Non-invasive detection of mycobacterium tuberculosis using IR and NIR spectroscopy

Tuberculosis (TB) is a tropical disease affecting millions of lives. Conventional chemical based methods such as acid-fast technique for the detection of TB are time consuming and suffer from low sensitivity and specificity. We present a non-invasive detection technique for TB using IR and NIR spectroscopy. Spectral fingerprints of Mycobacterium are compared with E-Coli, Smegmatis and BCG bacteria. Principal component analysis on the spectral data shows that the bacteria can be distinguished easily due to their slightly different composition of cellular constituents. Non-invasive method presented here provides an attractive pathway for low cost, rapid portable diagnostic systems for TB and other diseases.