Figure 3 from GPSai: A Clinically Validated AI Tool for Tissue of Origin Prediction during Routine Tumor Profiling

Hassan Ghani; Anthony Helmstetter; Jennifer R. Ribeiro; Todd Maney; Stephanie Rock; Rebecca Feldman; Jeffrey Swensen; Farah Abdulla; David Spetzler; Elena Florento; Ari M. Vanderwalde; Patricia Pittman; Milan Radovich; Jaclyn F. Hechtman; Casey Bales; George W. Sledge; Myra M. George; David Bryant; Jim Abraham; Matthew J. Oberley

doi:10.1158/2767-9764.30022013

Key Points

GPSai achieved notable accuracy in predicting tumor types, aligning closely with TCGA labels.
The analysis included 6,820 samples from TCGA, with a confusion matrix illustrating prediction concordance.
Validation utilized the confusion matrix to analyze the performance across various cancer categories.
These findings support the potential clinical use of AI in enhancing tumor profiling accuracy.

Implication

Assessment of GPSai demonstrates its effectiveness in predicting tumor types from the TCGA dataset, suggesting reliable tissue classification.

Abstract

Validation of GPSai in the TCGA dataset. The confusion matrix shows GPSai concordance for each cancer category among 6,820 TCGA samples. Numbers of each predicted tumor type are shown in the boxes, with numbers of correctly predicted tumors (matching TCGA label) in each category shown in the diagonal. CNS, central nervous system.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Ghani et al. (2025) studied this question.

synapsesocial.com/papers/68bb49cc6d6d5674bccffab7 https://doi.org/10.1158/2767-9764.30022013

View Full Paper

Ask AI

Bookmark

Share

Figure 3 from GPSai: A Clinically Validated AI Tool for Tissue of Origin Prediction during Routine Tumor Profiling

Key Points

Implication

Abstract

Discussion

Cite This Study

Figure 3 from GPSai: A Clinically Validated AI Tool for Tissue of Origin Prediction during Routine Tumor Profiling

Key Points

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading