. No radiotherapy needed. He was clear. Christou is still processing the year’s impact on his health, work, and time. He built Keragon, his AI-powered platform for medical administrative automation, before this happened. Going through the system as a patient gave him new perspective. He watched nurses and doctors buried under non-care tasks. He received the same chemo protocol as an 80-year-old woman, side effects managed through cascading drugs. He says we’ll look back at this era and cringe. He takes Sundays off now, trying to be present. A VC friend’s advice — “Be happy now” — kept replaying during treatment. He says it’s among the hardest things but finally appreciates its importance. He’d be happy to talk to anyone going through something similar. “It’s not happening in 10 years,” he says of AI’s potential. “It’s happening today.”

Reflections and future implications

Christou’s experience highlights both the promise and peril of AI in healthcare. While tools like Claude can process vast datasets and identify patterns humans might miss, experts caution against overreliance. The 60% false-positive rate in PET scans underscores systemic gaps in diagnostic precision. For rare conditions, where specialists may see only one case annually, AI’s access to comprehensive medical literature offers a unique advantage. However, the lack of rigorous evaluation for personalized diagnoses raises ethical and practical concerns.

Christou’s recovery also reflects broader shifts in patient agency. His refusal to accept a single opinion mirrors trends in precision medicine, where genomic data and AI-driven insights are reshaping treatment paths. Yet his story raises questions about accessibility: not all patients have networks to consult 12 doctors or access to advanced wearables. The future of AI in healthcare may hinge on balancing innovation with equitable access, ensuring tools augment rather than replace human expertise.

His company, Keragon, now operates with renewed urgency. Automating administrative tasks isn’t just about efficiency — it’s about freeing healthcare workers to focus on care, reducing the very inefficiencies Christou witnessed. As AI becomes more integrated into clinical workflows, stories like his may redefine how patients and providers collaborate, blending data-driven rigor with human judgment.