H Akai1, K Yasaka1, A Kunimatsu1, M Nojima2, T Kokudo3, N Kokudo4, K Hasegawa3, O Abe5, K Ohtomo6, S Kiryu7. 1. Department of Radiology, Institute of Medical Science, University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, 108-8639 Tokyo, Japan. 2. Division of Advanced Medicine Promotion, The Advanced Clinical Research Center, Institute of Medical Science, University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, 108-8639 Tokyo, Japan. 3. Division of Hepato-Biliary-Pancreatic Surgery, Department of Surgery, Graduate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyou-ku, 113-8655 Tokyo, Japan. 4. Department of Surgery, Center Hospital of the National Center of Global Health and Medicine, 1-21-1 Toyama, Shinnjuku-ku, 162-8655 Tokyo, Japan. 5. Department of Radiology, Graduate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, 113-8655 Tokyo, Japan. 6. International University of Health and Welfare, 2600-1 Kitakanemaru, Ohtawara City, 324-8501 Tochigi, Japan. 7. Department of Radiology, International University of Health and Welfare Hospital, 537-3 Iguchi, Nasushiobara, 329-2763 Tochigi, Japan. Electronic address: kiryu-tky@umin.ac.jp.
Abstract
RATIONALE AND OBJECTIVES: To investigate the impact of random survival forest (RSF) classifier trained by radiomics features over the prediction of the overall survival of patients with resectable hepatocellular carcinoma (HCC). MATERIALS AND METHODS: The dynamic computed tomography data of 127 patients (97 men, 30 women; mean age, 68 years) newly diagnosed with resectable HCC were retrospectively analyzed. After manually setting the region of interest to include the tumor within the slice at its maximum diameter, texture analyses were performed with or without a Laplacian of Gaussian filter. Using the extracted 96 histogram based texture features, RSFs were trained using 5-fold cross-validation to predict the individual risk for each patient on disease free survival (DFS) and overall survival (OS). The associations between individual risk and DFS or OS were evaluated using Kaplan-Meier analysis. The effects of the predicted individual risk and clinical variables upon OS were analyzed using a multivariate Cox proportional hazards model. RESULTS: Among the 96 histogram based texture features, RSF extracted 8 of high importance for DFS and 15 for OS. The RSF trained by these features distinguished two patient groups with high and low predicted individual risk (P=1.1×10-4 for DFS, 4.8×10-7 for OS). Based on the multivariate Cox proportional hazards model, high predicted individual risk (hazard ratio=1.06 per 1% increase, P=8.4×10-8) and vascular invasion (hazard ratio=1.74, P=0.039) were the only unfavorable prognostic factors. CONCLUSIONS: The combination of radiomics analysis and RSF might be useful in predicting the prognosis of patients with resectable HCC.
RATIONALE AND OBJECTIVES: To investigate the impact of random survival forest (RSF) classifier trained by radiomics features over the prediction of the overall survival of patients with resectable hepatocellular carcinoma (HCC). MATERIALS AND METHODS: The dynamic computed tomography data of 127 patients (97 men, 30 women; mean age, 68 years) newly diagnosed with resectable HCC were retrospectively analyzed. After manually setting the region of interest to include the tumor within the slice at its maximum diameter, texture analyses were performed with or without a Laplacian of Gaussian filter. Using the extracted 96 histogram based texture features, RSFs were trained using 5-fold cross-validation to predict the individual risk for each patient on disease free survival (DFS) and overall survival (OS). The associations between individual risk and DFS or OS were evaluated using Kaplan-Meier analysis. The effects of the predicted individual risk and clinical variables upon OS were analyzed using a multivariate Cox proportional hazards model. RESULTS: Among the 96 histogram based texture features, RSF extracted 8 of high importance for DFS and 15 for OS. The RSF trained by these features distinguished two patient groups with high and low predicted individual risk (P=1.1×10-4 for DFS, 4.8×10-7 for OS). Based on the multivariate Cox proportional hazards model, high predicted individual risk (hazard ratio=1.06 per 1% increase, P=8.4×10-8) and vascular invasion (hazard ratio=1.74, P=0.039) were the only unfavorable prognostic factors. CONCLUSIONS: The combination of radiomics analysis and RSF might be useful in predicting the prognosis of patients with resectable HCC.
Authors: Bradley Spieler; Carl Sabottke; Ahmed W Moawad; Ahmed M Gabr; Mustafa R Bashir; Richard Kinh Gian Do; Vahid Yaghmai; Radu Rozenberg; Marielia Gerena; Joseph Yacoub; Khaled M Elsayes Journal: Abdom Radiol (NY) Date: 2021-03-31
Authors: Emily Harding-Theobald; Jeremy Louissaint; Bharat Maraj; Edward Cuaresma; Whitney Townsend; Mishal Mendiratta-Lala; Amit G Singal; Grace L Su; Anna S Lok; Neehar D Parikh Journal: Aliment Pharmacol Ther Date: 2021-08-12 Impact factor: 9.524
Authors: Taiga Wakabayashi; Farid Ouhmich; Cristians Gonzalez-Cabrera; Emanuele Felli; Antonio Saviano; Vincent Agnus; Peter Savadjiev; Thomas F Baumert; Patrick Pessaux; Jacques Marescaux; Benoit Gallix Journal: Hepatol Int Date: 2019-08-31 Impact factor: 9.029