Cough count as a marker for patient recovery from pulmonary tuberculosis

In regions of the world where tuberculosis (TB) poses the greatest disease burden, clinicians often lack access to skilled laboratories. This is particularly problematic for patients with drug-resistant tuberculosis, as these patients will otherwise receive standard TB medication and will not respond to treatment. Thus, a lab-free method for assessing patient recovery during treatment would be of great benefit. We hypothesize that cough analysis may provide such a test, and have carried out a pilot study to record coughs from a cohort of patients in Lima, Peru. We describe algorithm development for cough data analysis and compare several event detection and classification strategies. Results from our ongoing validation efforts suggest that cough count (cough/hour) decreases noticeably after the start of treatment in drug-responsive patients. Our long-term goal is development of a low-cost ambulatory cough analysis system that will help identify patients with drug-resistant tuberculosis.