Deep learning for diagnosis and survival prediction in soft tissue sarcoma
S Foersch1, M Eckstein2, D-C Wagner1
1Institute of Pathology, University Medical Center Mainz, Mainz, Germany.
Summary
Deep learning (DL) models can accurately diagnose soft tissue sarcoma (STS) subtypes and predict leiomyosarcoma (LMS) survival. This technology aids pathologists, improving diagnostic speed and accuracy for better STS patient management.
Area of Science:
- Digital pathology
- Artificial intelligence in oncology
- Machine learning for cancer diagnosis
Background:
- Soft tissue sarcoma (STS) presents significant clinical management challenges.
- Digital pathology and deep learning (DL) offer novel approaches for STS diagnosis and prognosis.
Purpose of the Study:
- To evaluate the efficacy of a DL model (DLM) in diagnosing STS subtypes.
- To assess the DLM's capability in predicting prognosis for leiomyosarcoma (LMS).
- To determine the DLM's impact on pathologists' diagnostic performance.
Main Methods:
- Retrospective, multicenter study with 506 histopathological slides from 291 STS patients.
- Training and validation using The Cancer Genome Atlas cohort; testing on an independent multicenter cohort.
- Evaluation of DLM by nine pathologists; prognosis prediction for LMS using 139 slides from 85 patients.
Main Results:
- DLM achieved high accuracy (79.9%) and AUROC (0.97) in diagnosing common STS subtypes.
- DLM significantly improved pathologist accuracy from 46.3% to 87.1%, increasing speed and certainty.
- DLM demonstrated strong prognostic prediction for LMS (AUROC 0.91, accuracy 88.9%) and was an independent prognostic factor.
Conclusions:
- DL accurately diagnoses frequent STS subtypes from histopathological slides.
- DL shows potential for prognosis prediction in LMS, improving clinical management.
- DL enhances pathologist diagnostic accuracy and efficiency, benefiting STS patient care.
Related Concept Videos
01:21Cancer Survival Analysis
490
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...
490
02:07Tumor Progression
6.7K
Tumor progression is a phenomenon where the pre-formed tumor acquires successive mutations to become clinically more aggressive and malignant. In the 1950s, Foulds first described the stepwise progression of cancer cells through successive stages.
Colon cancer is one of the best-documented examples of tumor progression. Early mutation in the APC gene in colon cells causes a small growth on the colon wall called a polyp. With time, this polyp grows into a benign, pre-cancerous tumor. Further...
Colon cancer is one of the best-documented examples of tumor progression. Early mutation in the APC gene in colon cells causes a small growth on the colon wall called a polyp. With time, this polyp grows into a benign, pre-cancerous tumor. Further...
6.7K