Huihui Xie1, Shuai Ma1, Xiaoying Wang1, Xiaodong Zhang2. 1. Department of Radiology, Peking University First Hospital, No. 8, Xishiku Street, Xicheng District, Beijing, 100034, China. 2. Department of Radiology, Peking University First Hospital, No. 8, Xishiku Street, Xicheng District, Beijing, 100034, China. zhxd2009@gmail.com.
Abstract
OBJECTIVES: To develop a radiomics model for predicting hematoma expansion in patients with intracerebral hemorrhage (ICH) and to compare its predictive performance with a conventional radiological feature-based model. METHODS: We retrospectively analyzed 251 consecutive patients with acute ICH. Two radiologists independently assessed baseline noncontrast computed tomography (NCCT) images. For each radiologist, a radiological model was constructed from radiological variables; a radiomics score model was constructed from high-dimensional quantitative features extracted from NCCT images; and a combined model was constructed using both radiological variables and radiomics score. Development of models was constructed in a primary cohort (n = 177). We then validated the results in an independent validation cohort (n = 74). The primary outcome was hematoma expansion. We compared the three models for predicting hematoma expansion. Predictive performance was assessed with the receiver operating characteristic (ROC) curve analysis. RESULTS: In the primary cohort, combined model and radiomics model showed greater AUCs than radiological model for both readers (all p < .05). In the validation cohort, combined model and radiomics model showed greater AUCs, sensitivities, and accuracies than radiological model for reader 2 (all p < .05). Combined model showed greater AUC than radiomics model for reader 1 only in the primary cohort (p = .03). Performance of three models was comparable between reader 1 and reader 2 in both cohorts (all p > .05). CONCLUSIONS: NCCT-based radiomics model showed high predictive performance and outperformed radiological model in the prediction of early hematoma expansion in ICH patients. KEY POINTS: • Radiomics model showed better performance for prediction of hematoma expansion in patients with intracerebral hemorrhage than radiological feature-based model. • Hematomas which expanded in follow-up NCCT tended to be larger in baseline volume, more irregular in shape, more heterogeneous in composition, and coarser in texture. • A radiomics model provides a convenient and objective tool for prediction of hematoma expansion that helps to define subsets of patients who would benefit from anti-expansion therapy.
OBJECTIVES: To develop a radiomics model for predicting hematoma expansion in patients with intracerebral hemorrhage (ICH) and to compare its predictive performance with a conventional radiological feature-based model. METHODS: We retrospectively analyzed 251 consecutive patients with acute ICH. Two radiologists independently assessed baseline noncontrast computed tomography (NCCT) images. For each radiologist, a radiological model was constructed from radiological variables; a radiomics score model was constructed from high-dimensional quantitative features extracted from NCCT images; and a combined model was constructed using both radiological variables and radiomics score. Development of models was constructed in a primary cohort (n = 177). We then validated the results in an independent validation cohort (n = 74). The primary outcome was hematoma expansion. We compared the three models for predicting hematoma expansion. Predictive performance was assessed with the receiver operating characteristic (ROC) curve analysis. RESULTS: In the primary cohort, combined model and radiomics model showed greater AUCs than radiological model for both readers (all p < .05). In the validation cohort, combined model and radiomics model showed greater AUCs, sensitivities, and accuracies than radiological model for reader 2 (all p < .05). Combined model showed greater AUC than radiomics model for reader 1 only in the primary cohort (p = .03). Performance of three models was comparable between reader 1 and reader 2 in both cohorts (all p > .05). CONCLUSIONS: NCCT-based radiomics model showed high predictive performance and outperformed radiological model in the prediction of early hematoma expansion in ICHpatients. KEY POINTS: • Radiomics model showed better performance for prediction of hematoma expansion in patients with intracerebral hemorrhage than radiological feature-based model. • Hematomas which expanded in follow-up NCCT tended to be larger in baseline volume, more irregular in shape, more heterogeneous in composition, and coarser in texture. • A radiomics model provides a convenient and objective tool for prediction of hematoma expansion that helps to define subsets of patients who would benefit from anti-expansion therapy.
Authors: Dar Dowlatshahi; Andrea Morotti; Fahad S Al-Ajlan; Gregoire Boulouis; Andrew D Warren; William Petrcich; Richard I Aviv; Andrew M Demchuk; Joshua N Goldstein Journal: Stroke Date: 2019-05 Impact factor: 7.914
Authors: Charlotte Jj van Asch; Merel Ja Luitse; Gabriël Je Rinkel; Ingeborg van der Tweel; Ale Algra; Catharina Jm Klijn Journal: Lancet Neurol Date: 2010-01-05 Impact factor: 44.182
Authors: Dylan Blacquiere; Andrew M Demchuk; Mohammed Al-Hazzaa; Anirudda Deshpande; William Petrcich; Richard I Aviv; David Rodriguez-Luna; Carlos A Molina; Yolanda Silva Blas; Imanuel Dzialowski; Anna Czlonkowska; Jean-Martin Boulanger; Cheemun Lum; Gord Gubitz; Vasantha Padma; Jayanta Roy; Carlos S Kase; Rohit Bhatia; Michael D Hill; Dar Dowlatshahi Journal: Stroke Date: 2015-10-08 Impact factor: 7.914
Authors: C D Barras; B M Tress; S Christensen; M Collins; P M Desmond; B E Skolnick; S A Mayer; S M Davis Journal: AJNR Am J Neuroradiol Date: 2013-01-10 Impact factor: 3.825
Authors: Xia Wang; Hisatomi Arima; Rustam Al-Shahi Salman; Mark Woodward; Emma Heeley; Christian Stapf; Pablo M Lavados; Thompson Robinson; Yining Huang; Jiguang Wang; Candice Delcourt; Craig S Anderson Journal: Stroke Date: 2014-12-11 Impact factor: 7.914