METHODS: A systematic literature search of brain tumours in the context of fMRI methods applied to pre-operative mapping for language functional areas was conducted using PubMed/MEDLINE and Scopus electronic database following PRISMA guidelines. The article search was conducted between the earliest record and March 1, 2019. References and citations were checked in Google Scholar database.
RESULTS: Twenty-nine independent studies were identified, comprising 1031 adult participants with 976 patients characterised with different types and sizes of brain tumours, and the remaining 55 being healthy controls. These studies evaluated functional language areas in patients with brain tumours prior to surgical interventions using language-based fMRI. Results demonstrated that 86% of the studies used a Word Generation Task (WGT) to evoke functional language areas during pre-operative mapping. Fifty-seven percent of the studies that used language-based paradigms in conjunction with fMRI as a pre-operative mapping tool were in agreement with intra-operative results of language localization.
CONCLUSIONS: WGT was most commonly utilised and is proposed as a suitable and useful technique for a language-based paradigm fMRI for pre-operative mapping. However, based on available evidence, WGT alone is not sufficient. We propose a combination and convergence paradigms for a more sensitive and specific map of language function for pre-operative mapping. A standard guideline for clinical applications should be established.
METHODS: A systematic search was performed in the PubMed, Scopus, and Web of Science (WoS) databases in June 2022. Patients with head and neck cancer treated with radiotherapy and periodic rs-fMRI assessments were included. A meta-analysis was performed to determine the potential of rs-fMRI for detecting brain changes.
RESULTS: Ten studies with a total of 513 subjects (head and neck cancer patients, n = 437; healthy controls, n = 76) were included. A significance of rs-fMRI for detecting brain changes in the temporal and frontal lobes, cingulate cortex, and cuneus was demonstrated in most studies. These changes were reported to be associated with dose (6/10 studies) and latency (4/10 studies). A strong effect size (r = 0.71, p
PURPOSE: To demonstrate automatic detection of BM on three MRI datasets using a deep learning-based approach. To improve the performance of the network is iteratively co-trained with datasets from different domains. A systematic approach is proposed to prevent catastrophic forgetting during co-training.
STUDY TYPE: Retrospective.
POPULATION: A total of 156 patients (105 ground truth and 51 pseudo labels) with 1502 BM (BrainMetShare); 121 patients with 722 BM (local); 400 patients with 447 primary gliomas (BrATS). Training/pseudo labels/validation data were distributed 84/51/21 (BrainMetShare). Training/validation data were split: 121/23 (local) and 375/25 (BrATS).
FIELD STRENGTH/SEQUENCE: A 5 T and 3 T/T1 spin-echo postcontrast (T1-gradient echo) (BrainMetShare), 3 T/T1 magnetization prepared rapid acquisition gradient echo postcontrast (T1-MPRAGE) (local), 0.5 T, 1 T, and 1.16 T/T1-weighted-fluid-attenuated inversion recovery (T1-FLAIR) (BrATS).
ASSESSMENT: The ground truth was manually segmented by two (BrainMetShare) and four (BrATS) radiologists and manually annotated by one (local) radiologist. Confidence and volume based domain adaptation (CAVEAT) method of co-training the three datasets on a 3D nonlocal convolutional neural network (CNN) architecture was implemented to detect BM.
STATISTICAL TESTS: The performance was evaluated using sensitivity and false positive rates per patient (FP/patient) and free receiver operating characteristic (FROC) analysis at seven predefined (1/8, 1/4, 1/2, 1, 2, 4, and 8) FPs per scan.
RESULTS: The sensitivity and FP/patient from a held-out set registered 0.811 at 2.952 FP/patient (BrainMetShare), 0.74 at 3.130 (local), and 0.723 at 2.240 (BrATS) using the CAVEAT approach with lesions as small as 1 mm being detected.
DATA CONCLUSION: Improved sensitivities at lower FP can be achieved by co-training datasets via the CAVEAT paradigm to address the problem of data sparsity.
LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 2.