CAD4TB Print

Computer Aided Detection for TB (CAD4TB - v6 CE certified) developed by Radboud University of Nijmegen in cooperation with Delft Imaging Systems outperforms trained human readers to detect abnormalities consistent with TB. This means CAD can be used cost effectively as a 1 minute low cost triage tool before Xpert. Partial financing for this R&D was obtained from the Dutch Ministry of Economic Affairs & Innovations with support from CheckTB!

At the recent UNION Conferences CAD4TB sensitivity and specificity scores on data sets from Bangladesh, Cameroon (WHO/TB REACH), Philippines (WHO), Tanzania (USAID) and South Africa (GLOBAL FUND) were presented, confirming the potential of CAD4TB as a rapid triage before Xpert MTB/RIF. CAD4TB is currently under evaluation by WHO.

http://www.stoptb.org/wg/new_diagnostics/assets/documents/F.vanDoren_CAD%20Digital%20X-ray.pdf

 

Computers improve quality and efficiency of screening

ü 90% of lesions initially missed by human readers were visible;

ü less than 50% of lesions < 1 cm are seen by human reader[1]


Although operational research is still in progress, CAD is particularly efficient whilst used in high throughput screening programs when:

ü numerous chest X-rays need to be evaluated in a short period;

ü human readers are often neither fast nor accurate enough.


[1] Manning DJ et al, Br J Rad 2004; Muhm JR et al Radiology 1983

CAD objectives & use

ü Fit for rapid triage in high risk groups;

ü Sensitivity & specificity equal or better than trained human reader;

ü Reading and report within 1 minute at almost 0 expenses;

ü CAD abnormality threshold (%) can be set for Xpert® eligibility depending on Xpert® capacity (systems and number of cartridges).

 

CAD design

The CAD system follows the general machine learning approach of supervised learning. In supervised learning the system learns from a set of examples, called the training data. The training data consists of a number of labelled images of which the TB status is known. For the CAD4TB system there are two classes and the labels are normal and suspect for TB. A number of features is extracted from these images and combined to form a feature vector for each image. These feature vectors are used to train a TB classifier. This TB classifier incorporating state-of-the-art pattern recognition techniques can then be used to classify images of which the label is not known.

Computer Aided Diagnosis will enable automated accurate pre-screening on site by the digital X-ray system allowing radiologists or pulmonologists to focus more on analysing the TB suspect images. In addition CAD4TB can contribute to fewer "over-reading" by less experienced radiographers or clinical officers. CAD in combination with molecular tests such as Xpert innovates screening for early case detection. This CAD-Xpert combination enables rapid and more cost effective active case finding in high risk groups such as PLWH, diabetecs, mine workers and inmates.

Possibilities for Future Use CAD4TB

To increase TB prevalence survey accuracy and to reduce the cost of risk group screening programs CAD offers the following diagnostic functionality:

  • provide a probability of abnormalities consistent with TB between 0 and 100
  • provide markers around suspect regions
  • integrate with CRRS scoring system or PACS
  • select and present similar images for reference

CAD4TB also allows for pre-selecting its sensitivity, which is of special value for prevalence surveys. Now that the CAD4TB protocol is optimised the number of CXR requiring human reading can be reduced to less than 20%. In upcoming TB prevalence surveys in Botswana and Mozambique CAD4TB will be used as the radiological gold standard.

The progress

The “Computer Aided Detection for TB (CAD4TB)” project is still in development as the detection of other lung diseases by the algorithm will further increase its specificity. The University of Nijmegen/Thirona is now involved on the Dutch side together with Delft Imaging Systems they collaborate to further optimize the accuracy of CAD using deep learning techniques.

In order to achieve this, tens of thousands of digital images taken with the same technology direct digital X-ray systems were analysed in the same systematic way. Various state-of-the-art pattern recognition techniques are applied to continuously improve sensitivity and specificity scores.

Expected Outcome

The objective of the joint research was to achieve an electronic TB screening capability with a sensitivity of 90% and a specificity of 80%. The specificity target is the biggest challenge for the scientists. It is expected that on the basis of reading thousands of images (including access to patient information like HIV status), that also patterns consistent with TB typical for co-infected patients will be found. This will further enhance the role of X-ray as a screening tool to effectively detect TB suspects in people living with HIV.

CAD improves TB screening and supports active case finding by:

A high percentage of TB suspect cases is identified automatically wherever the digital camera is used.

    - Supporting radiographers and/or clinical officers with first image reading;
    - Reducing number of human image interpretations and under- or over reading;
    - Providing a powerful tool in prevalence studies;
    - Offering an effective quality assessment tool for health workers and NTP;
    - Cutting screening and diagnostics costs.

Reference images in the database will support radiographers to interpret Chest images on the spot with higher accuracy.
Only a limited number of complex Chest images will require external radiological expertise.

While developed for TB, this system will be enhanced in the future to spot other lung diseases as well such as COPD. It is thus expected to contribute to a broad healthcare strengthening impact in areas where this is most needed.

More than 25 countries already use CAD4TB in their TB programs. In below graph the role of CAD in these programs is illustrated. IRD in Pakistan - one of the early adopters of CAD4TB with WHO/TB REACH support - is now planning to screen up to 5 million people using CAD4TB as a rapid triage.

Philippines, Southern Africa, UK, The Gambia and Tanzania

More recent studies with digital chest images from Southern Africa, London Find & Treat, the Gambian and Tanzanian TB prevalence surveys, confirm that CAD is already as accurate as the trained human reader. In the Find & Treat database CAD scored 93 sensitivity at 65% specificity. At the UNION Conference in Kuala Lumpur a linear relationship between CAD score and TB detection among different strata of patients was presented by Zambart confirming the potential value to use CAD as a screening tool before Xpert.

Study World Health Organisation in Philippines

A feasibility study was being performed on CAD4TB performance in the Philippines to automatically interpret digital chest images made for migrant TB screening programs. This study is a collaboration between Radboud University and WHO with initial liaison through KNCV Tuberculosis Foundation and CheckTB!.

12,256 CXRs from Palawan provincial areas of the Philippines5; CAD4TB achieved 90.0% sensitivity at 80.0% specificity;Research performed in cooperation with WHO Philippines. 46th Union World Conference on Lung Health, 2015; Rick H.H.M. Philipsen, Clara I. Sánchez, P. Maduskar, J. Melendez, B. van Ginneken and W.J. Lew  Computerized Chest Radiography Screening to Detect Tuberculosis in the Philippines, 2015. Radboud University Medical Center, Nijmegen, the Netherlands and WHO Country Office, Manilla, Philippines.

12,256 CXRs from Palawan provincial areas of the Philippines5; CAD4TB achieved 90.0% sensitivity at 80.0% specificity;Research performed in cooperation with WHO Philippines. 46th Union World Conference on Lung Health, 2015; Rick H.H.M. Philipsen, Clara I. Sánchez, P. Maduskar, J. Melendez, B. van Ginneken and W.J. Lew  Computerized Chest Radiography Screening to Detect Tuberculosis in the Philippines, 2015. Radboud University Medical Center, Nijmegen, the Netherlands and WHO Country Office, Manilla, Philippines.

Conclusions

From the various studies it can be concluded, that CAD4TB already performs similar or better than the human readers in classifying radiographs as being normal or suspect for TB. The performance of both human readers and the CAD system is expected to improve significantly with more training. CAD as a rapid triage before Xpert has the potential to cost-effectively scale up systematic screening of high risk groups.

Can your organization participate in the study?

As TB can manifest itself in different forms in people of different regional and ethnic backgrounds, the researchers include images from different countries. If your organization would want to participate and patient privacy rules would allow this, then please contact This e-mail address is being protected from spambots. You need JavaScript enabled to view it This e-mail address is being protected from spambots. You need JavaScript enabled to view it for further information.

CAD4TB availability to support TB programs