Microsoft Portrait of Hoifung Poon
It is great to see the new publication from Hoifung Poon and Microsoft Research about all-in-one biomedical image analysis, feature in this month’s Nature Methods Magazine.
I was assigned to photograph portraits of Hoifung Poon and capture him whilst speaking at a conference.
In cancer diagnosis or advanced treatments like immunotherapy, every detail in a medical image counts. Radiologists and pathologists rely on these images to track tumours, understand their boundaries, and analyse how they interact with surrounding cells. This work demands pinpoint accuracy across several tasks—identifying whether a tumour is present, locating it precisely, and mapping its contours on complex CT scans or pathology slides.
Object recognition, detection, and segmentation are often tackled separately, which can limit the depth of analysis.
BiomedParse is a new approach for holistic image analysis by treating object as the first-class citizen. By unifying object recognition, detection, and segmentation into a single framework, BiomedParse allows users to specify what they’re looking for through a simple, natural-language prompt. The result is a more cohesive, intelligent way of analysing medical images that supports faster, more integrated clinical insights.
You can read the full article at – www.microsoft.com/en-us/research/blog/biomedparse-a-foundation-model-for-smarter-all-in-one-biomedical-image-analysis
