TY - GEN
T1 - A MINIMALLY SUPERVISED APPROACH FOR MEDICAL IMAGE QUALITY ASSESSMENT IN DOMAIN SHIFT SETTINGS
AU - Yang, Huijuan
AU - Coyner, Aaron S.
AU - Guretno, Feri
AU - Mien, Ivan Ho
AU - Foo, Chuan Sheng
AU - Campbell, J. Peter
AU - Ostmo, Susan
AU - Chiang, Michael F.
AU - Krishnaswamy, Pavitra
N1 - Publisher Copyright:
© 2022 IEEE
PY - 2022
Y1 - 2022
N2 - Accurate disease diagnosis requires objective assessment of clinical image quality. Automated image quality assessment (IQA) could enhance screening and diagnosis workflows. However, development of generalizable quality assessment tools requires large labeled clinical image datasets from different sites. Obtaining these datasets is often infeasible; and quality indicators may vary with acquisition settings due to domain shift. We introduce a minimally-supervised image quality assessment (MIQA) approach that can learn effectively with small datasets and limited labels in class-imbalanced domain shift scenarios. We formulate the IQA task as an anomaly detection problem, and use a small number of target domain images to identify a compact subset of source domain data for better representation of acceptable quality features. For this compact source domain dataset, we extract features with a pre-trained CNN, perform adaptive feature selection, and develop a one-class classifier to detect poor quality images. We evaluate our approach on two ophthalmology datasets, and show substantial AUC gains and improved cross-site generalizability over competitive baselines. Our approach has implications for improved image quality audit in many clinical settings.
AB - Accurate disease diagnosis requires objective assessment of clinical image quality. Automated image quality assessment (IQA) could enhance screening and diagnosis workflows. However, development of generalizable quality assessment tools requires large labeled clinical image datasets from different sites. Obtaining these datasets is often infeasible; and quality indicators may vary with acquisition settings due to domain shift. We introduce a minimally-supervised image quality assessment (MIQA) approach that can learn effectively with small datasets and limited labels in class-imbalanced domain shift scenarios. We formulate the IQA task as an anomaly detection problem, and use a small number of target domain images to identify a compact subset of source domain data for better representation of acceptable quality features. For this compact source domain dataset, we extract features with a pre-trained CNN, perform adaptive feature selection, and develop a one-class classifier to detect poor quality images. We evaluate our approach on two ophthalmology datasets, and show substantial AUC gains and improved cross-site generalizability over competitive baselines. Our approach has implications for improved image quality audit in many clinical settings.
KW - Medical image quality assessment
KW - class imbalance
KW - data scarcity
KW - domain shift
KW - minimally supervised
UR - http://www.scopus.com/inward/record.url?scp=85131239019&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85131239019&partnerID=8YFLogxK
U2 - 10.1109/ICASSP43922.2022.9746837
DO - 10.1109/ICASSP43922.2022.9746837
M3 - Conference contribution
AN - SCOPUS:85131239019
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 1286
EP - 1290
BT - 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022
Y2 - 23 May 2022 through 27 May 2022
ER -