BACKGROUND Molecular characterization of glioma has implications for prognosis, treatment planning, and prediction of treatment response. Current histopathology is limited by intratumoral heterogeneity and variability in detection methods. Advances in computational techniques have led to interest in mining quantitative imaging features to noninvasively detect genetic mutations. OBJECTIVE To evaluate the diagnostic accuracy of machine learning (ML) models in molecular subtyping gliomas on preoperative magnetic resonance imaging (MRI). METHODS A systematic search was performed following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) guidelines to identify studies up to April 1, 2020. Methodological quality of studies was assessed using the Quality Assessment for Diagnostic Accuracy Studies (QUADAS)-2. Diagnostic performance estimates were obtained using a bivariate model and heterogeneity was explored using metaregression. RESULTS Forty-four original articles were included. The pooled sensitivity and specificity for predicting isocitrate dehydrogenase (IDH) mutation in training datasets were 0.88 (95% CI 0.83-0.91) and 0.86 (95% CI 0.79-0.91), respectively, and 0.83 to 0.85 in validation sets. Use of data augmentation and MRI sequence type were weakly associated with heterogeneity. Both O6-methylguanine-DNA methyltransferase (MGMT) gene promoter methylation and 1p/19q codeletion could be predicted with a pooled sensitivity and specificity between 0.76 and 0.83 in training datasets. CONCLUSION ML application to preoperative MRI demonstrated promising results for predicting IDH mutation, MGMT methylation, and 1p/19q codeletion in glioma. Optimized ML models could lead to a noninvasive, objective tool that captures molecular information important for clinical decision making. Future studies should use multicenter data, external validation and investigate clinical feasibility of ML models.

No abstract

Background There has been increasing focus to improve the quality of recovery following anterior cervical spine surgery (ACSS). Postoperative pain and nausea are the most common reasons for prolonged hospital stay and readmission after ACSS. Superficial cervical plexus block (SCPB) provides site-specific analgesia with minimal side effects, thereby improving the quality of recovery. The aim of our study was to investigate the effect bilateral cervical plexus block has on postoperative recovery in patients undergoing ACSS. Methods The study is a pragmatic, multi-centre, blinded, parallel-group, randomised placebo-controlled trial. 136 eligible patients (68 in each group) undergoing ACSS will be included. Patients randomised to the intervention group will have a SCPB administered under ultrasound guidance with a local anaesthetic solution (0.2% ropivacaine, 15mL); patients randomised to the placebo group will be injected in an identical manner with a saline solution. The primary outcome is the 40-item quality of recovery questionnaire score at 24 h after surgery. In addition, comparisons between groups will be made for a 24-h opioid usage and length of hospital stay. Neck pain intensity will be quantified using the numeric rating scale at 1, 3, 6 and at 24 h postoperatively. Incidence of nausea, vomiting, dysphagia or hoarseness in the first 24 h after surgery will also be measured. Discussion By conducting a blinded placebo trial, we aim to control for the bias inherently associated with a tangible medical intervention and show the true treatment effect of SCPB in ACSS. A statistically significant result will indicate an overall improved quality of recovery for patients; alternatively, if no benefit is shown, this trial will provide evidence that this intervention is unnecessary. Trial registration ClinicalTrials.gov ACTRN12619000028101. Prospectively registered on 11 January 2019 with Australia New Zealand Clinical Trials Registry