Pushkar1853/FaM-SAM

Repository for Fast Medical SAM (FaM-SAM).

FaM-SAM

Repository for Fast Medical SAM (FaM-SAM). Consists of Notebooks and datasets links for the results in the document.

Selected Datasets:

• https://www.kaggle.com/datasets/andrewmvd/covid19-ct-scans
• https://www.kaggle.com/datasets/tawsifurrahman/covid19-radiography-database
• https://www.kaggle.com/datasets/ashutoshnaik67/hovernetdataset
• https://warwick.ac.uk/fac/cross_fac/tia/data/glascontest/

Inference Models:

• Fast-SAM
• SAM
• Med-SAM

We create extrapolation from the Fast-SAM and Med-SAM performance results on these medical segmentation datasets and implement these to show contrast from the Vanilla Segment Anything Model. The conclusion led to the formation of Fast Medical SAM. (FaM-SAM)

Notebooks and codes:

• Notebooks/FastSAM - SAM/CT-Scan-Covid-19.ipynb
• Notebooks/FastSAM - SAM/Lung-Covid-19.ipynb
• Notebooks/MedSAM/Lung-Covid-19.ipynb
• Notebooks/MedSAM/breast-cancer.ipynb

Also, check https://github.com/Pushkar1853/HF-Medical-imaging for other scripts and notebooks used in the document.