Deep Learning-Based Tumor Segmentation of Murine Magnetic Resonance Images of Prostate Cancer Patient-Derived Xenografts
The slice classifier network achieved 90% accuracy in identifying slices containing tumors. Among the various segmentation architectures tested, the dense residual recurrent U-Net achieved the highest performance in kidney tumors. When evaluated across the kidneys, tibia, and liver, this architecture performed the best when trained on all data as compared to training on only data from a single site (and inferring on a multi-site tumor images), achieving a Dice score of 0.924 across the test set.
Authors
- Renuka Sriram (ORCID: https://orcid.org/0000-0003-3505-2479)
- Nikhil Deveshwar (ORCID: https://orcid.org/0000-0002-5516-3769)
- Avantika Sinha (ORCID: https://orcid.org/0000-0002-9349-6345)
- Madeline Hess (ORCID: https://orcid.org/0000-0002-3313-932X)
- Peder E. Z. Larson (ORCID: https://orcid.org/0000-0003-4183-3634)
- Sule Sahin (ORCID: https://orcid.org/0000-0002-7006-4793)
- Abhejit Rajagopal (ORCID: https://orcid.org/0000-0003-0402-4643)
- Matthew Gibbons (ORCID: https://orcid.org/0000-0003-4108-6157)
- Sunita Nayak
- Ernesto Diaz (ORCID: https://orcid.org/0000-0002-3716-7758)
- Henry Salkever
Institutions
- University of California, San Francisco (US)
Publication Details
- Journal
- Tomography
- Published
- 2025-02-22
- DOI
- https://doi.org/10.3390/tomography11030021
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00