Published September 25, 2024 | Version v2
Model Open

Pre-trained Swin UNETR, nnU-Net, and AIMOS Models for Mouse Organ Segmentation

  • 1. ROR icon University of California, San Francisco
  • 2. ROR icon University of California, Los Angeles
  • 3. ROR icon City of Hope

Description

This repository contains pre-trained Swin UNETR, nnU-Net, and AIMOS models for mouse organ segmentation. These models were trained using native micro-CT and contrast-enhanced micro-CT scans from a publicly available preclinical database. The models are capable of segmenting key abdominal structures, including the bladder, lungs, heart, liver, intestine, kidneys, and spleen.

The public data are openly available at https://www.nature.com/articles/sdata2018294.

The models are based on the Swin UNETR[1], nnU-Net[2], and AIMOS[3] frameworks, with modifications to adapt them to our specific needs. 

[1] Hatamizadeh A, Nath V, Tang Y, et al. Swin unetr: Swin transformers for semantic segmentation of brain tumors in mri images[C]//International MICCAI brainlesion workshop. Cham: Springer International Publishing, 2021: 272-284.

[2] Isensee F, Jaeger P F, Kohl S A A, et al. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation[J]. Nature methods, 2021, 18(2): 203-211.

[3] Schoppe O, Pan C, Coronel J, et al. Deep learning-enabled multi-organ segmentation in whole-body mouse scans[J]. Nature communications, 2020, 11(1): 5626.

Files

Mouse Organ Segmentation.zip

Files (1.3 GB)

Name Size Download all
md5:1a8a2bb867d5f21fbefca852264417ab
1.3 GB Preview Download

Additional details

Software

Programming language
Python