Predictive and discriminative localization of pathology using high resolution class activation maps with CNNs

Abstract

Purpose Existing class activation mapping (CAM) techniques extract the feature maps only from a single layer of the convolutional neural net (CNN), generally from the final layer and then interpolate to upsample to the original image resolution to locate the discriminative regions. Consequently these provide a coarse localization that may not be able to capture subtle abnormalities in medical images. To alleviate this, our work proposes a technique called high resolution class activation mapping (HR-CAMs) that can provide enhanced visual explainability to the CNN models. Methods HR-CAMs fuse feature maps by training a network using the input from multiple layers of a trained CNN, thus gaining information from every layer that can localize abnormalities with greater details in original image resolution. The technique is validated qualitatively and quantitatively on a simulated dataset of 8,000 images followed by applications on multiple image analysis tasks that include (1) skin lesion classification (ISIC open dataset—25,331 cases) and (2) predicting bone fractures (MURA open dataset—40,561 images) (3) predicting Parkinson’s disease (PD) from neuromelanin sensitive MRI (small cohort-80 subjects). Results We demonstrate that our model creates clinically interpretable subject specific high resolution discriminative localizations when compared to widely used CAMs and Gradient-CAMs. Conclusion HR-CAMs provide finer delineation of abnormalities thus facilitating superior explainability to CNNs as has been demonstrated from its rigorous validation.

A multicenter proof-of-concept study on deep learning-based intraoperative discrimination of primary central nervous system lymphoma

“…Fourthly, this is only a proof-of-concept study that needs for future analyses on a suitable number of pathologists and on better-selected cases (small biopsies and difficult neuroimaging interpretation) to draw more reliable conclusions. Lastly, while the logistic regression model provides biological interpretability for the deep learning model, more advanced visual methods, such as high-resolution class activation mapping 34 , could offer further insights into the model functioning.…”

Deep learning-based intraoperative differentiation of primary CNS lymphoma and glioma: a discovery, multicenter validation, and proof-of concept study

“…Thirdly, the online platform still needs to be re ned before it can be deployed for real clinical use. Lastly, while the logistic regression model provides biological interpretability for the deep learning model, more advanced visual methods, such as high-resolution class activation mapping 25 , may be used to further understand the model.…”

Deep Learning in the Identification of Electroencephalogram Sources Associated with Sexual Orientation

“…For feature extraction, the class activation maps allow to seek for distinctive features in the input data for deep learning based networks. The class activation maps show weighted heatmaps of the input data that allow for the identification of the most important data points for classification tasks, e.g., in medical imaging [34]. Their successor, Grad-CAM allows for the same approach without the necessity of global average pooling before the softmax layer by using gradient information that flows into the last convolutional layer of the network.…”