Self-supervised multicontrast super-resolution for diffusion-weighted prostate MRI.

PURPOSE: This study addresses the challenge of low resolution and signal-to-noise ratio (SNR) in diffusion-weighted images (DWI), which are pivotal for cancer detection. Traditional methods increase SNR at high b-values through multiple acquisitions, but this results in diminished image resolution due to motion-induced variations. Our research aims to enhance spatial resolution by exploiting the global structure within multicontrast DWI scans and millimetric motion between acquisitions. METHODS: We introduce a novel approach employing a "Perturbation Network" to learn subvoxel-size motions between scans, trained jointly with an implicit neural representation (INR) network. INR encodes the DWI as a continuous volumetric function, treating voxel intensities of low-resolution acquisitions as discrete samples. By evaluating this function with a finer grid, our model predicts higher-resolution signal intensities for intermediate voxel locations. The Perturbation Network's motion-correction efficacy was validated through experiments on biological phantoms and in vivo prostate scans. RESULTS: Quantitative analyses revealed significantly higher structural similarity measures of super-resolution images to ground truth high-resolution images compared to high-order interpolation (p < $$ < $$ 0.005). In blind qualitative experiments, 96 . 1 % $$ 96.1\% $$ of super-resolution images were assessed to have superior diagnostic quality compared to interpolated images. CONCLUSION: High-resolution details in DWI can be obtained without the need for high-resolution training data. One notable advantage of the proposed method is that it does not require a super-resolution training set. This is important in clinical practice because the proposed method can easily be adapted to images with different scanner settings or body parts, whereas the supervised methods do not offer such an option.

Directional and inter-acquisition variability in diffusion-weighted imaging and editing for restricted diffusion.

PURPOSE: To evaluate and quantify inter-directional and inter-acquisition variation in diffusion-weighted imaging (DWI) and emphasize signals that report restricted diffusion to enhance cancer conspicuity, while reducing the effects of local microscopic motion and magnetic field fluctuations. METHODS: Ten patients with biopsy-proven prostate cancer were studied under an Institutional Review Board-approved protocol. Individual acquisitions of DWI signal intensities were reconstructed to calculate inter-acquisition distributions and their statistics, which were compared for healthy versus cancer tissue. A method was proposed to detect and filter the acquisitions affected by motion-induced signal loss. First, signals that reflect restricted diffusion were separated from the acquisitions that suffer from signal loss, likely due to microscopic motion, by imposing a cutoff value. Furthermore, corrected apparent diffusion coefficient maps were calculated by employing a weighted sum of the multiple acquisitions, instead of conventional averaging. These weights were calculated by applying a soft-max function to the set of acquisitions per-voxel, making the analysis immune to acquisitions with significant signal loss, even if the number of such acquisitions is high. RESULTS: Inter-acquisition variation is much larger than the Rician noise variance, local spatial variations, and the estimates of diffusion anisotropy based on the current data, as well as the published values of anisotropy. The proposed method increases the contrast for cancers and yields a sensitivity of 98 . 8 % $$ 98.8\% $$ with a false positive rate of 3 . 9 % $$ 3.9\% $$ . CONCLUSION: Motion-induced signal loss makes conventional signal-averaging suboptimal and can obscure signals from areas with restricted diffusion. Filtering or weighting individual acquisitions prior to image analysis can overcome this problem.