4 papers · 1 filter
Few-Shot, Now for Real: Medical VLMs Adaptation without Balanced Sets or Validation
Julio Silva-Rodríguez, Fereshteh Shakeri, Houda Bahig +2
Vision-language models (VLMs) are gaining attention in medical image analysis. These are pre-trained on large, heterogeneous data sources, yielding rich and transferable representa…
Few-shot Adaptation of Medical Vision-Language Models
Fereshteh Shakeri, Yunshi Huang, Julio Silva-Rodríguez +4
Integrating image and text data through multi-modal learning has emerged as a new approach in medical imaging research, following its successful deployment in computer vision. Whil…
Boosting Vision-Language Models for Histopathology Classification: Predict all at once
Maxime Zanella, Fereshteh Shakeri, Yunshi Huang +2
The development of vision-language models (VLMs) for histo-pathology has shown promising new usages and zero-shot performances. However, current approaches, which decompose large s…
Beyond pixel-wise supervision for segmentation: A few global shape descriptors might be surprisingly good!
Hoel Kervadec, Houda Bahig, Laurent Letourneau-Guillon +2
Standard losses for training deep segmentation networks could be seen as individual classifications of pixels, instead of supervising the global shape of the predicted segmentation…