Online Tool for Anatomic Detection in Pituitary Surgery

We’re happy to announce the development of a website where researchers and clinicians can upload their videos of pituitary surgery and obtain detections of various anatomical structures and surgical instruments.

Try it here:

https://surgicalvision.bmic.ethz.ch/upload

Our tool utilizes the YOLOv7 network, which has been trained and validated on 166 videos of transsphenoidal adenomectomy, encompassing approximately 19,000 labeled frames.

To further support research in computer vision in neurosurgery, we also provide access to the model's weights. For more information, please refer to this paper:

external page Live Image-Based Neurosurgical Guidance and Roadmap Generation Using Unsupervised Embedding

 

 

JavaScript has been disabled in your browser