Publication


Summary of IWBI presentations using Transpara technology

4th June 2020

Summary of IWBI presentations using Transpara technology

As the leaders in evidence based AI decision support, we are proud that Transpara technology which is FDA cleared for both 2D and 3D mammography, featured in 6 scientific presentations at the recent IWBI virtual workshop in Leuven.  Here is a brief summary below. 

Christiana Balta, Alejandro Rodriguez-Ruiz, Christoph Mieskes, Nico Karssemeijer, Sylvia H. Heywang-Köbrunner, “Going from double to single reading for screening exams labeled as likely normal by AI: what is the impact?,” Proc. SPIE 11513, 15th International Workshop on Breast Imaging
(IWBI2020), 115130D (22 May 2020); doi: 10.1117/12.2564179

Findings:  Transpara can improve breast cancer screening efficiency by pre-selecting likely normal exams where double-reading can be safely replaced by single reading resulting in 32.6% reading workload reduction.

Magnus Dustler, Victor Dahlblom, Anders Tingberg, Sophia Zackrisson, “The effect of breast density on the performance of deep learning-based breast cancer detection methods for mammography,” Proc. SPIE 11513, 15th International Workshop on Breast Imaging (IWBI2020), 1151324 (22 May 2020); doi: 10.1117/12.2564328

Findings:  Transpara risk categorization is not affected by density.  Transpara has the potential to improve screening sensitivity even for women with high breast density.

Victor Dahlblom, Anders Tingberg, Sophia Zackrisson, Magnus Dustler, “Personalised breast cancer screening with selective addition of digital breast
tomosynthesis through artificial intelligence,” Proc. SPIE 11513, 15th International Workshop on Breast Imaging (IWBI2020), 115130C (22 May 2020); doi: 10.1117/12.2564344

Findings:  Using DBT only for cases marked as highly suspicious by Transpara in mammography could be an alternative to a complete DBT screening.  If adding DBT for women with Transpara Score 10 in mammography, 12% of the women would have DBT added and 26% more cancers would be detected, with a 21% increase in false positives.

Alejandro Rodríguez-Ruiz, Kristina Lång, Albert Gubern-Merida, Mireille Broeders, Gisella Gennaro, Paola Clauser, Thomas Helbich, Thomas Mertelmeier, Margarita Chevalier, Matthew Wallis, Ingvar Andersson, Sophia Zackrisson, R. M. Mann, I. Sechopoulos, “Can AI serve as an independent second reader of mammograms? a simulation study,” Proc. SPIE 11513, 15th
International Workshop on Breast Imaging (IWBI2020), 115130O (22 May 2020); doi: 10.1117/12.2564114

Findings:  Using Transpara as a second reader in a double reading setting, workload could be reduced by 44%, without an impact in sensitivity and with a possible specificity increase by 5.3%.

Anna Bejnö, Gustav Hellgren, Alejandro Rodriguez-Ruiz, Predrag R. Bakic, Sophia Zackrisson, Anders Tingberg, Magnus Dustler, “Artificial intelligence
together with mechanical imaging in mammography,” Proc. SPIE 11513, 15th International Workshop on Breast Imaging (IWBI2020), 1151320 (22 May 2020); doi: 10.1117/12.2564107

Findings:  Mechanical imaging (MI) estimates the relative stiffness of suspicious breast abnormalities by measuring the distribution of pressure on the compressed breast.  MI combined with Transpara has the potential to increase the accuracy of mammography screening.

Koen Dercksen, Michiel Kallenberg, Jaap Kroes, “Robust multi-vendor breast region segmentation using deep learning,” Proc. SPIE 11513, 15th International Workshop on Breast Imaging (IWBI2020), 115131A (22 May 2020); doi: 10.1117/12.2564108

Findings:  Transpara can provide robust breast region segmentations in a multimodal multi-vendor setting.