Delft Imaging Systems' innovative CAD4TB software is designed to help (non) expert readers in the detection of TB. Within one minute from exposure, the software automatically analyses chest X-rays, detects abnormalities and indicates the likelihood of active TB.
• Servicing in remote areas • No skilled personnel required • Increased accuracy of TB case findings • Low-cost per screening (€1.50) • Higher sensitivity than sputum-smear microscopy test • Quick diagnosis (within 1 minute) • Up to 300 examinations daily • Diagnosis and treatment in the very same spot • No additional running costs
With annually 9 million new cases and 1.5 million deaths worldwide, TB is the second cause of death after HIV/AIDS.
In countries with a high TB burden, sputum smear microscopy is the most common means for diagnosis. Provided the relatively high workload and poor sensitivity of this approach, digital chest radiography is a good alternative with large potential in TB case findings. Also, digital chest radiography uses no film, chemicals or water permitting the reduction of costs and increasing usability of the technique for low-resource areas. Complementary, Delft Imaging Systems CAD4TB, Computer Aided Detection for TB, is a software package that automatically analyses chest X-rays and is capable of detecting abnormalities that may likely indicate TB. CAD4TB detection software classifies normal versus abnormal CXRs and marks suspect regions, combines shape-texture-symmetry abnormality scores and calculates an abnormality probability for TB. The software will automatically calculate and display the abnormality score on a scale from 0-100, which is a valuable detection tool to mitigate the risk of under- and over reading. As such, CAD4TB, supports (non-) experts with the interpretation of chest X-rays, enabling reliable screening of populations in underserved areas.
CAD4TB has been developed by Delft Imaging Systems in cooperation with the Radboud University of Nijmegen and the Lung Institute in Cape Town. The involvement of academic researchers from the onset demonstrates the efforts of delivering a quality and reliable product. For the academic publications on CAD4TB please click on the following links:
"Automated chest-radiography as a triage for Xpert testing in resource-constrained settings: a prospective study of diagnostic accuracy and costs", Scientific Reports 5:12215 DOI: 10.1038/srep12215 Abstract/PDF
"Diagnostic accuracy of computer-aided detection of pulmonary tuberculosis in chest radiographs: a validation study from sub-saharan Africa", PLoS One 2014;9:e106381 Abstract/PDF
"Automated Scoring of Chest Radiographs for Tuberculosis Prevalence Surveys: A Combined Approach", Abstract/PDF
"Detection of Tuberculosis with Digital Chest Radiography: Automatic Reading versus Interpretation by Clinical Officers", International Journal of Tuberculosis and Lung Disease 2013:17: 1613-1620 Abstract/PDF
"The Sensitivity and Specificity of Using a Computer Aided Diagnosis Program for Automatically Scoring Chest X-Rays of Presumptive TB Patients Compared with Xpert MTB/RIF in Lusaka Zambia", Plos One 2015:e93757 Abstract/PDF
"Advances in Tuberculosis Diagnostics", Current Tropical Medicine Reports 2015:2: 54-61 Abstract/PDF
CAD4TB is incorporated in the WHO e-Health Compendium. Also, CAD4TB has been certifiied as of the 2nd of April 2015 (issued to Radboud University Medical Center as the research partner in the consortium). The certificate can be viewed by clicking here.