Deep Learning–Based Automated Imaging Classification of ADPKD

• Published in: Kidney international reports Vol. 9; no. 6; pp. 1802 - 1809
• Main Authors: Kim, Youngwoo, Bu, Seonah, Tao, Cheng, Bae, Kyongtae T., Steinman, Theodore, Wei, Jesse, Czarnecki, Peter, Pedrosa, Ivan, Braun, William, Nurko, Saul, Remer, Erick, Chapman, Arlene, Martin, Diego, Rahbari-Oskoui, Frederic, Mittal, Pardeep, Torres, Vicente, Hogan, Marie C., El-Zoghby, Ziad, Harris, Peter, Glockner, James, King, Bernard, Perrone, Ronald, Halin, Neil, Miskulin, Dana, Schrier, Robert, Brosnahan, Godela, Gitomer, Berenice, Kelleher, Cass, Masoumi, Amirali, Patel, Nayana, Winklhofer, Franz, Grantham, Jared, Yu, Alan, Wang, Connie, Wetzel, Louis, Moore, Charity G., Bost, James E., Bae, Kyongtae, Abebe, Kaleab Z., Miller, J. Philip, Thompson, Paul A., Briggs, Josephine, Flessner, Michael, Meyers, Catherine M., Star, Robert, Shayman, James, Henrich, William, Greene, Tom, Leonard, Mary, McCullough, Peter, Moe, Sharon, Rocco, Michael, Wendler, David
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.06.2024; Elsevier
• Subjects: atypical cyst; Clinical Research; deep learning; explainable artificial intelligence; polycystic kidney disease; risk factors; total kidney volume; explainable artificial intelligence; total kidney volume; deep learning; risk factors; atypical cyst; polycystic kidney disease
• ISSN: 2468-0249; 2468-0249
• DOI: 10.1016/j.ekir.2024.04.002

The Mayo imaging classification model (MICM) requires a prestep qualitative assessment to determine whether a patient is in class 1 (typical) or class 2 (atypical), where patients assigned to class 2 are excluded from the MICM application. We developed a deep learning–based method to automatically classify class 1 and 2 from magnetic resonance (MR) images and provide classification confidence utilizing abdominal T2-weighted MR images from 486 subjects, where transfer learning was applied. In addition, the explainable artificial intelligence (XAI) method was illustrated to enhance the explainability of the automated classification results. For performance evaluations, confusion matrices were generated, and receiver operating characteristic curves were drawn to measure the area under the curve. The proposed method showed excellent performance for the classification of class 1 (97.7%) and 2 (100%), where the combined test accuracy was 98.01%. The precision and recall for predicting class 1 were 1.00 and 0.98, respectively, with F1-score of 0.99; whereas those for predicting class 2 were 0.87 and 1.00, respectively, with F1-score of 0.93. The weighted averages of precision and recall were 0.98 and 0.98, respectively, showing the classification confidence scores whereas the XAI method well-highlighted contributing regions for the classification. The proposed automated method can classify class 1 and 2 cases as accurately as the level of a human expert. This method may be a useful tool to facilitate clinical trials investigating different types of kidney morphology and for clinical management of patients with autosomal dominant polycystic kidney disease (ADPKD). [Display omitted]