On-Field Test of Tuberculosis Diagnosis through Exhaled Breath Analysis with a Gas Sensor Array

Area of Science:

• Pulmonary Medicine
• Biotechnology
• Diagnostic Technology

Background:

• Tuberculosis (TB) remains a significant global health threat, particularly in low-middle-income countries.
• Current diagnostic methods like smear microscopy have limitations in sensitivity and widespread accessibility.
• There is a critical need for improved, low-cost diagnostic tools for early pulmonary TB detection.

Purpose of the Study:

• To evaluate the diagnostic performance of an electronic nose (EN) for identifying pulmonary TB.
• To assess the feasibility of using EN technology for TB diagnosis in a real-world clinical setting.
• To explore the potential of exhaled volatile organic compounds (VOCs) as biomarkers for TB.

Main Methods:

• An electronic nose (EN) device utilizing sensor technology was deployed in a hospital in Cameroon.
• Breath samples were collected from pulmonary TB patients (46), healthy controls (38), and TB suspects (16).
• Machine learning algorithms were applied to analyze the sensor array data for pattern recognition.

Main Results:

• The EN achieved 88% accuracy, 90.8% sensitivity, and 85.7% specificity in distinguishing pulmonary TB patients from healthy controls.
• A high Area Under the Curve (AUC) of 0.88 was recorded, indicating strong diagnostic capability.
• The trained model demonstrated consistent performance when applied to TB suspects with negative TB-LAMP results.

Conclusions:

• Electronic nose technology shows significant potential as an effective and accessible diagnostic tool for pulmonary tuberculosis.
• The analysis of exhaled volatile organic compounds (VOCs) offers a promising non-invasive approach for TB detection.
• Further investigation and clinical integration of ENs could enhance global TB control strategies, especially in resource-limited areas.