Maroufi, S. Farzad

OBJECTIVE Epilepsy affects nearly 50 million individuals worldwide, with one-third of cases resistant to antiseizure medications. For these patients, surgical intervention offers a potential path to seizure freedom. While resective surgery has been the gold standard, laser interstitial thermal therapy (LITT) has emerged as a minimally invasive alternative. The aim of this study was to evaluate the efficacy and safety of LITT versus resective surgery in patients with nontumoral epilepsy. METHODS A systematic review and meta-analysis were conducted using PubMed, Embase, and Scopus, including studies comparing seizure freedom rates, complications, and procedural outcomes between LITT and open surgery in nontumoral epilepsy. Eleven studies met the inclusion criteria, comprising 389 LITT and 557 open surgery patients with varying epilepsy etiologies, including temporal lobe epilepsy, focal cortical dysplasia, and tuberous sclerosis. Statistical analysis was performed using a random-effects model to assess seizure freedom, complications, and reoperation rates. RESULTS Open surgery demonstrated higher rates of complete seizure freedom, although not reaching significance (68.1% vs 53.7%, RR 0.81, p = 0.07). This outcome was sensitive to influential analysis and reached significance in the epileptogenic zone–directed resection subgroup analysis. Although adequate seizure freedom was comparable between the groups (LITT: 63.0% vs open: 74.0%, RR 0.90, p = 0.11), the open surgery group had higher rates of control in the pediatric and non–temporal lobe epilepsy subgroups. Complication rates were significantly higher in the open surgery group (30.0% vs 18.3%, RR 0.55, p < 0.01). LITT patients had significantly shorter hospital stays (3.4 vs 6.8 days, standardized mean difference −0.93, p < 0.01). Reoperation rates were comparable between groups (13.1% for LITT vs 13.4%, RR 1.59, p = 0.26). CONCLUSIONS While LITT offers a less invasive approach with reduced hospitalization and morbidity, open surgery remains slightly superior in achieving long-term seizure freedom. Patient selection remains critical, and further studies are needed to refine decision-making criteria based on epilepsy subtype and lesion characteristics.

BACKGROUND AND OBJECTIVES: Aneurysm risk prediction remains an imprecise science that places patients at risk for either over or undertreatment. Machine learning (ML) models may improve clinical practice by adding precision to risk assessment. This study aims to comprehensively assess the current landscape of ML applications in predicting the risk of aneurysm rupture and compare the performance with the widely used PHASES score. METHODS: A systematic review of PubMed, Scopus, and Web of Science was conducted. All studies using ML tools to predict the rupture risk of intracranial aneurysms were included. Meta-analysis was conducted with consideration to the ML algorithms and compared with the PHASES score. RESULTS: Thirty-six studies involving 22 462 patients were included in the final analysis. ML techniques, including 124 models using 25 algorithms, were employed. Among various ML models, while they had comparable diagnostic performance, deep learning exhibited a slightly better performance profile (sensitivity = 0.792, specificity = 0.788, and accuracy = 0.778 in external validation). Based on our analysis, ML, regardless of the algorithm, provides comparable sensitivity (0.743 vs 0.771, P = .60) and higher specificity (0.763 vs 0.507, P < .01) compared with the PHASES score. Consistently, pooling the area under the receiver operating characteristic curve (AUC) for 60 ML models and 5 PHASES score data, ML models exhibited higher AUC (0.84 vs 0.64, P < .01). Using hemodynamic parameters as input for models improved specificity ( P < .01) in the test sets without any significant changes in the sensitivity. The later improvement was not observed in the external validation sets. CONCLUSION: ML techniques have the potential to enhance the prediction of intracranial aneurysm rupture compared with traditional approaches, like the PHASES score. Incorporating hemodynamic parameters may further enhance the accuracy of ML models. Feature prospective studies are required to validate the utility of ML models for clinical integration.