Accuracy of artificial intelligence for the detection of intracranial hemorrhage and chronic cerebral microbleeds: a systematic review and pooled analysis
Matsoukas S, Scaggiante J, Schuldt BR, Smith CJ, Chennareddy S, Kalagara R, Majidi S, Bederson JB, Fifi JT, Mocco J, et al
La Radiologia medica. 2022
BACKGROUND Artificial intelligence (AI)-driven software has been developed and become commercially available within the past few years for the detection of intracranial hemorrhage (ICH) and chronic cerebral microbleeds (CMBs). However, there is currently no systematic review that summarizes all of these tools or provides pooled estimates of their performance. METHODS In this PROSPERO-registered, PRISMA compliant systematic review, we sought to compile and review all MEDLINE and EMBASE published studies that have developed and/or tested AI algorithms for ICH detection on non-contrast CT scans (NCCTs) or MRI scans and CMBs detection on MRI scans. RESULTS In total, 40 studies described AI algorithms for ICH detection in NCCTs/MRIs and 19 for CMBs detection in MRIs. The overall sensitivity, specificity, and accuracy were 92.06%, 93.54%, and 93.46%, respectively, for ICH detection and 91.6%, 93.9%, and 92.7% for CMBs detection. Some of the challenges encountered in the development of these algorithms include the laborious work of creating large, labeled and balanced datasets, the volumetric nature of the imaging examinations, the fine tuning of the algorithms, and the reduction in false positives. CONCLUSIONS Numerous AI-driven software tools have been developed over the last decade. On average, they are characterized by high performance and expert-level accuracy for the diagnosis of ICH and CMBs. As a result, implementing these tools in clinical practice may improve workflow and act as a failsafe for the detection of such lesions. REGISTRATION-URL: https://www.crd.york.ac.uk/prospero/ Unique Identifier: CRD42021246848.
Minimally Invasive Surgery for Intracerebral Hemorrhage
Scaggiante J, Zhang X, Mocco J, Kellner CP
Background and Purpose- Minimally invasive surgery (MIS) for intracerebral hemorrhage (ICH) has been evaluated in numerous clinical trials. Although meta-analyses for this strategy have been performed in the past, recent trials add important information to results of the comparison and permit strategy-specific analyses, including evaluation of endoscopic evacuation and stereotactic thrombolysis. Methods- Major scientific databases including but not limited to Pubmed, the CENTRAL (Cochrane Central Register of Controlled Trials), Embase, Web of Science, Scopus, the ICTRP (International Clinical Trials Registry Platform), the Internet Stroke Center, and the CNKI (Chinese National Knowledge Infrastructure) were searched in October of 2017 for randomized controlled trials of MIS treatment of supratentorial spontaneous ICH. The primary outcome was defined as death or dependence at the end of follow-up, and the secondary outcome was defined as death. Results- The initial search yielded 958 reports, which were reduced to 15 high-quality randomized controlled trials involving 2152 patients. We analyzed odds ratios for MIS overall, endoscopic surgery, and stereotactic thrombolysis compared with conventional treatment, including medical treatment and conventional craniotomy. The odds ratio and CIs of the primary and secondary outcomes were 0.46 (0.36-0.57) and 0.59 (0.45-0.76) for MIS versus conventional treatment; 0.40 (0.25-0.66) and 0.37 (0.20-0.67) for endoscopic surgery versus conventional treatment; 0.47 (0.34-0.65) and 0.76 (0.56-1.04) for stereotactic thrombolysis versus conventional treatment; and 0.44 (0.29-0.67) and 0.56 (0.37-0.84) for MIS versus craniotomy. We also conducted subgroup analyses focusing on time to evacuation for MIS versus conventional treatment and found 0.36 (0.22-0.59) and 0.59 (0.34-1.00) for evacuations performed within 24 hours and 0.49 (0.38-0.63) and 0.57 (0.43-0.76) for evacuations performed within 72 hours. Conclusions- This meta-analysis demonstrates that select patients with supratentorial ICH benefit from MIS over other treatments. This beneficial effect remains true when analyzing specific techniques and evacuation timing subgroups.