Prediction of Tuberculosis using U-Net and segmentation techniques

TL;DR

This paper presents an automated approach for tuberculosis diagnosis using U-Net segmentation of bacterial cords in lens-free microscopy images, aiming to improve ease of use and accuracy in resource-limited settings.

Contribution

It introduces a novel application of U-Net for segmenting TB bacteria in lens-free microscopy images to automate diagnosis, which is a new approach in this context.

Findings

U-Net achieved promising segmentation accuracy.

The method demonstrated good predictive performance for TB.

Lens-free microscopy simplifies the diagnostic process.

Abstract

One of the most serious public health problems in Peru and worldwide is Tuberculosis (TB), which is produced by a bacterium known as Mycobacterium tuberculosis. The purpose of this work is to facilitate and automate the diagnosis of tuberculosis using the MODS method and using lens-free microscopy, as it is easier to calibrate and easier to use by untrained personnel compared to lens microscopy. Therefore, we employed a U-Net network on our collected data set to perform automatic segmentation of cord shape bacterial accumulation and then predict tuberculosis. Our results show promising evidence for automatic segmentation of TB cords, and thus good accuracy for TB prediction.