Jennifer Nitsch1,2,3, Jordan Sack4, Michael W Halle5, Jan H Moltz6, April Wall7, Anna E Rutherford4, Ron Kikinis6,8,5, Hans Meine6,8. 1. Fraunhofer MEVIS Institute for Digital Medicine, Bremen, Germany. jennifer.nitsch@online.de. 2. Medical Image Computing Group, University of Bremen, Bremen, Germany. jennifer.nitsch@online.de. 3. Surgical Planning Laboratory, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA. jennifer.nitsch@online.de. 4. Division of Gastroenterology, Hepatology, and Endoscopy, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA. 5. Surgical Planning Laboratory, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA. 6. Fraunhofer MEVIS Institute for Digital Medicine, Bremen, Germany. 7. Department of Medicine, Brigham and Women's Hospital, Boston, MA, USA. 8. Medical Image Computing Group, University of Bremen, Bremen, Germany.
Abstract
PURPOSE: We aimed to develop a predictive model of disease severity for cirrhosis using MRI-derived radiomic features of the liver and spleen and compared it to the existing disease severity metrics of MELD score and clinical decompensation. The MELD score is compiled solely by blood parameters, and so far, it was not investigated if extracted image-based features have the potential to reflect severity to potentially complement the calculated score. METHODS: This was a retrospective study of eligible patients with cirrhosis ([Formula: see text]) who underwent a contrast-enhanced MR screening protocol for hepatocellular carcinoma (HCC) screening at a tertiary academic center from 2015 to 2018. Radiomic feature analyses were used to train four prediction models for assessing the patient's condition at time of scan: MELD score, MELD score [Formula: see text] 9 (median score of the cohort), MELD score [Formula: see text] 15 (the inflection between the risk and benefit of transplant), and clinical decompensation. Liver and spleen segmentations were used for feature extraction, followed by cross-validated random forest classification. RESULTS: Radiomic features of the liver and spleen were most predictive of clinical decompensation (AUC 0.84), which the MELD score could predict with an AUC of 0.78. Using liver or spleen features alone had slightly lower discrimination ability (AUC of 0.82 for liver and AUC of 0.78 for spleen features only), although this was not statistically significant on our cohort. When radiomic prediction models were trained to predict continuous MELD scores, there was poor correlation. When stratifying risk by splitting our cohort at the median MELD 9 or at MELD 15, our models achieved AUCs of 0.78 or 0.66, respectively. CONCLUSIONS: We demonstrated that MRI-based radiomic features of the liver and spleen have the potential to predict the severity of liver cirrhosis, using decompensation or MELD status as imperfect surrogate measures for disease severity.
PURPOSE: We aimed to develop a predictive model of disease severity for cirrhosis using MRI-derived radiomic features of the liver and spleen and compared it to the existing disease severity metrics of MELD score and clinical decompensation. The MELD score is compiled solely by blood parameters, and so far, it was not investigated if extracted image-based features have the potential to reflect severity to potentially complement the calculated score. METHODS: This was a retrospective study of eligible patients with cirrhosis ([Formula: see text]) who underwent a contrast-enhanced MR screening protocol for hepatocellular carcinoma (HCC) screening at a tertiary academic center from 2015 to 2018. Radiomic feature analyses were used to train four prediction models for assessing the patient's condition at time of scan: MELD score, MELD score [Formula: see text] 9 (median score of the cohort), MELD score [Formula: see text] 15 (the inflection between the risk and benefit of transplant), and clinical decompensation. Liver and spleen segmentations were used for feature extraction, followed by cross-validated random forest classification. RESULTS: Radiomic features of the liver and spleen were most predictive of clinical decompensation (AUC 0.84), which the MELD score could predict with an AUC of 0.78. Using liver or spleen features alone had slightly lower discrimination ability (AUC of 0.82 for liver and AUC of 0.78 for spleen features only), although this was not statistically significant on our cohort. When radiomic prediction models were trained to predict continuous MELD scores, there was poor correlation. When stratifying risk by splitting our cohort at the median MELD 9 or at MELD 15, our models achieved AUCs of 0.78 or 0.66, respectively. CONCLUSIONS: We demonstrated that MRI-based radiomic features of the liver and spleen have the potential to predict the severity of liver cirrhosis, using decompensation or MELD status as imperfect surrogate measures for disease severity.
Authors: Lili He; Hailong Li; Jonathan A Dudley; Thomas C Maloney; Samuel L Brady; Elanchezhian Somasundaram; Andrew T Trout; Jonathan R Dillman Journal: AJR Am J Roentgenol Date: 2019-05-23 Impact factor: 3.959
Authors: Chintan Parmar; Patrick Grossmann; Derek Rietveld; Michelle M Rietbergen; Philippe Lambin; Hugo J W L Aerts Journal: Front Oncol Date: 2015-12-03 Impact factor: 6.244
Authors: Uri Kartoun; Kathleen E Corey; Tracey G Simon; Hui Zheng; Rahul Aggarwal; Kenney Ng; Stanley Y Shaw Journal: PLoS One Date: 2017-10-25 Impact factor: 3.240
Authors: Xiaojian Chen; Kiyoko Oshima; Diane Schott; Hui Wu; William Hall; Yingqiu Song; Yalan Tao; Dingjie Li; Cheng Zheng; Paul Knechtges; Beth Erickson; X Allen Li Journal: PLoS One Date: 2017-06-02 Impact factor: 3.240
Authors: Ali A Mokdad; Alan D Lopez; Saied Shahraz; Rafael Lozano; Ali H Mokdad; Jeff Stanaway; Christopher J L Murray; Mohsen Naghavi Journal: BMC Med Date: 2014-09-18 Impact factor: 8.775