ABSTRACT
OBJECTIVE: To review the literature regarding the current state and clinical applicability of machine learning (ML) models in prognosticating the outcomes of patients with mild traumatic brain injury (mTBI) in the early clinical presentation. DESIGN: Databases were searched for studies including ML and mTBI from inception to March 10, 2023. Included studies had a primary outcome of predicting post-mTBI prognosis or sequalae. The Prediction model study Risk of Bias for Predictive Models assessment tool (PROBAST) was used for assessing the risk of bias and applicability of included studies. RESULTS: Out of 1235 articles, 10 met the inclusion criteria, including data from 127,929 patients. The most frequently used modeling techniques were Support Vector Machine (SVM) and Artificial Neural Network (NN) and Area Under the Curve (AUC) ranged from 0.66-0.889. Despite promise, several limitations to studies exist such as low sample sizes, database restrictions, inconsistencies in patient presentation definitions and lack of comparison to traditional clinical judgment or tools. CONCLUSION: ML models show potential in early stage mTBI prognostication, but to achieve widespread adoption, future clinical studies prognosticating mTBI using ML need to reduce bias, provide clarity and consistency in defining patient populations targeted, and validate against established benchmarks.
ABSTRACT
PURPOSE: To analyze the effectiveness of image-guided energy ablation techniques with and without concurrent therapies in providing palliative pain relief in patients with bone metastases. MATERIALS AND METHODS: Ovid Embase, Ovid Medline, and Pubmed were searched from inception to April 14, 2023, using search terms related to bone lesions and MeSH terms regarding ablation therapy. English peer-reviewed primary articles were included that reported pain scores following image-guided energy-based ablation of bone metastases. Exclusion criteria included nonpalliative treatment, pain scores associated with specific treatment modalities not reported, and nonmetastatic bone lesions. Mean percentage reduction in pain score was calculated. RESULTS: Of the 1,396 studies screened, 54 were included. All but 1 study demonstrated decreased pain scores at final follow-up. Mean reductions in pain scores at final follow-up were 49% for radiofrequency (RF) ablation, 58% for RF ablation and adjunct, 54% for cryoablation (CA), 72% for cryoablation and adjunct (CA-A), 48% for microwave ablation (MWA), 81% for microwave ablation and adjunct (MWA-A), and 64% for high-intensity focused ultrasound (US). Postprocedural adverse event rates were 4.9% for RF ablation, 34.8% for RF ablation and adjunct, 9.6% for CA, 12.0% for CA-A, 48.9% for MWA, 33.5% for MWA-A, and 17.0% for high-intensity focused US. CONCLUSIONS: Image-guided energy ablation demonstrated consistently strong reduction in pain across all modalities, with variable postprocedural adverse event rates. Owing to heterogeneity of included studies, quantitative analysis was not appropriate. Future primary research should focus on creating consistent prospective studies with established statistical power, explicit documentation, and comparison with other techniques.
