AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

A case for reframing automated medical image classification as segmentation

Sarah Hooper, Mayee F Chen, Khaled Kamal Saab, Kush Bhatia, Curtis Langlotz, Christopher Re

NeurIPS 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use rule-based segmentation-for-classification instead of classification network, 10% labelssegmentation-for-classification · CANDIDmean AUROC: 10.0%Improved
use rule-based segmentation-for-classification instead of classification network, 10% labelssegmentation-for-classification · ISICmean AUROC: 6.9%Improved
use rule-based segmentation-for-classification instead of classification network, 10% labelssegmentation-for-classification · SPINEmean AUROC: 2.8%Improved
use rule-based segmentation-for-classification instead of classification network, all labelssegmentation-for-classification · SPINEmean AUROC: 6.1%Improved
use rule-based segmentation-for-classification instead of classification network, all labelssegmentation-for-classification · ISICmean AUROC: 0.6%Improved
segmentation instead of classification for small melanoma lesionssegmentation-for-classification · ISICrecall: 0.18 -> 0.47Improved
segmentation instead of classification when chest tube spurious correlation absentsegmentation-for-classification · CANDIDAUROC: 0.58 -> 0.84Improved
segmentation instead of classification across lung regions (worst-case region)segmentation-for-classification · CANDIDrecall: 0.71 -> 0.87Improved
use threshold, fully connected or classifier summarizing functions on segmentation output vs on embeddings (ablation)segmentation-for-classification · CANDID, ISIC, SPINEmean AUROCImproved
use multitask network (Y-Net) instead of segmentation-for-classification (ablation)multitask network · CANDID, ISIC, SPINEmean AUROCHurt
add semi-supervised segmentation with unlabeled data (ablation)segmentation-for-classification · CANDID, ISIC, SPINEmean AUROCImproved
add boosted semi-supervision using classification labels (ablation)segmentation-for-classification · CANDID, ISIC, SPINEmean AUROCImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy