Re: how many sputum specimens are necessary to diagnose pulmonary tuberculosis

Distinguishing infectivity in patients with pulmonary tuberculosis using deep learning

Introduction

This study aimed to develop and assess a deep-learning model based on CT images for distinguishing infectivity in patients with pulmonary tuberculosis (PTB).

Methods

We labeled all 925 patients from four centers with weak and strong infectivity based on multiple sputum smears within a month for our deep-learning model named TBINet's training. We compared TBINet's performance in identifying infectious patients to that of the conventional 3D ResNet model. For model explainability, we used gradient-weighted class activation mapping (Grad-CAM) technology to identify the site of lesion activation in the CT images.

Results

The TBINet model demonstrated superior performance with an area under the curve (AUC) of 0.819 and 0.753 on the validation and external test sets, respectively, compared to existing deep learning methods. Furthermore, using Grad-CAM, we observed that CT images with higher levels of consolidation, voids, upper lobe involvement, and enlarged lymph nodes were more likely to come from patients with highly infectious forms of PTB.

Conclusion

Our study proves the feasibility of using CT images to identify the infectivity of PTB patients based on the deep learning method.

How Can the Tuberculosis Laboratory Aid in the Patient-Centered Diagnosis and Management of Tuberculosis?

In 2019, tuberculosis is still a global source of morbidity and mortality. To determine and provide the most effective treatment regimen to patients, the tuberculosis laboratory needs to rapidly but reliably answer 2 main questions: (1) Is Mycobacterium tuberculosis detectable in the patient specimen? and (2) If so, is the strain detected drug susceptible or does it show any form of drug resistance? In cases of drug resistance, health care providers need to have access to minimal inhibitory concentration results and to the type of mutation conferring drug resistance to tailor the most appropriate drug regimen.