As a (resident) radiologist, I’ve seen that it can take weeks for patients to get into onc after I call a cancer on their Emergency Room scan. That’s why, two years ago, I built a small tool to see how much of the oncologist’s job an LLM could do :)
https://github.com/jjanizek/blog_posts/blob/main/AutomatingOncology.md
Radiologist builds LLM tool to automate oncology scan triage
AIA resident radiologist built a small tool to test how much of an oncologist's work an LLM can do, citing waits of weeks for cancer patients to see oncology after an Emergency Room scan. The post links to a write-up on GitHub.
@joejanizek
As an oncologist, I often see a patient right after their CT scan, and the scan hasn't been read yet. Reads can take a day or two. So I built a small tool to see what an open model can do. You ask a question about a chest CT in plain words, and it shows you the answer and the exact spot it looked at. I typed "find the nodules in the lungs" and it found and boxed the nodule. It can also compare two scans, which is what I care about most. As an oncologist I want to know: is it growing, shrinking, or stable? It's a research demo on a 4 billion parameter open model. Not a medical device. I haven't tested sensitivity or specificity here, and I still want a radiologist's eyes on every scan. But for a quick look at new nodules or progression in clinic, this is promising. (The video is trimmed: the model took about 10 seconds to answer.)
Source details
Joseph D. Janizek · x.comPublished · added here
Original title: As a (resident) radiologist, I’ve seen that it can take weeks for patients to get into onc after I call a cancer on their Emergency Room ...