Beyond Classification: Elaborating Network Predictions for Better Weakly Supervised Quantization

Chih-Chieh Chen, Chang-Fu Kuo

Keywords: ICH classification, heatmap generations, weakly supervised localization.

Abstract:

For clinical applications, more detailed information such as specific locations and ROI
volumes is preferred. However, most of the time only classification annotations are avail-
able. Class Activation Mapping (CAM) and its variants are the most commonly used
techniques for weakly supervised localization tasks. In this study, we assessed both tradi-
tional and modern network architectures regarding classification accuracy and CAM visu-
alization. Although all networks achieved high AUROC scores and their heatmaps closely
corresponded to pathology locations, we observed that the heatmaps were influenced by the
particular network architectures and pretrained weights used. Additionally, current models
produce heatmaps from small latent spaces (e.g. 16 × 16), which limits the precision of
these heatmaps for further detailed analysis.
Based on the observations mentioned above, we designed a UNet-style architecture that
utilizes pretrained classification networks as the encoder and produces heatmaps within a
latent space of size 128 × 128. We observed that the generated heatmaps are more detailed
and suitable for weakly supervised segmentation. We validated the effectiveness of our
approach using the intracerebral hemorrhage (ICH) dataset.

Primary Subject Area: Learning with Noisy Labels and Limited Data

Secondary Subject Area: Interpretability and Explainable AI

Registration Requirement: Yes

Reproducibility: https://github.com/chihchiehchen/Beyond-Classification-Elaborating-Network-Predictions-for-Better-Weakly-Supervised-Quantization

Visa & Travel: No

Read CFP & Author Instructions: Yes

Originality Policy: Yes

Single-blind & Not Under Review Elsewhere: Yes

LLM Policy: Yes

Submission Number: 3