TY - GEN
T1 - Individualized and Generalized Learner Models for Predicting Missed Hepatic Metastases
AU - Pillai, Parvathy Sudhir
AU - Hsieh, Scott
AU - Holmes, David
AU - Carter, Rickey
AU - Fletcher, Joel G.
AU - McCollough, Cynthia
N1 - Publisher Copyright:
© 2022 SPIE. All rights reserved.
PY - 2022
Y1 - 2022
N2 - The diagnostic performance of radiologist readers exhibits substantial variation that cannot be explained by CT acquisition protocol differences. Studying reader detectability from CT images may help identify why certain types of lesions are missed by multiple or specific readers. Ten subspecialized abdominal radiologists marked all suspected metastases in a multi-reader-multi-case study of 102 deidentified contrast-enhanced CT liver scans at multiple radiation dose levels. A reference reader marked ground truth metastatic and benign lesions with the aid of histopathology or tumor progression on later scans. Multi-slice image patches and 3D radiomic features were extracted from the CT images. We trained deep convolutional neural networks (CNN) to predict whether an average (generalized) or individual radiologist reader would detect or miss a specific metastasis from an image patch containing it. The individualized CNN showed higher performance with an area under the receiver operating characteristic curve (AUC) of 0.82 compared to a generalized one (AUC = 0.78) in predicting reader-specific detectability. Random forests were used to build the respective versions from radiomic features. Both the individualized (AUC = 0.64) and generalized (AUC = 0.59) predictors from radiomic features showed limited ability to differentiate detected from missed lesions. This shows that CNN can identify and learn automated features that are better predictors of reader detectability of lesions than radiomic features. Individualized prediction of difficult lesions may allow targeted training of idiosyncratic weaknesses but requires substantial training data for each reader.
AB - The diagnostic performance of radiologist readers exhibits substantial variation that cannot be explained by CT acquisition protocol differences. Studying reader detectability from CT images may help identify why certain types of lesions are missed by multiple or specific readers. Ten subspecialized abdominal radiologists marked all suspected metastases in a multi-reader-multi-case study of 102 deidentified contrast-enhanced CT liver scans at multiple radiation dose levels. A reference reader marked ground truth metastatic and benign lesions with the aid of histopathology or tumor progression on later scans. Multi-slice image patches and 3D radiomic features were extracted from the CT images. We trained deep convolutional neural networks (CNN) to predict whether an average (generalized) or individual radiologist reader would detect or miss a specific metastasis from an image patch containing it. The individualized CNN showed higher performance with an area under the receiver operating characteristic curve (AUC) of 0.82 compared to a generalized one (AUC = 0.78) in predicting reader-specific detectability. Random forests were used to build the respective versions from radiomic features. Both the individualized (AUC = 0.64) and generalized (AUC = 0.59) predictors from radiomic features showed limited ability to differentiate detected from missed lesions. This shows that CNN can identify and learn automated features that are better predictors of reader detectability of lesions than radiomic features. Individualized prediction of difficult lesions may allow targeted training of idiosyncratic weaknesses but requires substantial training data for each reader.
KW - Convolutional Neural Network
KW - Liver metastasis detection
KW - Low contrast detection
KW - Observer Performance
UR - http://www.scopus.com/inward/record.url?scp=85131877701&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85131877701&partnerID=8YFLogxK
U2 - 10.1117/12.2612745
DO - 10.1117/12.2612745
M3 - Conference contribution
AN - SCOPUS:85131877701
T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE
BT - Medical Imaging 2022
A2 - Mello-Thoms, Claudia R.
A2 - Mello-Thoms, Claudia R.
A2 - Taylor-Phillips, Sian
PB - SPIE
T2 - Medical Imaging 2022: Image Perception, Observer Performance, and Technology Assessment
Y2 - 21 March 2022 through 27 March 2022
ER -