Extending Grad-CAM to DualNet.
Authors
Schulz A,Krettek C,Wolff D
Affiliations (2)
Affiliations (2)
- Peter L. Reichertz Institute for Medical Informatics of TU Braunschweig and Hannover Medical School, Hannover Medical School, Hannover, Germany.
- Professor em., Trauma Department, Hannover Medical School, Hannover, Germany.
Abstract
The machine learning architecture DualNet was introduced for analysing both frontal and lateral X-rays of a patient for disease classification. Since the model lacks explainability methods, we extended the pytorch-grad-cam package for the application on DualNet. Here, we show in a proof-of-principle that our implementation technically works in an artificial scenario.
Topics
Machine LearningJournal Article