Background: Intracerebral hemorrhages (ICH) and perihematomal edema (PHE) are respective imaging markers of primary and secondary brain injury in hemorrhagic stroke. In this study, we explored the potential added value of PHE radiomic features for prognostication in ICH patients. Methods: Using a multicentric trial cohort of acute supratentorial ICH (n = 852) patients, we extracted radiomic features from ICH and PHE lesions on admission non-contrast head CTs. We trained and tested combinations of different machine learning classifiers and feature selection methods for prediction of poor outcome—defined by 4-to-6 modified Rankin Scale scores at 3-month follow-up—using five different input strategies: (a) ICH radiomics, (b) ICH and PHE radiomics, (c) admission clinical predictors of poor outcomes, (d) ICH radiomics and clinical variables, and (e) ICH and PHE radiomics with clinical variables. Models were trained on 500 patients, tested, and compared in 352 using the receiver operating characteristics Area Under the Curve (AUC), Integrated Discrimination Index (IDI), and Net Reclassification Index (NRI). Results: Comparing the best performing models in the independent test cohort, both IDI and NRI demonstrated better individual-level risk assessment by addition of PHE radiomics as input to ICH radiomics (both p < 0.001), but with insignificant improvement in outcome prediction (AUC of 0.74 versus 0.71, p = 0.157). The addition of ICH and PHE radiomics to clinical variables also improved IDI and NRI risk-classification (both p < 0.001), but with a insignificant increase in AUC of 0.85 versus 0.83 (p = 0.118), respectively. All machine learning models had greater or equal accuracy in outcome prediction compared to the widely used ICH score. Conclusions: The addition of PHE radiomics to hemorrhage lesion radiomics, as well as radiomics to clinical risk factors, can improve individual-level risk assessment, albeit with an insignificant increase in prognostic accuracy. Machine learning models offer quantitative and immediate risk stratification—on par with or more accurate than the ICH score—which can potentially guide patients’ selection for interventions such as hematoma evacuation.

Purpose HPV-positive and HPV-negative OPSCC are biologically distinct entities, with different prognosis and divergent AJCC/UICC staging schemes. Radiomics refers to automated extraction of shape, intensity and texture features from target lesions on medical images – inaccessible to visual interpretation. We applied machine learning classifiers to devise radiomics signatures to determine the HPV-status in OPSCC. Methods Imaging data was retrieved form The Cancer Imaging Archive and our institutional archives. Patients with OPSCC, known HPV/p16-status, and pre-treatment FDG‑PET / non‑contrast CT were included. The primary tumors were delineated on PET and CT scans. 1040 radiomics features describing texture, shape and signal intensity characteristics were extracted from each tumor and per imaging modality. For HPV-status prediction, LASSO regression feature selection (LASSO) and random forest (RF) machine learning classifiers were applied in 10-fold cross validation, repeated 10x. The area under the receiver operating characteristic curve (AUC-ROC) averaged across validation folds is reported. Conclusions A total of 244 HPV-positive and 82 HPV-negative OPSCC patients were included: 46 T1, 119 T2, 107 T3, and 54 T4 UICC/AJCC stage primary tumors were analyzed. The LASSO / RF machine learning algorithm achieved an averaged AUC-ROC of 0.79 (PET/CT), 0.74 (CT) and 0.70 (PET). Conclusion Radiomics feature extraction from PET and CT scans, combined with machine learning classifiers can generate imaging biomarkers for HPV in OPSCC primary tumors, which may aid in HPV-classification if standard immunohistochemical staining is equivocal, or supplement the immunohistochemical tests in subjects requiring second-line testing. Poster-PDF A-1164.PDF Publication Date: 10 June 2020 (online) © . © Georg Thieme Verlag KG Stuttgart · New York