# Deep Learning from Label Proportions for Emphysema Quantification

arXiv: https://arxiv.org/abs/1807.08601 · HTML: https://mlsanity.com/paper/1807.08601/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Jul 2018.

## Built on by (most cited first, 13 of 13)

- [Learning from Label Proportions by Learning with Label Noise](https://mlsanity.com/paper/2203.02496/) (2022)
- [Easy Learning from Label Proportions](https://mlsanity.com/paper/2302.03115/) (2023)
- [A Unified Approach to Count-Based Weakly-Supervised Learning](https://mlsanity.com/paper/2311.13718/) (2023)
- [Fast learning from label proportions with small bags](https://mlsanity.com/paper/2110.03426/) (2021)
- [LLP-Bench: A Large Scale Tabular Benchmark for Learning from Label Proportions](https://mlsanity.com/paper/2310.10096/) (2023)
- [Nearly Optimal Sample Complexity for Learning with Label Proportions](https://mlsanity.com/paper/2505.05355/) (2025)
- [Proportion Estimation by Masked Learning from Label Proportion](https://mlsanity.com/paper/2405.04815/) (2024)
- [Training image classifiers using Semi-Weak Label Data](https://mlsanity.com/paper/2103.10608/) (2021)
- [Learning from Label Proportions with Instance-wise Consistency](https://mlsanity.com/paper/2203.12836/) (2022)
- [Forming Auxiliary High-confident Instance-level Loss to Promote Learning from Label Proportions](https://mlsanity.com/paper/2411.10364/) (2024)
- [Hardness of Learning Boolean Functions from Label Proportions](https://mlsanity.com/paper/2403.19401/) (2024)
- [FRACTAL: Fine-Grained Scoring from Aggregate Text Labels](https://mlsanity.com/paper/2404.04817/) (2024)
- [Learning from Label Proportions and Covariate-shifted Instances](https://mlsanity.com/paper/2411.12334/) (2024)

## Abstract (arXiv)

We propose an end-to-end deep learning method that learns to estimate emphysema extent from proportions of the diseased tissue. These proportions were visually estimated by experts using a standard grading system, in which grades correspond to intervals (label example: 1-5% of diseased tissue). The proposed architecture encodes the knowledge that the labels represent a volumetric proportion. A custom loss is designed to learn with intervals. Thus, during training, our network learns to segment the diseased tissue such that its proportions fit the ground truth intervals. Our architecture and loss combined improve the performance substantially (8% ICC) compared to a more conventional regression network. We outperform traditional lung densitometry and two recently published methods for emphysema quantification by a large margin (at least 7% AUC and 15% ICC), and achieve near-human-level performance. Moreover, our method generates emphysema segmentations that predict the spatial distribution of emphysema at human level.