Blog
Medical Imaging Modalities
An overview of medical imaging modalities
The challenge of medical image analysis
What's unique about medical imaging
Is LLM the Cure for Medical Image Analysis?
If it's the cure, what is the problem?
Zhang Xiangyu's Talk on Multimodality & Reasoning
A truly inspirational view on CV -> Multimodality -> Reasoning
The two types of uncertainty
Can we solve diagnosis with infinite data and model?
What does cancer look like?
An intuition about what cancer looks like in medical imaging.
Leveraging Prior Knowledge
the not-so-secret technique to juice more nutrients out of data
Thoughts on data annotation
frustrations and challenges, tears and sweat
Anatomical Positional Embedding
Exploring anatomical positional embedding for medical imaging applications - a novel approach to encoding spatial and anatomical context for improved medical image analysis.
GPT Example: A Rare Lesion in CT
GPT-O3 demonstrates impressive diagnostic capabilities for rare mediastinal lesions.
O3 Medical Image Analysis Example
It might be a toy, but it is a futuristic toy.
O3 Can Be a Great Radiologist, If You Let It
GPT-O3 demonstrates impressive diagnostic reasoning when given the proper tools and context.
GEMeX-ThinkVG
Using RL to get the bbox right.
CT-to-Report Does NOT Work, Yet
It could, just not with current methodologies.
Green Score
An okay metric for radiology report generation.
Reason Like a Radiologist
Structured thinking first, writing the report last.
Sequential Diagnosis with Language Models
An interesting study on longitudinal diagnosis