Quality Assessment of Radiomics Studies on Functional Outcomes After Acute Ischemic Stroke-A Systematic Review.

OBJECTIVE: Radiomics is a machine-learning method that extracts features from medical images. The objective of the present systematic review was to assess the quality of existing studies that use radiomics methods to predict functional outcomes in patients after acute ischemic stroke. METHODS: Studies using radiomics-extracted features to predict functional outcomes among patients with acute ischemic stroke using the modified Rankin Scale were included. PubMed, Scopus, Web of Science, and Embase were screened using the terms "radiomics" and "texture" in combination with "stroke." Quality scores were calculated based on Radiomics Quality Score, the IBSI (Image Biomarkers Standardization Initiative), and the QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2). RESULTS: Fourteen studies were included. The median total Radiomics Quality Score was 14.5 (13-16) out of 36. Domains 1, 5, and 6 on protocol quality and stability of imaging and segmentation, level of evidence, and use of open science and data, respectively, were poor. Median IBSI score was 2.5 (1-5) out of 6. Few studies included bias-field correction algorithms, isovoxel resampling, skull stripping, or gray-level discretization. Of 14 studies, none received +6 points, 1 received +5 points, 5 received +4 points, 1 study received +3 points, 5 received +2 points, 2 received +1 points, and none received 0 points. As per the QUADAS-2, 6/14 (42.9%) studies had a risk of bias concern and 0/14 (0%) had applicability concern. CONCLUSIONS: The quality of the included studies was low to moderate. With increasing use of radiomics, future studies should attempt to adhere to and report established radiomics quality guidelines.

Machine learning can accurately predict risk factors for all-cause reoperation after ACLR: creating a clinical tool to improve patient counseling and outcomes.

PURPOSE: Identifying predictive factors for all-cause reoperation after anterior cruciate ligament reconstruction could inform clinical decision making and improve risk mitigation. The primary purposes of this study are to (1) determine the incidence of all-cause reoperation after anterior cruciate ligament reconstruction, (2) identify predictors of reoperation after anterior cruciate ligament reconstruction using machine learning methodology, and (3) compare the predictive capacity of the machine learning methods to that of traditional logistic regression. METHODS: A longitudinal geographical database was utilized to identify patients with a diagnosis of new anterior cruciate ligament injury. Eight machine learning models were appraised on their ability to predict all-cause reoperation after anterior cruciate ligament reconstruction. Model performance was evaluated via area under the receiver operating characteristics curve. To explore modeling interpretability and radiomic feature influence on the predictions, we utilized a game-theory-based method through SHapley Additive exPlanations. RESULTS: A total of 1400 patients underwent anterior cruciate ligament reconstruction with a mean postoperative follow-up of 9 years. Two-hundred and eighteen (16%) patients experienced a reoperation after anterior cruciate ligament reconstruction, of which 6% of these were revision ACL reconstruction. SHapley Additive exPlanations plots identified the following risk factors as predictive for all-cause reoperation: diagnosis of systemic inflammatory disease, distal tear location, concomitant medial collateral ligament repair, higher visual analog scale pain score prior to surgery, hamstring autograft, tibial fixation via radial expansion device, younger age at initial injury, and concomitant meniscal repair. Pertinent negatives, when compared to previous studies, included sex and timing of surgery. XGBoost was the best-performing model (area under the receiver operating characteristics curve of 0.77) and outperformed logistic regression in this regard. CONCLUSIONS: All-cause reoperation after anterior cruciate ligament reconstruction occurred at a rate of 16%. Machine learning models outperformed traditional statistics and identified diagnosis of systemic inflammatory disease, distal tear location, concomitant medial collateral ligament repair, higher visual analog scale pain score prior to surgery, hamstring autograft, tibial fixation via radial expansion device, younger age at initial injury, and concomitant meniscal repair as predictive risk factors for reoperation. Pertinent negatives, when compared to previous studies, included sex and timing of surgery. These models will allow surgeons to tabulate individualized risk for future reoperation for patients undergoing anterior cruciate ligament reconstruction. LEVEL OF EVIDENCE: III.