Implementation of a low-cost automatic baciloscopy system for the diagnosis of tuberculosis

According to the World Health Organization (WHO), tuberculosis continues to be a global problem of great importance. In Peru, around 31 120 cases of TB were registered annually, with an incidence of 116 new cases per 100 000 inhabitants. Peru still occupies first place in Latin America in the number of resistant tuberculosis cases. Overcrowding coupled with the lack of rapid diagnostic equipment accessible to the neediest populations has resulted in the disease remaining a problem. The diagnostic process is manual, leading to great variability in the results because it depends on the laboratory staff's experience and concentration. Although not the most sensitive, conventional baciloscopy techniques are the most widely used due to their speed and cost. These techniques involve a staining stage and a microscopy stage. Ziehl Neelsen (ZN) staining method is used in smear microscopy because it is effective and simple. This article shows a prototype that performs tuberculosis diagnosis for 4 samples, consisting of two automatical modules: the staining module and the microscope module. The staining module consists of a sample positioning sub-system, a reagent dispensing sub-system, and a heating sub-system. The microscope module captures fields' images of up to 4 samples stained using the Ziehl-Neelsen method and determines the diagnosis of Tuberculosis (TB) through image processing algorithms. The mechanical system of this module is made up of three subsystems: the main support that houses the electronic circuits, a mobile platform that moves the samples, and the microscope holding structure. Tests have been performed on 206 samples at the Dos de Mayo National Hospital in Peru. These tests involved several adjustments that allowed reducing the staining times and the quantities of reagents used.