METHODS: Knee image is first oversegmented to produce homogeneous superpixels. Then, a ranking model is developed to rank the superpixels according to their affinities to standard priors, wherein background superpixels would have lower ranking values. Finally, seed labels are generated on the background superpixel using Fuzzy C-Means method.
RESULTS: SAGE has achieved better interobserver DSCs of 0.94 ± 0.029 and 0.93 ± 0.035 in healthy and OA knee segmentation, respectively. Good segmentation performance has been reported in femoral (Healthy: 0.94 ± 0.036 and OA: 0.93 ± 0.034), tibial (Healthy: 0.91 ± 0.079 and OA: 0.88 ± 0.095) and patellar (Healthy: 0.88 ± 0.10 and OA: 0.84 ± 0.094) cartilage segmentation. Besides, SAGE has demonstrated greater mean readers' time of 80 ± 19 s and 80 ± 27 s in healthy and OA knee segmentation, respectively.
CONCLUSIONS: SAGE enhances the efficiency of segmentation process and attains satisfactory segmentation performance compared to manual and random walks segmentation. Future works should validate SAGE on progressive image data cohort using OA biomarkers.
METHODS: Sixty-six women who underwent HIFU treatment were divided into groups A (NPVr ≥90%; n = 26) and B (NPVr <90%, n = 40). Multivariate logistic regression analyses of MRI features were conducted to identify the potential predictors of an NPVr ≥90%.
RESULTS: Generalized estimating equation (GEE) analysis was used to model the prediction of an NPVr ≥90% with four significant predictors from multivariate analyses: the thickness of the subcutaneous fat layer, adenomyosis volume, T2 signal intensity (SI) ratio of adenomyosis to myometrium, and the Ktrans ratio of adenomyosis to myometrium. Clinical efficacy was significantly greater in group A than in group B. The findings showed no serious AEs and no significant differences between AMH concentrations before and 6 months after treatment.
CONCLUSIONS: The present retrospective study demonstrated that achievement of NPVr ≥90% as a measure of clinical treatment success in MRI-guided HIFU treatment of adenomyosis using multivariate analyses and a prediction model is clinically possible without compromising the safety of patients.
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.
METHODS: We developed a hybrid algorithm that combines features of empirical mode decomposition (EMD) with principal component analysis (PCA) to reduce the BCG artefact. The algorithm does not require extra electrocardiogram (ECG) or electrooculogram (EOG) recordings to extract the BCG artefact.
RESULTS: The method was tested with both simulated and real EEG data of 11 participants. From the simulated data, the similarity index between the extracted BCG and the simulated BCG showed the effectiveness of the proposed method in BCG removal. On the other hand, real data were recorded with two conditions, i.e. resting state (eyes closed dataset) and task influenced (event-related potentials (ERPs) dataset). Using qualitative (visual inspection) and quantitative (similarity index, improved normalized power spectrum (INPS) ratio, power spectrum, sample entropy (SE)) evaluation parameters, the assessment results showed that the proposed method can efficiently reduce the BCG artefact while preserving the neuronal signals.
COMPARISON WITH EXISTING METHODS: Compared with conventional methods, namely, average artefact subtraction (AAS), optimal basis set (OBS) and combined independent component analysis and principal component analysis (ICA-PCA), the statistical analyses of the results showed that the proposed method has better performance, and the differences were significant for all quantitative parameters except for the power and sample entropy.
CONCLUSIONS: The proposed method does not require any reference signal, prior information or assumption to extract the BCG artefact. It will be very useful in circumstances where the reference signal is not available.