# Weakly Supervised Medical Diagnosis and Localization from Multiple Resolutions

arXiv: https://arxiv.org/abs/1803.07703 · HTML: https://mlsanity.com/paper/1803.07703/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Mar 2018.

## Built on by (most cited first, 22 of 22)

- [Delving into Masked Autoencoders for Multi-Label Thorax Disease Classification](https://mlsanity.com/paper/2210.12843/) (2022)
- [Learning Hierarchical Attention for Weakly-supervised Chest X-Ray Abnormality Localization and Diagnosis](https://mlsanity.com/paper/2112.12349/) (2021)
- [Label-Assemble: Leveraging Multiple Datasets with Partial Labels](https://mlsanity.com/paper/2109.12265/) (2021)
- [SwinCheX: Multi-label classification on chest X-ray images with transformers](https://mlsanity.com/paper/2206.04246/) (2022)
- [Probabilistic Integration of Object Level Annotations in Chest X-ray Classification](https://mlsanity.com/paper/2210.06980/) (2022)
- [I-AI: A Controllable & Interpretable AI System for Decoding Radiologists' Intense Focus for Accurate CXR Diagnoses](https://mlsanity.com/paper/2309.13550/) (2023)
- [Learning Inductive Attention Guidance for Partially Supervised Pancreatic Ductal Adenocarcinoma Prediction](https://mlsanity.com/paper/2105.14773/) (2021)
- [Weakly-supervised High-resolution Segmentation of Mammography Images for Breast Cancer Diagnosis](https://mlsanity.com/paper/2106.07049/) (2021)
- [Robust Classification from Noisy Labels: Integrating Additional Knowledge for Chest Radiography Abnormality Assessment](https://mlsanity.com/paper/2104.05261/) (2021)
- [Beyond Fine-tuning: Classifying High Resolution Mammograms using Function-Preserving Transformations](https://mlsanity.com/paper/2101.07945/) (2021)
- [ThoraX-PriorNet: A Novel Attention-Based Architecture Using Anatomical Prior Probability Maps for Thoracic Disease Classification](https://mlsanity.com/paper/2210.02998/) (2022)
- [Reciprocal Landmark Detection and Tracking with Extremely Few Annotations](https://mlsanity.com/paper/2101.11224/) (2021)
- [Learning Low-Rank Feature for Thorax Disease Classification](https://mlsanity.com/paper/2404.18933/) (2024)
- [Deep Reinforcement Learning Framework for Thoracic Diseases Classification via Prior Knowledge Guidance](https://mlsanity.com/paper/2306.01232/) (2023)
- [Computer-Aided Diagnosis of Thoracic Diseases in Chest X-rays using hybrid CNN-Transformer Architecture](https://mlsanity.com/paper/2404.11843/) (2024)
- [Searching for Pneumothorax in X-Ray Images Using Autoencoded Deep Features](https://mlsanity.com/paper/2102.06096/) (2021)
- [Many-to-One Distribution Learning and K-Nearest Neighbor Smoothing for Thoracic Disease Identification](https://mlsanity.com/paper/2102.13269/) (2021)
- [DeepChest: Dynamic Gradient-Free Task Weighting for Effective Multi-Task Learning in Chest X-ray Classification](https://mlsanity.com/paper/2505.23595/) (2025)
- [CLARiTy: A Vision Transformer for Multi-Label Classification and Weakly-Supervised Localization of Chest X-ray Pathologies](https://mlsanity.com/paper/2512.16700/) (2025)
- [Weighing Features of Lung and Heart Regions for Thoracic Disease Classification](https://mlsanity.com/paper/2105.12430/) (2021)
- [BRAIxDet: Learning to Detect Malignant Breast Lesion with Incomplete Annotations](https://mlsanity.com/paper/2301.13418/) (2023)
- [StructCore: Structure-Aware Image-Level Scoring for Training-Free Unsupervised Anomaly Detection](https://mlsanity.com/paper/2602.17048/) (2026)

## Abstract (arXiv)

Diagnostic imaging often requires the simultaneous identification of a multitude of findings of varied size and appearance. Beyond global indication of said findings, the prediction and display of localization information improves trust in and understanding of results when augmenting clinical workflow. Medical training data rarely includes more than global image-level labels as segmentations are time-consuming and expensive to collect. We introduce an approach to managing these practical constraints by applying a novel architecture which learns at multiple resolutions while generating saliency maps with weak supervision. Further, we parameterize the Log-Sum-Exp pooling function with a learnable lower-bounded adaptation (LSE-LBA) to build in a sharpness prior and better handle localizing abnormalities of different sizes using only image-level labels. Applying this approach to interpreting chest x-rays, we set the state of the art on 9 abnormalities in the NIH's CXR14 dataset while generating saliency maps with the highest resolution to date.