TB Predictor TBXNet++

Tuberculosis (TB) is a major global public health concern, particularly in low- and middle-income countries. According to the World Health Organization (WHO), TB is one of the top 10 causes of death worldwide, and in 2020, an estimated 10 million people fell ill with TB, resulting in 1.5 million deaths. Addressing TB requires a comprehensive approach involving strong healthcare systems, increased awareness, improved diagnostics, access to quality treatment, and ongoing research and development for new tools and strategies. Deep learning techniques, including machine learning algorithms and neural networks, have been increasingly utilized in medical research and clinical applications, including anomaly detection and prediction. Inspired by deep learning, we developed TBXNet++ in order to detect and predic TB. Our model is trained on large datasets of chest X-ray images, along with corresponding clinical information, to detect patterns indicative of TB infection. Few steps are required to perform prediction using our developed model as shown on this page.

Intelligently understands & Automatically Recommends

Quick Search and Select disease

Select the desired indicator for performing prediction for instance Tuberculosis.

Select GPU

A GPU is automatically allocated to train your data before performing prediction

Upload Images

Upload your dataset containing the chest X-ray of the patient.

Perform Prediction

Perform prediction and view the results. The results will tell you whether a patient has TB or not. The visual representation and the confidence score of the results are also given.

Scientific Research Importance​
  • Our developed algorithm TBXNet has achieved an accuracy of 98.98%, which is comparatively better than all the state-of-the-art methods.
  • Grad-CAM technique has been utilized for Visual explanations by highlighting important regions in the image for predicting TB​