{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T18:55:45Z","timestamp":1785610545072,"version":"3.56.0"},"reference-count":26,"publisher":"Wiley","issue":"8","license":[{"start":{"date-parts":[[2023,3,13]],"date-time":"2023-03-13T00:00:00Z","timestamp":1678665600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2023,8]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Background<\/jats:title><jats:p>von Hippel\u2013Lindau syndrome (VHL) is an autosomal dominant hereditary syndrome with an increased predisposition of developing numerous cysts and tumors, almost exclusively clear cell renal cell carcinoma (ccRCC). Considering the lifelong surveillance in such patients to monitor the disease, patients with VHL are preferentially imaged using MRI to eliminate radiation exposure.<\/jats:p><\/jats:sec><jats:sec><jats:title>Purpose<\/jats:title><jats:p>Segmentation of kidney and tumor structures on MRI in VHL patients is useful in lesion characterization (e.g., cyst vs. tumor), volumetric lesion analysis, and tumor growth prediction. However, automated tasks such as ccRCC segmentation on MRI is sparsely studied. We develop segmentation methodology for ccRCC on T1 weighted precontrast, corticomedullary, nephrogenic, and excretory contrast phase MRI.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We applied a new neural network approache using a novel differentiable decision forest, called hinge forest (HF), to segment kidney parenchyma, cyst, and ccRCC tumors in 117 images from 115 patients. This data set represented an unprecedented 504 ccRCCs with 1171 cystic lesions obtained at five different MRI scanners. The HF architecture was compared with U\u2010Net on 10 randomized splits with 75% used for training and 25% used for testing. Both methods were trained with Adam using default parameters () over 1000 epochs. We further demonstrated some interpretability of our HF method by exploiting decision tree structure.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The HF achieved an average kidney, cyst, and tumor Dice similarity coefficient (DSC) of 0.75 \u00b1 0.03, 0.44 \u00b1 0.05, 0.53 \u00b1 0.04, respectively, while U\u2010Net achieved an average kidney, cyst, and tumor DSC of 0.78 \u00b1 0.02, 0.41 \u00b1 0.04, 0.46 \u00b1 0.05, respectively. The HF significantly outperformed U\u2010Net on tumors while U\u2010Net significantly outperformed HF when segmenting kidney parenchymas ().<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>For the task of ccRCC segmentation, the HF can offer better segmentation performance compared to the traditional U\u2010Net architecture. The leaf maps can glean hints about deep learning features that might prove to be useful in other automated tasks such as tumor characterization.<\/jats:p><\/jats:sec>","DOI":"10.1002\/mp.16303","type":"journal-article","created":{"date-parts":[[2023,3,1]],"date-time":"2023-03-01T03:10:30Z","timestamp":1677640230000},"page":"5020-5029","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":8,"title":["Deep learning\u2010based decision forest for hereditary clear cell renal cell carcinoma segmentation on MRI"],"prefix":"10.1002","volume":"50","author":[{"given":"Nathan","family":"Lay","sequence":"first","affiliation":[{"name":"Artificial Intelligence Resource Molecular Imaging Branch National Cancer Institute Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pouria Yazdian","family":"Anari","sequence":"additional","affiliation":[{"name":"Radiology and Imaging Sciences Clinical Center National Institutes of Health Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aditi","family":"Chaurasia","sequence":"additional","affiliation":[{"name":"Radiology and Imaging Sciences Clinical Center National Institutes of Health Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fatemeh Dehghani","family":"Firouzabadi","sequence":"additional","affiliation":[{"name":"Radiology and Imaging Sciences Clinical Center National Institutes of Health Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stephanie","family":"Harmon","sequence":"additional","affiliation":[{"name":"Artificial Intelligence Resource Molecular Imaging Branch National Cancer Institute Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Evrim","family":"Turkbey","sequence":"additional","affiliation":[{"name":"Radiology and Imaging Sciences Clinical Center National Institutes of Health Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rabindra","family":"Gautam","sequence":"additional","affiliation":[{"name":"Urologic Oncology Branch Center for Cancer Research National Cancer Institute Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Safa","family":"Samimi","sequence":"additional","affiliation":[{"name":"Radiology and Imaging Sciences Clinical Center National Institutes of Health Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maria J.","family":"Merino","sequence":"additional","affiliation":[{"name":"Laboratory of Pathology Center for Cancer Research National Cancer Institute Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mark W.","family":"Ball","sequence":"additional","affiliation":[{"name":"Urologic Oncology Branch Center for Cancer Research National Cancer Institute Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"William Marston","family":"Linehan","sequence":"additional","affiliation":[{"name":"Urologic Oncology Branch Center for Cancer Research National Cancer Institute Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Baris","family":"Turkbey","sequence":"additional","affiliation":[{"name":"Artificial Intelligence Resource Molecular Imaging Branch National Cancer Institute Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ashkan A.","family":"Malayeri","sequence":"additional","affiliation":[{"name":"Radiology and Imaging Sciences Clinical Center National Institutes of Health Bethesda Maryland USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2023,3,13]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.1038\/ejhg.2010.175"},{"key":"e_1_2_8_3_1","doi-asserted-by":"crossref","unstructured":"RonnebergerO FischerP BroxT.U\u2010Net: convolutional networks for biomedical image segmentation. In:International Conference on Medical Image Computing and Computer\u2010assisted Intervention.2015:234\u2010241.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_8_4_1","volume-title":"Automated Detection of Renal Masses in Contrast\u2010Enhanced MRI using Deep Learning Methods","author":"Agarwal A.","year":"2021"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2019.01.012"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101821"},{"key":"e_1_2_8_7_1","unstructured":"HellerN SathianathenNJ KalaparaAA et\u00a0al.The KiTS19 challenge data: 300 kidney tumor cases with clinical context CT semantic segmentations and surgical outcomes. arXiv preprint arXiv:1904.00445 2019."},{"key":"e_1_2_8_8_1","doi-asserted-by":"crossref","unstructured":"IsenseeF JaegerPF KohlSAA PetersenJP Maier\u2010HeinKH.nnU\u2010Net: self\u2010adapting framework for u\u2010net\u2010based medical image segmentation. Bildverarbeitung f\u00fcr die Medizin 2019: Algorithmen\u2013Systeme\u2013Anwendungen. Proceedings des Workshops vom 17. bis 19. M\u00e4rz 2019 in L\u00fcbeck. Springer Fachmedien Wiesbaden 2019","DOI":"10.1007\/978-3-658-25326-4_7"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-0-387-84858-7"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1109\/34.817409"},{"key":"e_1_2_8_11_1","unstructured":"FrosstN HintonG.Distilling a neural network into a soft decision tree. CEUR Workshop Proceedings 2017."},{"key":"e_1_2_8_12_1","unstructured":"HazimehH PonomarevaN MolP TanZ MazumderR.The tree ensemble layer: differentiability meets conditional computation. In:ICML.2020:4138\u20104148."},{"key":"e_1_2_8_13_1","doi-asserted-by":"crossref","unstructured":"KontschiederP FiterauM CriminisiA BuloSR.Deep neural decision forests. In:ICCV.IEEE;2015:1467\u20101475.","DOI":"10.1109\/ICCV.2015.172"},{"key":"e_1_2_8_14_1","unstructured":"LeeG\u2010H JaakkolaTS.Oblique decision trees from derivatives of ReLU networks. In:ICLR.2019."},{"key":"e_1_2_8_15_1","unstructured":"TannoR ArulkumaranK AlexanderD CriminisiA NoriA.Adaptive neural trees. In:ICML.2019:6166\u20106175."},{"key":"e_1_2_8_16_1","unstructured":"YangY MorilloIG HospedalesTM.Deep neural decision trees. ICML Workshop on Human Interpretability in Machine Learning (WHI 2018) Stockholm Sweden 2018."},{"key":"e_1_2_8_17_1","unstructured":"ZantedeschiV KusnerM NiculaeV.Learning binary decision trees by argmin differentiation. In:International Conference on Machine Learning.PMLR;2021:12298\u201012309."},{"key":"e_1_2_8_18_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.1910340618"},{"key":"e_1_2_8_19_1","first-page":"1","article-title":"Statistical comparisons of classifiers over multiple data sets","volume":"7","author":"Dem\u0161ar J","year":"2006","journal-title":"J Mach Learn Res"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2019.05.002"},{"key":"e_1_2_8_21_1","unstructured":"IoffeS SzegedyC.Batch normalization: accelerating deep network training by reducing internal covariate shift. In:International Conference on Machine Learning.PMLR;2015:448\u2010456."},{"key":"e_1_2_8_22_1","unstructured":"SimonyanK ZissermanA.Very deep convolutional networks for large\u2010scale image recognition. International Conference on Learning Representations 2015."},{"key":"e_1_2_8_23_1","unstructured":"KingmaDP BaJ.Adam: a method for stochastic optimization. International Conference on Learning Representations 2015."},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2009.2013851"},{"key":"e_1_2_8_25_1","unstructured":"SimonyanK VedaldiA ZissermanA.Deep inside convolutional networks: visualising image classification models and saliency maps. Workshop at International Conference on Learning Representations 2014."},{"key":"e_1_2_8_26_1","doi-asserted-by":"crossref","unstructured":"ZeilerMD FergusR.Visualizing and understanding convolutional networks. European Conference on Computer Vision.2014:818\u2010833.","DOI":"10.1007\/978-3-319-10590-1_53"},{"key":"e_1_2_8_27_1","doi-asserted-by":"crossref","unstructured":"SelvarajuRR CogswellM DasA VedantamR ParikhD BatraD.Grad\u2010CAM: visual explanations from deep networks via gradient\u2010based localization. In:Proceedings of the IEEE International Conference on Computer Vision.2017:618\u2010626.","DOI":"10.1109\/ICCV.2017.74"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.16303","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.16303","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.16303","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,22]],"date-time":"2025-08-22T14:54:36Z","timestamp":1755874476000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.16303"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,3,13]]},"references-count":26,"journal-issue":{"issue":"8","published-print":{"date-parts":[[2023,8]]}},"alternative-id":["10.1002\/mp.16303"],"URL":"https:\/\/doi.org\/10.1002\/mp.16303","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,3,13]]},"assertion":[{"value":"2022-07-18","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2023-01-04","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2023-03-13","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}