CAD4TB
Computer-Aided Detection for Tuberculosis
CAD4TB was designed to help (non-expert) readers detect tuberculosis more accurately and cost-effectively. The speed of digital X-rays combined with deep learning and remote expertise make CAD4TB a valuable asset in the fight against TB.
Tuberculosis solution
Tuberculosis (TB) is one of the deadliest infectious disease in the world. It affects over 10 million people with a mortality rate of 1.4 million every year. However, when diagnosed in time, TB is generally curable within 6 months. Unfortunately, about one-third of patients fail to receive a quick and accurate diagnosis or effective treatment, making them more likely to die from this curable disease.
Solving the problem of tuberculosis requires a diagnostic solution that is affordable, fast, accurate and easy to implement in resource-constrained and hard-to-reach regions.
The World Health Organization
In March 2021, the WHO recommended for the first time that CAD software may be used in stead of human readers for the interpretation of digital chest X-rays for screening and triage for TB diseases, and CAD4TB is one of the three products included in the technical evaluation. Additionally, CAD4TB meets the WHO’s Target Product Profile (TPP) of triage tests (≥90% sensitivity and ≥70% specificity) . CAD4TB has been extensively used for TB projects worldwide supported by the Global Fund, USAID, Stop TB Partnership / TB REACH, EU and the World Bank.
CAD4TB platform: Enable AI and digitalise TB screening data management
To further support the END-TB strategy by finding the missing cases, Delft has developed additional screening analytics and management tools to enable users to rapidly screen risk groups and support prevalence surveys, and collect all information into one platform. CAD4TB now includes not only artificial intelligence, but also modules for patient registration, symptoms registration (through customisable forms) and patient reporting. Apart from helping to digitalise the TB screening process, the ‘insights’ module featured in the CAD4TB platform provides valuable insights to management on the program, reflecting those screened in the program through distribution in gender, age and CAD4TB scores.
Diagnosis and treatment within one day.
Deep learning for TB detection
The CAD4TB software starts with a quick inspection of a new X-ray image. It normalizes the image to correct for input from various modalities, and then detects the many different abnormal structures that may be related to TB. The system has been trained in the detection of these abnormalities by applying deep learning to thousands of healthy and diseased X-rays from over 15 countries. CAD4TB also received expert feedback from lung specialists, making it extremely accurate and reliable.
Active TB Screening
The output of the CAD4TB software is a score between 0 and 100, indicating the likelihood that the subject on the image has active TB. The score also includes a texture heat map overlaying the original X-ray image which indicates abnormalities. Only if the output score is higher than a certain threshold will the patient need a subsequent sputum test to finalize the diagnosis. This process brings down both the cost and the time-to-treatment.
Scientific Support for CAD4TB
Delft’s innovative approach to screening and diagnosing tuberculosis has been supported by many of our colleagues in the scientific community. Read the in-depth evaluations of CAD4TB in scientific articles:
Within 30 seconds, the CAD4TB software analyses a chest X-ray, detects abnormalities and indicates the likelihood that active TB is present. This allows costly TB tests to be used only in suspected cases.
Bi-directional Tuberculosis/COVID-19 Screening
The COVID-19 pandemic is threatening the global goals to fight to end TB by 2030. Thereupon, the importance to implement prioritized mitigation measures and innovations for TB programs have been indicated from across the globe. This includes prioritizing bi-directional screening/testing for TB and COVID-19, using X-ray with CAD (computer-aided detection).
At the start of the pandemic, we developed CAD4COVID which uses the same technical core and quality standard as our proven CAD4TB software. With years of digital X-ray and CAD field implementation expertise, our response to the bi-directional screening equation pertaining to TB and COVID-19 is apparent.
Delft Projects Using CAD4TB
Lagos State
In February 2021, we successfully completed the last phase of the Lagos State Ministry of Health project which began in December 2020. Altogether 21 digital x-ray installations, including 3 mobile screening vans, 8 semi-mobile containerized x-ray systems and 10 room installations were made.
Jakarta – Prisons
In collaboration with Fullerton Health, the CAD4TB software was used in several prisons and a factory in Jakarta. Prisoners were first screened on symptoms and in the case symptoms were present that could indicate tuberculosis, a chest X-ray (CXR) was taken in a mobile van.
Mozambique
Delft provided an EasyDR X-ray machine with our Computer Aided Detection for Tuberculosis software (CAD4TB). The X-ray screening on TB and silicosis, audio and other tests will be free of charge for all the migrating mine workers.
eSwatini
Early 2016, the National Tuberculosis Programme of Swaziland ordered a mobile OneStopTB clinic to screen around 100.000 inhabitants in Swaziland. They combined the EasyDR digital X-ray with the innovative CAD4TB software, which allowed for an instant triage of suspects. Besides this Delft delivered two EasyPortable X-rays to reach the remotest areas in the mountainous country.
Sierra Leone
In 2018, 12 digital X-ray systems were installed at the Connaught and at 10 district hospitals, through a tender by IDA Foundation that was Global Fund supported. These systems are entirely solar powered and include charging batteries that allow 4-6 hours of operation at night when grid power may be unavailable.
Bangladesh
The focus was on increasing early case-detection, reporting and correct treatment of TB, serving poor communities in Dhaka, Bangladesh. The project was rolled out using an innovative pay-per-use scheme. The lung screening center’s charged a modest fee for every “lung screening” and part of this fee supported the cost of the X-ray units.
Malawi
We provided as many as 7 OneStopTB clinics, making this our biggest OneStopTB project in East-Africa yet. All 7 OneStopTB clinics have an X-ray compartment with a control room, a separate laboratory for the GeneXpert test (both air-conditioned), teleradiology and computers with our CAD4TB software.
