Detection and Prognosis of Pulmonary Tuberculosis Using ML Classifiers with Soft Computing

Several intriguing research works on soft computing techniques have recently been presented and for the creation of a pulmonary TB diagnosis system, researchers continue to build an accurate and trustworthy intelligent decision-making approach. The current diagnostic testing system processes are cumbersome and time-consuming to analyse. Since increased sensitivity and specificity can be attained, the diagnosis of TB still has to be improved upon using new speedy and accurate diagnostic models and procedures, which will help with disease prevention and control. Therefore, this review paper's significance lies in its ability to discriminate between the many soft computing methodologies used to help the diagnosis, identification, prediction, and intelligent categorization of pulmonary TB illness. Researchers and medical professionals anticipate employing soft computing methods in the field. “Artificial neural networks, genetic algorithms, support vector machines, fuzzy logic”, and others are a few of these. The most recent diagnostic techniques involve artificial neural networks. Predictive inference accuracy and easy-to-optimize, adoptable non-linear modelling of large data sets are some additional advantages of artificial neural networks, showing that they could be an important decision support tool in a variety of industries, including medicine.