Fibre tract segmentation for intraoperative diffusion MRI in neurosurgical patients using tract-specific orientation atlas and tumour deformation modelling.

PURPOSE: Intraoperative diffusion MRI could provide a means of visualising brain fibre tracts near a neurosurgical target after preoperative images have been invalidated by brain shift. We propose an atlas-based intraoperative tract segmentation method, as the standard preoperative method, streamline tractography, is unsuitable for intraoperative implementation. METHODS: A tract-specific voxel-wise fibre orientation atlas is constructed from healthy training data. After registration with a target image, a radial tumour deformation model is applied to the orientation atlas to account for displacement caused by lesions. The final tract map is obtained from the inner product of the atlas and target image fibre orientation data derived from intraoperative diffusion MRI. RESULTS: The simple tumour model takes only seconds to effectively deform the atlas into alignment with the target image. With minimal processing time and operator effort, maps of surgically relevant tracts can be achieved that are visually and qualitatively comparable with results obtained from streamline tractography. CONCLUSION: Preliminary results demonstrate feasibility of intraoperative streamline-free tract segmentation in challenging neurosurgical cases. Demonstrated results in a small number of representative sample subjects are realistic despite the simplicity of the tumour deformation model employed. Following this proof of concept, future studies will focus on achieving robustness in a wide range of tumour types and clinical scenarios, as well as quantitative validation of segmentations.

Combined Denoising and Suppression of Transient Artifacts in Arterial Spin Labeling MRI Using Deep Learning.

BACKGROUND: Arterial spin labeling (ASL) is a useful tool for measuring cerebral blood flow (CBF). However, due to the low signal-to-noise ratio (SNR) of the technique, multiple repetitions are required, which results in prolonged scan times and increased susceptibility to artifacts. PURPOSE: To develop a deep-learning-based algorithm for simultaneous denoising and suppression of transient artifacts in ASL images. STUDY TYPE: Retrospective. SUBJECTS: 131 pediatric neuro-oncology patients for model training and 11 healthy adult subjects for model evaluation. FIELD STRENGTH/SEQUENCE: 3T / pseudo-continuous and pulsed ASL with 3D gradient-and-spin-echo readout. ASSESSMENT: A denoising autoencoder (DAE) model was designed with stacked encoding/decoding convolutional layers. Reference standard images were generated by averaging 10 pairwise ASL subtraction images. The model was trained to produce perfusion images of a similar quality using a single subtraction image. Performance was compared against Gaussian and non-local means (NLM) filters. Evaluation metrics included SNR, peak SNR (PSNR), and structural similarity index (SSIM) of the CBF images, compared to the reference standard. STATISTICAL TESTS: One-way analysis of variance (ANOVA) tests for group comparisons. RESULTS: The DAE model was the only model to produce a significant increase in SNR compared to the raw images (P < 0.05), providing an average SNR gain of 62%. The DAE model was also effective at suppressing transient artifacts, and was the only model to show a significant improvement in accuracy in the generated CBF images, as assessed using PSNR values (P < 0.05). In addition, using data from multiple inflow time acquisitions, the DAE images produced the best fit to the Buxton kinetic model, offering a 75% reduction in the fitting error compared to the raw images. DATA CONCLUSION: Deep-learning-based algorithms provide superior accuracy when denoising ASL images, due to their ability to simultaneously increase SNR and suppress artifactual signals in raw ASL images. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 1.