Utilizing Deep Learning to Distinguish Between Non-Tuberculous Mycobacterial Lung Disease and Pulmonary Tuberculosis via Chest CT Imaging - Takeaways - MDSpire

Utilizing Deep Learning to Distinguish Between Non-Tuberculous Mycobacterial Lung Disease and Pulmonary Tuberculosis via Chest CT Imaging

  • By

  • Bingchuan Hu

  • Bin Wu

  • Yuwei Zhou

  • Zherui Shao

  • Qingning Wang

  • Binyu Luo

  • Zhuo Yu

  • Dawei Zheng

  • April 21, 2026

  • 0 min

Share

  • 1

    Differentiating non-tuberculous mycobacterial lung disease from pulmonary tuberculosis is clinically challenging due to overlapping features.

  • 2

    The study developed a 3D ResNeXt deep learning model that outperformed six other architectures in distinguishing NTM-LD from PTB.

  • 3

    The 3D ResNeXt model achieved an AUC of 0.83 and accuracy of 0.84 on the independent test set, indicating strong diagnostic performance.

  • 4

    Grad-CAM visualizations revealed specific imaging features for NTM-LD and PTB, aiding in the interpretability of the model's decisions.

  • 5

    The findings suggest that the 3D ResNeXt model could serve as a valuable clinical decision-support tool, pending further validation.

Original Source(s)

Utilizing Deep Learning to Distinguish Between Non-Tuberculous Mycobacterial Lung Disease and Pulmonary Tuberculosis via Chest CT Imaging

  • by Bingchuan Hu, Bin Wu, Yuwei Zhou, Zherui Shao, Qingning Wang, Binyu Luo, Zhuo Yu, Dawei Zheng

  • April 21, 2026

Related Content

Quantification of pulmonary involvement in COVID-19 pneumonia by means of a cascade of two U-nets: training and assessment on multiple datasets using different annotation criteria

CT-Based Assessment of Malignancy Risk in Part-Solid Pulmonary Nodules Utilizing Vascular Distortion and Interruption Analysis

Mandatory Training Modules Deserve a Harder Look

A JAMA Viewpoint argues that 4 million physician-hours spent on compliance modules each year warrants institutional reassessment.