Institute of Fundamental Technological Research
Polish Academy of Sciences


John Erdman

University of Illinois at Urbana-Champaign (US)

Recent publications
1.  Han A., Byra M., Heba E., Andre M.P., Erdman J.W.Jr., Loomba R., Sirlin C.B., O'Brien W.D.Jr., Noninvasive diagnosis of nonalcoholic fatty liver disease and quantification of liver fat with radiofrequency ultrasound data using one-dimensional convolutional neural networks, Radiology, ISSN: 0033-8419, DOI: 10.1148/radiol.2020191160, Vol.295, No.2, pp.342-350, 2020

Background: Radiofrequency ultrasound data from the liver contain rich information about liver microstructure and composition. Deep learning might exploit such information to assess nonalcoholic fatty liver disease (NAFLD). Purpose: To develop and evaluate deep learning algorithms that use radiofrequency data for NAFLD assessment, with MRI-derived proton density fat fraction (PDFF) as the reference. Materials and Methods: A HIPAA-compliant secondary analysis of a single-center prospective study was performed for adult participants with NAFLD and control participants without liver disease. Participants in the parent study were recruited between February 2012 and March 2014 and underwent same-day US and MRI of the liver. Participants were randomly divided into an equal number of training and test groups. The training group was used to develop two algorithms via cross-validation: a classifier to diagnose NAFLD (MRI PDFF ≥ 5%) and a fat fraction estimator to predict MRI PDFF. Both algorithms used one-dimensional convolutional neural networks. The test group was used to evaluate the classifier for sensitivity, specificity, positive predictive value, negative predictive value, and accuracy and to evaluate the estimator for correlation, bias, limits of agreements, and linearity between predicted fat fraction and MRI PDFF. Results: A total of 204 participants were analyzed, 140 had NAFLD (mean age, 52 years ± 14 [standard deviation]; 82 women) and 64 were control participants (mean age, 46 years ± 21; 42 women). In the test group, the classifier provided 96% (95% confidence interval [CI]: 90%, 99%) (98 of 102) accuracy for NAFLD diagnosis (sensitivity, 97% [95% CI: 90%, 100%], 68 of 70; specificity, 94% [95% CI: 79%, 99%], 30 of 32; positive predictive value, 97% [95% CI: 90%, 99%], 68 of 70; negative predictive value, 94% [95% CI: 79%, 98%], 30 of 32). The estimator-predicted fat fraction correlated with MRI PDFF (Pearson r = 0.85). The mean bias was 0.8% (P = .08), and 95% limits of agreement were -7.6% to 9.1%. The predicted fat fraction was linear with an MRI PDFF of 18% or less (r = 0.89, slope = 1.1, intercept = 1.3) and nonlinear with an MRI PDFF greater than 18%. Conclusion: Deep learning algorithms using radiofrequency ultrasound data are accurate for diagnosis of nonalcoholic fatty liver disease and hepatic fat fraction quantification when other causes of steatosis are excluded.

Han A. - University of Illinois at Urbana-Champaign (US)
Byra M. - IPPT PAN
Heba E. - other affiliation
Andre M.P. - University of California (US)
Erdman J.W.Jr. - University of Illinois at Urbana-Champaign (US)
Loomba R. - University of California (US)
Sirlin C.B. - University of California (US)

Conference abstracts
1.  Byra M., Han A., Boehringer A., Zhang Y., Erdman J., Loomba R., Valasek M., Sirlin C., O Brien W., Andre M., Quantitative liver fat fraction measurement by multi-view sonography using deep learning and attention maps, ASA, 178th Meeting of the Acoustical Society of America, 2019-12-02/12-06, San Diego (US), DOI: 10.1121/1.5136936, Vol.146, No.4, pp.2809-1, 2019

Qualitative sonography is used to assess nonalcoholic fatty liver disease (NAFLD), an important health issue worldwide. We used B-mode image deep-learning to objectively assess NAFLD in 4 views of the liver (hepatic veins at confluence with inferior vena cava, right portal vein, right posterior portal vein and liver/kidney) in 135 patients with known or suspected NAFLD. Transfer learning with a deep convolutional neural network (CNN) was applied for quantifying fat fraction and diagnosing fatty liver (≥ 5%) using contemporaneous MRI-PDFF as ground truth. Single and multi-view learning approaches were compared. Class activation mapping generated attention maps to highlight regions important for deep learning-based recognition. The most accurate single view was hepatic veins, with area under the receiver operating characteristic curve (AUC) of 0.86 and Spearman’s rank correlation coefficient of 0.65. A multi-view ensemble of deep-learning models trained for each view separately improved AUC (0.93) and correlation coefficient (0.76). Attention maps highlighted regions known to be used by radiologists in their qualitative assessment, e.g., hepatic vein-parenchyma interface and liver-kidney interface. Machine learning of four liver views can automatically and objectively assess liver fat. Class activation mapping suggests that the CNN focuses on similar features as radiologists. [No. R01DK106419.]

Byra M. - IPPT PAN
Han A. - University of Illinois at Urbana-Champaign (US)
Boehringer A. - University of California (US)
Zhang Y. - University of California (US)
Erdman J. - University of Illinois at Urbana-Champaign (US)
Loomba R. - University of California (US)
Valasek M. - University of California (US)
Sirlin C. - University of California (US)
O Brien W. - University of Illinois at Urbana-Champaign (US)
Andre M. - University of California (US)

Category A Plus


logo ippt            Pawińskiego 5B, 02-106 Warsaw
  +48 22 826 12 81 (central)
  +48 22 826 98 15

Find Us

© Institute of Fundamental Technological Research Polish Academy of Sciences 2021