# Automatic lung segmentation in routine imaging is primarily a data diversity problem, not a methodology problem

arXiv: https://arxiv.org/abs/2001.11767 · HTML: https://mlsanity.com/paper/2001.11767/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jan 2020.

## Built on by (most cited first, 9 of 9)

- [Mediastinal lymph nodes segmentation using 3D convolutional neural network ensembles and anatomical priors guiding](https://mlsanity.com/paper/2102.06515/) (2021)
- [Meningioma segmentation in T1-weighted MRI leveraging global context and attention mechanisms](https://mlsanity.com/paper/2101.07715/) (2021)
- [An Explainable AI System for Automated COVID-19 Assessment and Lesion Categorization from CT-scans](https://mlsanity.com/paper/2101.11943/) (2021)
- [SGDA: Towards 3D Universal Pulmonary Nodule Detection via Slice Grouped Domain Attention](https://mlsanity.com/paper/2303.03625/) (2023)
- [Quantification of pulmonary involvement in COVID-19 pneumonia by means of a cascade oftwo U-nets: training and assessment on multipledatasets using different annotation criteria](https://mlsanity.com/paper/2105.02566/) (2021)
- [Airway measurement by refinement of synthetic images improves mortality prediction in idiopathic pulmonary fibrosis](https://mlsanity.com/paper/2208.14141/) (2022)
- [Cutting-Splicing data augmentation: A novel technology for medical image segmentation](https://mlsanity.com/paper/2210.09099/) (2022)
- [Unsupervised Cross-domain Pulmonary Nodule Detection without Source Data](https://mlsanity.com/paper/2304.01085/) (2023)
- [Respiratory Differencing: Enhancing Pulmonary Thermal Ablation Evaluation Through Pre- and Intra-Operative Image Fusion](https://mlsanity.com/paper/2408.04299/) (2024)

## Abstract (arXiv)

Automated segmentation of anatomical structures is a crucial step in image analysis. For lung segmentation in computed tomography, a variety of approaches exist, involving sophisticated pipelines trained and validated on different datasets. However, the clinical applicability of these approaches across diseases remains limited. We compared four generic deep learning approaches trained on various datasets and two readily available lung segmentation algorithms. We performed evaluation on routine imaging data with more than six different disease patterns and three published data sets. Using different deep learning approaches, mean Dice similarity coefficients (DSCs) on test datasets varied not over 0.02. When trained on a diverse routine dataset (n = 36) a standard approach (U-net) yields a higher DSC (0.97 ± 0.05) compared to training on public datasets such as Lung Tissue Research Consortium (0.94 ± 0.13, p = 0.024) or Anatomy 3 (0.92 ± 0.15, p = 0.001). Trained on routine data (n = 231) covering multiple diseases, U-net compared to reference methods yields a DSC of 0.98 ± 0.03 versus 0.94 ± 0.12 (p = 0.024).