Predicting microsatellite instability from whole slide images using texture features
Swanson, N.; Castro, M. A. A.; Robertson, A. G.; Shmulevich, I.; Tercan, B.
Show abstract
Identifying MSI in whole slide images (WSIs), one of the most widely used diagnostic imaging formats, is of great importance and in demand. In this study we employed color-based texture features to predict MSI on both a tile and sample based level. We found that within cohorts of hematoxylin and eosin (H&E) stained WSIs, texture morphology is able to predict MSI on a tile level with an AUC of up to 0.95 and on a sample level with an AUC of up to 0.98. This runs in contrast to other methods for predicting MSI in H&E WSIs which either utilized artificial intelligence based models, or achieved lower accuracy scores. Our results demonstrate that texture morphology is a significantly notable factor when it comes to identifying MSI in H&E WSIs, and should be used when constructing future models for MSI identification in a clinical setting.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Weakly supervised learning for multi-organ adenocarcinoma classification in whole slide images 96%
- Automated Clear Cell Renal Carcinoma Grade Classification with Prognostic Significance 94%
- Pixelwise H-score: a novel digital image analysis based-metric to quantify membrane biomarker expression from immunohistochemistry images 94%
Similar papers in this journal
- Using an Anomaly Detection Approach for the Segmentation of Colorectal Cancer Tumors in Whole Slide Images 97%
- Bladder Cancer Prognosis Using Deep Neural Networks and Histopathology Images 95%
- Development of an Interactive Web Dashboard to Facilitate the Reexamination of Pathology Reports for Instances of Underbilling of CPT Codes 93%
Similar papers in this journal
- Comparative Analysis of Pathology Foundation Models for Automated Detection of Tertiary Lymphoid Structures in H&E-Stained Digital Pathology Images 93%
- Topological embedding and directional feature importance in ensemble classifiers for multi-class classification 92%
- Wide and Deep Learning for Automatic Cell Type Identification 92%
Similar papers in this journal
- Cell graph neural networks enable the digital staging of tumor microenvironment and precise prediction of patient survival in gastric cancer 93%
- Cancer drug sensitivity prediction from routine histology images 92%
- A Deep Learning Model for Molecular Label Transfer that Enables Cancer Cell Identification from Histopathology Images 92%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.