When I turned 60, I knew something wasn’t right. I lost weight. I felt drained. I had no appetite. I had a gut feeling — literally, abdominal discomfort — and a sense that something serious was going on beneath the surface. So I asked my doctors to test everything.
Full body scans. Colonoscopy. Endoscopy. Cardiac function tests. Every lab test I could get approved.
The gastroenterologist removed a couple of polyps but said nothing was wrong other than mild gastritis. The cardiologist suggested that maybe I was stressed or depressed. Then the Palisades fire destroyed our Los Angeles house, and my wife and I were displaced to Palm Desert, Calif. A few weeks later, I was suddenly in severe pain that lasted a weekend. Nothing stayed down.
By Monday morning, I was a patient in the emergency room in a new health system. These doctors — strangers completely unfamiliar with my history — saw my case with fresh eyes. Within days, they found what everyone else had missed: an aggressive form of blood cancer in my bone marrow related to multiple myeloma. They had caught it early, but it was already starting to affect not only my bone marrow but also my kidneys, gut, and heart if we did not stop it fast.
Lying in a hospital bed for nine days, I couldn’t stop thinking. Why couldn’t my first doctors find anything? Would AI have caught my disease sooner?
We have built a vast, intricate, overburdened system that tries, heroically, to deliver care to people with infinite variability and demand using finite resources and time. The system is overwhelmed. It is structurally incapable of giving every person the kind of thoughtful, individualized, adaptive care that they need. That’s not a failure of any single doctor. It’s the result of an ecosystem designed to manage the population, not empower the patient.
Before I was a patient, I built technology. Two decades ago, I helped pioneer remote patient monitoring and chronic disease management systems. Later, I built educational AI agents that could navigate complex data and help people make sense of the world.
Now that world was my body; I was the data. I developed a medical AI agent named “Haley,” created to use underlying foundation models from OpenAI, Google, Anthropic, and xAI, but with layers of medical context to guide the knowledge exploration in combination with a carefully prepared set of all my medical history. I fed Haley the exact same data that all those doctors had seen just weeks earlier. My full MyChart history. My labs. The imaging results. The doctor notes.
Within minutes, Haley flagged a concerning pattern: mild anemia, elevated ferritin, low immunoglobulins — signs of immune dysfunction and bone marrow issues. Haley recommended a “serum free light chains” blood test and bone marrow biopsy. None of this had been previously suggested. Same data, new insights.
Then I expanded the team. I built a panel of AI agents — an oncologist, gastroenterologist, hematologist, ER doc, and many more — all trained to think like their human counterparts. I ran the same case through each of them, one at a time. I created a synthesis agent, Hippocrates, to serve as my chairman of the board. He listened to them all and gave me a consolidated recommendation.
I had created my own virtual multidisciplinary medical team. They illuminated the path that my doctors had missed.
Diagnosis is only the beginning. The harder part is what to do next. Medicine today is built around the standard of care, the set of best-practice guidelines meant to treat the average patient in the average case. It’s what insurance covers. It’s what doctors are trained to default to. And often, it works just fine.
But no one is an average patient. Everyone’s disease — especially when it’s cancer — is different. And my cancer wasn’t average, since it was a particularly lethal variant if not stopped early.
I didn’t want to wait and see if the standard first-line treatment would fail. It had seemed to work for a few weeks, but then it started to slow down. According to my mathematical models, I was not going to get a full response. Damaging toxic misfolded proteins would be circulating in my system for months, damaging my organs a little more every day.
I turned back to my growing pile of new medical records, blood tests, immune function tests, and, most importantly, the bone marrow biopsy. Deep in the scanned PDF attachments, I found my cytogenetics report — the DNA fingerprint of my cancer cells. It showed a long list of biomarkers and details on genetic translocations found in the plasma cell clone that had gone rogue.
This told a very different story. I discovered that I had genetic variants that might make the standard treatment less effective for me. There were other newer, more aggressive regimens — off-label combinations — that my AI agents suggested I might respond to better. But no one had told me. That information wasn’t part of the first-line protocol. No one had the time to sift through all of the research.
AI, though, had all the time in the world. It helped me to analyze the significance of those mutations, cross-reference them with clinical literature, find case reports and studies, and even draft rationales for pursuing alternative treatment strategies. It did not replace my doctors, but it supercharged my decision-making. It helped me skip the trial-and-error stage and suggested a plan that made sense for me.
It was a little scary to rely completely on AI agents. Mine appeared to have profoundly deep medical knowledge, but they had only seen one patient: me. I needed to find human specialists. I tracked down doctors at academic medical centers and leading cancer centers involved in clinical trials that seemed relevant. I sought out consultations to give me feedback on what I had learned. I wanted a second, third, and fourth opinion on the treatment plan I found with AI.
The collaboration between my doctors and me, powered by AI, reshaped my treatment. Instead of the standard protocol, I’m now on Daratumumab plus Venetoclax, a targeted but off-label combination that hasn’t had full clinical trials, simply because there aren’t enough people with this rare disease. The response has been exceptional: My key cancer marker is back in the normal range.
I’ve now turned to rebuilding the immune system that chemotherapy stripped away. Guided again by AI, I’ve added prophylactic measures to protect and restore immune function. Weekly blood tests show it’s paying off.
The doctors who missed my diagnosis aren’t in the picture anymore. My new hematologists and oncologists have been curious and open about AI, and I am writing a case study that I hope at least one of them will co-author. They view what I am doing as part of the larger shift toward shared decision-making, where patients and doctors work together more collaboratively on care. All of them are very pleased with my remarkable response, and the results speak for themselves.
I’ve started to share the app I’m calling CureWise with other patients in a private beta, where they, too, are beginning to uncover profound insights. AI doesn’t replace anyone; it augments everyone. It can give patients the ability to understand more, doctors the ability to synthesize more, and the system the ability to respond faster, earlier, and more intelligently than ever before.
This approach doesn’t hand the keys to machines. Instead, it builds a partnership between human judgment and machine intelligence — a partnership that sees us as individuals. When I wanted to know what else was being tried around the world, AI helped me to learn the paths others had gone down. AI can keep up with the incredible pace of medical discovery much more efficiently than humans. It can find the signal in the noise, even across disciplines. It can connect an immunology paper from Japan, a gene-editing trial from Boston, and a retrospective analysis from Sweden, all before breakfast. When that knowledge is layered on top of your own data, it becomes something extraordinary: actionable intelligence.
I went from being a victim of the disease to a proactive participant at every level of my care.
We don’t need more disruption. We need to amplify wisdom at every level of care. We need systems that work with us so that fewer people fall through the cracks. Now, finally, we have the tools to do it.
Steve Brown is the founder and CEO of CureWise.
Source link
