Purpose To develop a new technique that enables simultaneous quantification of whole‐brain T1, T2, T2∗, as well as susceptibility and synthesis of six contrast‐weighted images in a single 9.1‐minute scan. Methods The technique uses hybrid T2‐prepared inversion‐recovery pulse modules and multi‐echo gradient‐echo readouts to collect k‐space data with various T1, T2, and T2∗ weightings. The underlying image is represented as a six‐dimensional low‐rank tensor consisting of three spatial dimensions and three temporal dimensions corresponding to T1 recovery, T2 decay, and multi‐echo behaviors, respectively. Multiparametric maps were fitted from reconstructed image series. The proposed method was validated on phantoms and healthy volunteers, by comparing quantitative measurements against corresponding reference methods. The feasibility of generating six contrast‐weighted images was also examined. Results High quality, co‐registered T1, T2, and T2∗ susceptibility maps were generated that closely resembled the reference maps. Phantom measurements showed substantial consistency (R2 > 0.98) with the reference measurements. Despite the significant differences of T1 (p < .001), T2 (p = .002), and T2∗ (p = 0.008) between our method and the references for in vivo studies, excellent agreement was achieved with all intraclass correlation coefficients greater than 0.75. No significant difference was found for susceptibility (p = .900). The framework is also capable of synthesizing six contrast‐weighted images. Conclusion The MR Multitasking–based 3D brain mapping of T1, T2, T2∗, and susceptibility agrees well with the reference and is a promising technique for multicontrast and quantitative imaging.

Early detection is important in glaucoma management. By using optical coherence tomography (OCT), the subtle structural changes caused by glaucoma can be detected. Though OCT provided abundant parameters for comprehensive information, clinicians may be confused once the results conflict. Machine learning classifiers (MLCs) are good tools for considering numerous parameters and generating reliable diagnoses in glaucoma practice. Here we aim to compare different MLCs based on Spectralis OCT parameters, including circumpapillary retinal nerve fiber layer (cRNFL) thickness, Bruch’s membrane opening-minimum rim width (BMO-MRW), Early Treatment Diabetes Retinopathy Study (ETDRS) macular thickness, and posterior pole asymmetry analysis (PPAA), in discriminating normal from glaucomatous eyes. Five MLCs were proposed, namely conditional inference trees (CIT), logistic model tree (LMT), C5.0 decision tree, random forest (RF), and extreme gradient boosting (XGBoost). Logistic regression (LGR) was used as a benchmark for comparison. RF was shown to be the best model. Ganglion cell layer measurements were the most important predictors in early glaucoma detection and cRNFL measurements were more important as the glaucoma severity increased. The global, temporal, inferior, superotemporal, and inferotemporal sites were relatively influential locations among all parameters. Clinicians should cautiously integrate the Spectralis OCT results into the entire clinical picture when diagnosing glaucoma.

No abstract