# Progressive Curriculum Learning with Scale-Enhanced U-Net for Continuous Airway Segmentation

arXiv: https://arxiv.org/abs/2410.18456 · HTML: https://mlsanity.com/paper/2410.18456/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 58 references that are in the map.
- First posted on arXiv 24 Oct 2024.

## Builds on (most cited first, 19 of 19)

- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation](https://mlsanity.com/paper/1606.04797/) (2016)
- [3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation](https://mlsanity.com/paper/1606.06650/) (2016)
- [Alleviating Class-wise Gradient Imbalance for Pulmonary Airway Segmentation](https://mlsanity.com/paper/2011.11952/) (2020)
- [AirwayNet: A Voxel-Connectivity Aware Approach for Accurate Airway Segmentation Using Convolutional Neural Networks](https://mlsanity.com/paper/1907.06852/) (2019)
- [Learning Tubule-Sensitive CNNs for Pulmonary Airway and Artery-Vein Segmentation in CT](https://mlsanity.com/paper/2012.05767/) (2020)
- [Automatic Airway Segmentation in chest CT using Convolutional Neural Networks](https://mlsanity.com/paper/1808.04576/) (2018)
- [Fuzzy Attention Neural Network to Tackle Discontinuity in Airway Segmentation](https://mlsanity.com/paper/2209.02048/) (2022)
- [BREAK: Bronchi Reconstruction by gEodesic transformation And sKeleton embedding](https://mlsanity.com/paper/2202.00002/) (2022)
- [BronchoTrack: Airway Lumen Tracking for Branch-Level Bronchoscopic Localization](https://mlsanity.com/paper/2402.12763/) (2024)
- [Unsupervised Medical Image Alignment with Curriculum Learning](https://mlsanity.com/paper/2102.10438/) (2021)
- [Two-stage Contextual Transformer-based Convolutional Neural Network for Airway Extraction from CT Images](https://mlsanity.com/paper/2212.07651/) (2022)
- [Curriculum-Based Augmented Fourier Domain Adaptation for Robust Medical Image Segmentation](https://mlsanity.com/paper/2306.03511/) (2023)
- [Accurate Airway Tree Segmentation in CT Scans via Anatomy-aware Multi-class Segmentation and Topology-guided Iterative Learning](https://mlsanity.com/paper/2306.09116/) (2023)
- [Style Curriculum Learning for Robust Medical Image Segmentation](https://mlsanity.com/paper/2108.00402/) (2021)
- [LTSP: Long-Term Slice Propagation for Accurate Airway Segmentation](https://mlsanity.com/paper/2202.06260/) (2022)
- [PANS: Probabilistic Airway Navigation System for Real-time Robust Bronchoscope Localization](https://mlsanity.com/paper/2407.05554/) (2024)
- [A New Three-stage Curriculum Learning Approach to Deep Network Based Liver Tumor Segmentation](https://mlsanity.com/paper/1910.07895/) (2019)
- [Probing Perfection: The Relentless Art of Meddling for Pulmonary Airway Segmentation from HRCT via a Human-AI Collaboration Based Active Learning Method](https://mlsanity.com/paper/2407.03542/) (2024)

## Built on by (most cited first, 2 of 2)

- [BREATH-VL: Vision-Language-Guided 6-DoF Bronchoscopy Localization via Semantic-Geometric Fusion](https://mlsanity.com/paper/2601.03713/) (2026)
- [CalcSeg: Confidence-aware 3D Latent Context Curriculum Learning For Myocardial Scar Segmentation From Single-Stack LGE-CMRs](https://mlsanity.com/paper/2608.20305/) (2026)

## Abstract (arXiv)

Continuous and accurate segmentation of airways in chest CT images is essential for preoperative planning and real-time bronchoscopy navigation. Despite advances in deep learning for medical image segmentation, maintaining airway continuity remains a challenge, particularly due to intra-class imbalance between large and small branches and blurred CT scan details. To address these challenges, we propose a progressive curriculum learning pipeline and a Scale-Enhanced U-Net (SE-UNet) to enhance segmentation continuity. Specifically, our progressive curriculum learning pipeline consists of three stages: extracting main airways, identifying small airways, and repairing discontinuities. The cropping sampling strategy in each stage reduces feature interference between airways of different scales, effectively addressing the challenge of intra-class imbalance. In the third training stage, we present an Adaptive Topology-Responsive Loss (ATRL) to guide the network to focus on airway continuity. The progressive training pipeline shares the same SE-UNet, integrating multi-scale inputs and Detail Information Enhancers (DIEs) to enhance information flow and effectively capture the intricate details of small airways. Additionally, we propose a robust airway tree parsing method and hierarchical evaluation metrics to provide more clinically relevant and precise analysis. Experiments on both in-house and public datasets demonstrate that our method outperforms existing approaches, significantly improving the accuracy of small airways and the completeness of the airway tree. The code will be released upon publication.