3D Magnetic Resonance Imaging Research Articles

A modular motion compensation pipeline for prospective respiratory motion correction of multi-nuclear MR spectroscopy

Magnetic resonance (MR) acquisitions of the torso are frequently affected by respiratory motion with detrimental effects on signal quality. The motion of organs inside the body is typically decoupled from surface motion and is best captured using rapid MR imaging (MRI). We propose a pipeline for prospective motion correction of the target organ using MR image navigators providing absolute motion estimates in millimeters. Our method is designed to feature multi-nuclear interleaving for non-proton MR acquisitions and to tolerate local transmit coils with inhomogeneous field and sensitivity distributions. OpenCV object tracking was introduced for rapid estimation of in-plane displacements in 2D MR images. A full three-dimensional translation vector was derived by combining displacements from slices of multiple and arbitrary orientations. The pipeline was implemented on 3 T and 7 T MR scanners and tested in phantoms and volunteers. Fast motion handling was achieved with low-resolution 2D MR image navigators and direct implementation of OpenCV into the MR scanner’s reconstruction pipeline. Motion-phantom measurements demonstrate high tracking precision and accuracy with minor processing latency. The feasibility of the pipeline for reliable in-vivo motion extraction was shown on heart and kidney data. Organ motion was manually assessed by independent operators to quantify tracking performance. Object tracking performed convincingly on 7774 navigator images from phantom scans and different organs in volunteers. In particular the kernelized correlation filter (KCF) achieved similar accuracy (74%) as scored from inter-operator comparison (82%) while processing at a rate of over 100 frames per second. We conclude that fast 2D MR navigator images and computer vision object tracking can be used for accurate and rapid prospective motion correction. This and the modular structure of the pipeline allows for the proposed method to be used in imaging of moving organs and in challenging applications like cardiac magnetic resonance spectroscopy (MRS) or magnetic resonance imaging (MRI) guided radiotherapy.

3D‐SRDM: 3D super‐resolution of MRI volumes based on diffusion model

AbstractBrain magnetic resonance imaging (MRI) is crucial for diagnosing and understanding neurological disorders, but inherent limitations hinder the visualization of fine details in brain structures. The emergence of super‐resolution techniques, especially deep‐learning methods, has improved imaging quality of MRI, by increasing MRI spatial resolution. At present, deep‐learning algorithms mostly performed super‐resolution on 2D MRI images. However, considering 3D nature of MRI, 3D models are more suitable for brain MRI super‐resolution. To achieve finer brain structural details, this study proposes a 3D brain MRI super‐resolution method based on diffusion model (3D‐SRDM), which is a fast and easily trainable neural network for the generation of high‐resolution brain MRI images. In our 3D‐SRDM model, the self‐attention module in U‐Net is replaced with 3D spatial attention mechanism. The network structure of 3D‐SRDM is optimized to reduce training parameters. Moreover, accelerated sampling from denoising diffusion implicit model is also incorporated to reduce time consumption. By these optimizations, compared with original diffusion model, the proposed model can achieve about 10‐ and 5‐fold speed increase at 4× and 8× super‐resolution of 3D brain MRI volumes, respectively, almost without affecting image quality. Thus, 3D‐SRDM has potential application value in efficiently generating high‐resolution 3D brain MRI images, thus facilitating the doctors' diagnosis.

Association of uncertain significance genetic variants with myocardial mechanics and morphometrics in patients with nonischemic dilated cardiomyopathy

BackgroundCareful interpretation of the relation between phenotype changes of the heart and gene variants detected in dilated cardiomyopathy (DCM) is important for patient care and monitoring.ObjectiveWe sought to assess the association between cardiac-related genes and whole-heart myocardial mechanics or morphometrics in nonischemic dilated cardiomyopathy (NIDCM).MethodsIt was a prospective study consisting of patients with NIDCM. All patients were referred for genetic testing and a genetic analysis was performed using Illumina NextSeq 550 and a commercial gene capture panel of 233 genes (Systems Genomics, Cardiac-GeneSGKit®). It was analyzed whether there are significant differences in clinical, two-dimensional (2D) echocardiographic, and magnetic resonance imaging (MRI) parameters between patients with the genes variants and those without. 2D echocardiography and MRI were used to analyze myocardial mechanics and morphometrics.ResultsThe study group consisted of 95 patients with NIDCM and the average age was 49.7 ± 10.5. All echocardiographic and MRI parameters of myocardial mechanics (left ventricular ejection fraction 28.4 ± 8.7 and 30.7 ± 11.2, respectively) were reduced and all values of cardiac chambers were increased (left ventricular end-diastolic diameter 64.5 ± 5.9 mm and 69.5 ± 10.7 mm, respectively) in this group. It was noticed that most cases of whole-heart myocardial mechanics and morphometrics differences between patients with and without gene variants were in the genes GATAD1, LOX, RASA1, KRAS, and KRIT1. These genes have not been previously linked to DCM. It has emerged that KRAS and KRIT1 genes were associated with worse whole-heart mechanics and enlargement of all heart chambers. GATAD1, LOX, and RASA1 genes variants showed an association with better cardiac function and morphometrics parameters. It might be that these variants alone do not influence disease development enough to be selective in human evolution.ConclusionsCombined variants in previously unreported genes related to DCM might play a significant role in affecting clinical, morphometrics, or myocardial mechanics parameters.

Automatic segmentation of tumour and organs at risk in 3D MRI for cervical cancer radiation therapy with anatomical variations.

Cervical cancer is a common cancer in women globally, with treatment usually involving radiation therapy (RT). Accurate segmentation for the tumour site and organ-at-risks (OARs) could assist in the reduction of treatment side effects and improve treatment planning efficiency. Cervical cancer Magnetic Resonance Imaging (MRI) segmentation is challenging due to a limited amount of training data available and large inter- and intra- patient shape variation for OARs. The proposed Masked-Net consists of a masked encoder within the 3D U-Net to account for the large shape variation within the dataset, with additional dilated layers added to improve segmentation performance. A new loss function was introduced to consider the bounding box loss during training with the proposed Masked-Net. Transfer learning from a male pelvis MRI data with a similar field of view was included. The approaches were compared to the 3D U-Net which was widely used in MRI image segmentation. The data used consisted of 52 volumes obtained from 23 patients with stage IB to IVB cervical cancer across a maximum of 7 weeks of RT with manually contoured labels including the bladder, cervix, gross tumour volume, uterus and rectum. The model was trained and tested with a 5-fold cross validation. Outcomes were evaluated based on the Dice Similarity Coefficients (DSC), the Hausdorff Distance (HD) and the Mean Surface Distance (MSD). The proposed method accounted for the small dataset, large variations in OAR shape and tumour sizes with an average DSC, HD and MSD for all anatomical structures of 0.790, 30.19mm and 3.15mm respectively.

Imaging segmentation mechanism for rectal tumors using improved U-Net

ObjectiveIn radiation therapy, cancerous region segmentation in magnetic resonance images (MRI) is a critical step. For rectal cancer, the automatic segmentation of rectal tumors from an MRI is a great challenge. There are two main shortcomings in existing deep learning-based methods that lead to incorrect segmentation: 1) there are many organs surrounding the rectum, and the shape of some organs is similar to that of rectal tumors; 2) high-level features extracted by conventional neural networks often do not contain enough high-resolution information. Therefore, an improved U-Net segmentation network based on attention mechanisms is proposed to replace the traditional U-Net network.MethodsThe overall framework of the proposed method is based on traditional U-Net. A ResNeSt module was added to extract the overall features, and a shape module was added after the encoder layer. We then combined the outputs of the shape module and the decoder to obtain the results. Moreover, the model used different types of attention mechanisms, so that the network learned information to improve segmentation accuracy.ResultsWe validated the effectiveness of the proposed method using 3773 2D MRI datasets from 304 patients. The results showed that the proposed method achieved 0.987, 0.946, 0.897, and 0.899 for Dice, MPA, MioU, and FWIoU, respectively; these values are significantly better than those of other existing methods.ConclusionDue to time savings, the proposed method can help radiologists segment rectal tumors effectively and enable them to focus on patients whose cancerous regions are difficult for the network to segment.SignificanceThe proposed method can help doctors segment rectal tumors, thereby ensuring good diagnostic quality and accuracy.

Feature Importance Analysis and Machine Learning for Alzheimer’s Disease Early Detection: Feature Fusion of the Hippocampus, Entorhinal Cortex, and Standardized Uptake Value Ratio

Introduction: Alzheimer’s disease (AD) is a progressive neurological disorder characterized by mild memory loss and ranks as a leading cause of mortality in the USA, accounting for approximately 120,000 deaths per year. It is also the primary form of dementia. Early detection is critical for timely intervention as the neurodegenerative process often starts 15–20 years before cognitive symptoms manifest. This study focuses on determining feature importance in AD classification using fused texture features from 3D magnetic resonance imaging hippocampal and entorhinal cortex and standardized uptake value ratio (SUVR) derived from positron emission tomography (PET) images. Methods: To achieve this objective, we employed four distinct classifiers (Linear Support Vector Classification, Linear Discriminant Analysis, Logistic Regression, and Logistic Regression Classifier with Stochastic Gradient Descent Learning). These classifiers were used to derive both average and top-ranked importance scores for each feature based on their outputs. Our framework is designed to distinguish between two classes, AD-negative (or mild cognitive impairment stable [MCIs]) and AD-positive (or MCI conversion [MCIc]), using a probabilistic neural network classifier and the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Results: The findings from the feature importance highlight the crucial role of the GLCM texture features extracted from the hippocampus and entorhinal cortex, demonstrating their superior performance compared to the volume and SUVR. GLCM texture AD classification achieved approximately 90% sensitivity in identifying MCIc cases while maintaining low false positives (below 30%) when fused with other features. Moreover, the receiver operating characteristic curves validate the GLCMs’ superior performance in distinguishing between MCIs and MCIc. Additionally, fusing different types of features improved classification performance compared to relying solely on any single feature category. Conclusion: Our study emphasizes the pivotal role of GLCM texture features in early Alzheimer’s detection.

MRI super-resolution using similarity distance and multi-scale receptive field based feature fusion GAN and pre-trained slice interpolation network

Challenges arise in achieving high-resolution Magnetic Resonance Imaging (MRI) to improve disease diagnosis accuracy due to limitations in hardware, patient discomfort, long acquisition times, and high costs. While Convolutional Neural Networks (CNNs) have shown promising results in MRI super-resolution, they often don't look into the structural similarity and prior information available in consecutive MRI slices. By leveraging information from sequential slices, more robust features can be obtained, potentially leading to higher-quality MRI slices. We propose a multi-slice two-dimensional (2D) MRI super-resolution network that combines a Generative Adversarial Network (GAN) with feature fusion and a pre-trained slice interpolation network to achieve three-dimensional (3D) super-resolution. The proposed model requires consecutively acquired three low-resolution (LR) MRI slices along a specific axis, and achieves the reconstruction of the MRI slices in the remaining two axes. The network effectively enhances both in-plane and out-of-plane resolution along the sagittal axis while addressing computational and memory constraints in 3D super-resolution. The proposed generator has a in-plane and out-of-plane Attention (IOA) network that fuses both in-plane and out-plane features of MRI dynamically. In terms of out-of-plane attention, the network merges features by considering the similarity distance between features and for in-plane attention, the network employs a two-level pyramid structure with varying receptive fields to extract features at different scales, ensuring the inclusion of both global and local features. Subsequently, to achieve 3D MRI super-resolution, a pre-trained slice interpolation network is used that takes two consecutive super-resolved MRI slices to generate a new intermediate slice. To further enhance the network performance and perceptual quality, we introduce a feature up-sampling layer and a feature extraction block with Scaled Exponential Linear Unit (SeLU). Moreover, our super-resolution network incorporates VGG loss from a fine-tuned VGG-19 network to provide additional enhancement. Through experimental evaluations on the IXI dataset and BRATS dataset, using the peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM) and the number of training parameters, we demonstrate the superior performance of our method compared to the existing techniques. Also, the proposed model can be adapted or modified to achieve super-resolution for both 2D and 3D MRI data.

A Novel Evaluation for Vertebral Artery Course Using 3D Magnetic Resonance Imaging with Computed Tomography -like Bone Contrast and Magnetic Resonance Angiography: A Proof of Concept Study

Vertebral artery (VA) injury poses a significant risk in cervical spine surgery, necessitating accurate preoperative assessment. This study aims to introduce and validate a novel approach that combines the Fast field echo that resembles a computed tomography using restricted echo spacing (FRACTURE) sequence with Time of Flight (TOF) Magnetic Resonance Angiography (MRA) for comprehensive evaluation of VA courses in the cervical spine. A total of eight healthy volunteers and two patients participated in this study. The FRACTURE sequence provided high-resolution bone images of the cervical spine, while TOF MRA offered non-invasive vascular imaging. Fusion images were created by merging FRACTURE and MRA modalities to simultaneously visualize cervical spine structures and VA courses. Board-certified orthopedic spine surgeons independently evaluated images to assess the visibility of anatomical characteristics of the VA course by Likert-scale. The FRACTURE-MRA fusion images effectively depicted the extraosseous course of the VA at the craniovertebral junction, the intraosseous course of the VA at the craniovertebral junction, the VA entrance level to the transverse foramen, and the side-to-side asymmetry of bilateral VAs. Additionally, clinical cases demonstrated the utility of the proposed technique in identifying anomalies and guiding surgical interventions. The integration of the FRACTURE sequence and TOF MRA presents a promising methodology for the precise evaluation of VA courses in the cervical spine. This approach improves preoperative planning for cervical spine surgery with detailed anatomy and is a valuable alternative to conventional methods without contrast agents.

Reconstruction of 3D knee MRI using deep learning and compressed sensing: a validation study on healthy volunteers.

To investigate the potential of combining compressed sensing (CS) and artificial intelligence (AI), in particular deep learning (DL), for accelerating three-dimensional (3D) magnetic resonance imaging (MRI) sequences of the knee. Twenty healthy volunteers were examined using a 3-T scanner with a fat-saturated 3D proton density sequence with four different acceleration levels (10, 13, 15, and 17). All sequences were accelerated with CS and reconstructed using the conventional and a new DL-based algorithm (CS-AI). Subjective image quality was evaluated by two blinded readers using seven criteria on a 5-point-Likert-scale (overall impression, artifacts, delineation of the anterior cruciate ligament, posterior cruciate ligament, menisci, cartilage, and bone). Using mixed models, all CS-AI sequences were compared to the clinical standard (sense sequence with an acceleration factor of 2) and CS sequences with the same acceleration factor. 3D sequences reconstructed with CS-AI achieved significantly better values for subjective image quality compared to sequences reconstructed with CS with the same acceleration factor (p ≤ 0.001). The images reconstructed with CS-AI showed that tenfold acceleration may be feasible without significant loss of quality when compared to the reference sequence (p ≥ 0.999). For 3-T 3D-MRI of the knee, a DL-based algorithm allowed for additional acceleration of acquisition times compared to the conventional approach. This study, however, is limited by its small sample size and inclusion of only healthy volunteers, indicating the need for further research with a more diverse and larger sample. DRKS00024156. Using a DL-based algorithm, 54% faster image acquisition (178s versus 384s) for 3D-sequences may be possible for 3-T MRI of the knee. • Combination of compressed sensing and DL improved image quality and allows for significant acceleration of 3D knee MRI. •DL-based algorithm achieved better subjective image quality than conventional compressed sensing. •For 3D knee MRI at 3 T, 54% faster image acquisition may be possible.

3D cine-magnetic resonance imaging using spatial and temporal implicit neural representation learning (STINR-MR).

Objective. 3D cine-magnetic resonance imaging (cine-MRI) can capture images of the human body volume with high spatial and temporal resolutions to study anatomical dynamics. However, the reconstruction of 3D cine-MRI is challenged by highly under-sampled k-space data in each dynamic (cine) frame, due to the slow speed of MR signal acquisition. We proposed a machine learning-based framework, spatial and temporal implicit neural representation learning (STINR-MR), for accurate 3D cine-MRI reconstruction from highly under-sampled data.Approach. STINR-MR used a joint reconstruction and deformable registration approach to achieve a high acceleration factor for cine volumetric imaging. It addressed the ill-posed spatiotemporal reconstruction problem by solving a reference-frame 3D MR image and a corresponding motion model that deforms the reference frame to each cine frame. The reference-frame 3D MR image was reconstructed as a spatial implicit neural representation (INR) network, which learns the mapping from input 3D spatial coordinates to corresponding MR values. The dynamic motion model was constructed via a temporal INR, as well as basis deformation vector fields (DVFs) extracted from prior/onboard 4D-MRIs using principal component analysis. The learned temporal INR encodes input time points and outputs corresponding weighting factors to combine the basis DVFs into time-resolved motion fields that represent cine-frame-specific dynamics. STINR-MR was evaluated using MR data simulated from the 4D extended cardiac-torso (XCAT) digital phantom, as well as two MR datasets acquired clinically from human subjects. Its reconstruction accuracy was also compared with that of the model-based non-rigid motion estimation method (MR-MOTUS) and a deep learning-based method (TEMPEST).Main results. STINR-MR can reconstruct 3D cine-MR images with high temporal (<100 ms) and spatial (3 mm) resolutions. Compared with MR-MOTUS and TEMPEST, STINR-MR consistently reconstructed images with better image quality and fewer artifacts and achieved superior tumor localization accuracy via the solved dynamic DVFs. For the XCAT study, STINR reconstructed the tumors to a mean ± SD center-of-mass error of 0.9 ± 0.4 mm, compared to 3.4 ± 1.0 mm of the MR-MOTUS method. The high-frame-rate reconstruction capability of STINR-MR allows different irregular motion patterns to be accurately captured.Significance. STINR-MR provides a lightweight and efficient framework for accurate 3D cine-MRI reconstruction. It is a 'one-shot' method that does not require external data for pre-training, allowing it to avoid generalizability issues typically encountered in deep learning-based methods.

A Survey on AI-based Parkinson Disease Detection: Taxonomy, Case Study, and Research Challenges

Parkinson Disease (PD) is most common diseases from majority of disease encountered all over the world, with more than 7 million individuals being affected. PD is a type of progressive nervous system disease, causing deterioration in health or function. The timely identification of PD is a significant challenge because it rarely shows symptoms in the early stages. Moreover, it is typically encountered in older people where the symptoms sometimes coincide with age-related issues. Deep Learning (DL) can be integrated into many methodologies in diagnosing PD, such as Magnetic Resonance Imaging (MRI) and Single-Photon Emission Computed Tomography (SPECT). DL algorithms can detect PD based on observing some common symptoms. Moreover, it can also be detected using brain MRI images. So, in this study, we reviewed existing DL algorithms for timely identification of PD. We also have developed a CNN model for the timely identification of PD. We used 3D brain MRI images of PPMI datasets and achieved the 88% accuracy.

Diagnostic Methods for the Prenatal Detection of Cleft Lip and Palate: A Systematic Review.

Background: Accurate prenatal diagnosis of cleft lip and palate is essential to discuss severity prediction, perform appropriate parental counseling, and, at last, establish long-term treatment planning. The aim of this systematic review was to analyze the accuracy of various imaging techniques for the prenatal diagnosis of cleft lip and palate, assess the pregnancy phase for orofacial clefts diagnosis, and study the different cleft types in terms of diagnostic methods, timing, and predictability. Methods: A search of the PubMed, EMBASE, Scopus, and Web of Science databases was conducted to identify potentially relevant studies published until January 2024. The quality of the selected articles was assessed using the Newcastle-Ottawa scale for methodological quality assessment of cohort studies and the QUADAS-2 scale for diagnostic test studies. Results: A total of 18 studies met the eligibility criteria and were included in the review. The findings of this review indicate that the majority of studies showed improved diagnostic accuracy when supplementary techniques, such as 3D ultrasound or magnetic resonance imaging, were added to 2D ultrasound. Conclusions: The implementation of magnetic resonance imaging as a standard procedure could significantly improve the precision of diagnosing cleft lip and palate. Therefore, the diagnostic technique used will play a crucial role in the accuracy of the diagnosis.

3D MRI of the Wrist: Anatomy and Clinical Applications With Comparison to 2D MRI

The wrist joint has complex anatomy and is commonly involved in traumatic injuries. Magnetic resonance imaging (MRI) is routinely used to evaluate various wrist structures including the ligaments, triangular fibrocartilage complex (TFCC), tendons, and articular cartilage. While the 2D (2-dimensional) MR imaging offers high spatial resolution, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), it has some limitations due to relatively thick slices, limited fluid contrast, and the need for image acquisition in multiple planes. Isotropic 3D MR imaging can overcome these limitations by acquiring a volumetric dataset with thin contiguous slices enabling multiplanar reconstructions in arbitrary, though structure-specific planes. Recent advances have reduced the acquisition duration of 3D imaging datasets making them more time efficient than 2D MRI thus enabling clinical implementation. In this article, we will review the main concepts on wrist anatomy and imaging, present illustrative cases using recent technical advances of 3D MRI and compare 2D vs. 3D benefits for various wrist structures and their pathologies.

Novel 3D MRI-Based Volumetric Assessment of Rotator Cuff Musculature Demonstrates Stronger Correlation with Preoperative Functional Status When Compared to the Goutallier Grading Scheme

The Goutallier classification (GC) is used to assess fatty atrophy in rotator cuff (RC) tears, yet limitations exist. A battery of 3D-magnetic resonance imaging (MRI) volumetric scores (VS) was developed to provide comprehensive characterization of RC pathology. The purposes of this study were to: (1) Describe the correlation between GC and VS for supraspinatus changes in RC tears, (2) Characterize the chronicity of RC tears using the battery of 12 VS measurements, and (3) Compare GC and VS to determine which method most closely corresponds with preoperative patient reported outcome measures (PROMs). Preoperative shoulder MRIs were reviewed after arthroscopic RC repair. Preoperative GC stage and Patient-Reported Outcomes Measurement Information System (PROMIS) physical function (PF) and pain interference (PI) scores were collected. The battery of VS included fat infiltration (FIS), muscle size (MSS) and relative volume contribution (RCS) for each RC muscle. Backwards linear regression was performed to compare GC stage with preoperative PROMIS PF/PI to determine which VS measurement most closely correlated with preoperative PROMs. Eighty-two patients underwent RC repair (mean age 55±8.2 years, 63% male, 68% GC stage ≤1). In evaluation of the supraspinatus, there was a moderate positive correlation between GC and FIS (r = 0.459, p < 0.001); strong negative correlations were observed between MSS (r = -0.800, p < 0.001) and RCS (r = -0.745, p < 0.001) when compared to GC. A negligible linear correlation was observed between GC and preoperative PROMIS PF (r = -0.106, p = 0.343) and PI (r = -0.071, p = 0.528). On multivariate analysis, subscapularis MSS (beta > 0, p = 0.064) was a positive predictor, and subscapularis FIS (beta < 0, p = 0.137), teres minor MSS (beta < 0, p = 0.141) and FIS (beta < 0, p = 0.070) were negative predictors of preoperative PF (r = 0.343, p = 0.044); while supraspinatus MSS (beta > 0, p = 0.009) and FIS (beta > 0, p = 0.073), teres minor FIS (beta > 0, p = 0.072) and subscapularis FIS (beta > 0, p = 0.065) were positive predictors of preoperative PI (r = 0.410, p = 0.006). Although gold standard in evaluation of RC pathology, GC demonstrated negligible correlation with preoperative functional disability. Alternatively, a battery of 3D VS showed strong correlation with GC through a quantitative, comprehensive evaluation of the RC unit including several moderate predictors of preoperative functional disability.

Time-Saving 3D MR Imaging Protocols with Millimeter and Submillimeter Isotropic Spatial Resolution for Face and Neck Imaging as Implemented at a Single-Site Major Referral Center.

MR imaging has become the routine technique for staging nasopharyngeal carcinoma, evaluating perineural tumor spread, and detecting cartilage invasion in laryngeal carcinoma. However, these protocols traditionally require in the range of 25 to 35 minutes of acquisition time. 3D sequences offer the potential advantage of time savings through the acquisition of 1-mm or submillimeter resolution isotropic data followed by multiplanar reformats that require no further imaging time. We have iteratively optimized vendor product 3D T1-weighted MR imaging sequences for morphologic face and neck imaging, reducing the average acquisition time of our 3T protocols by 9 minutes 57 seconds (40.9%) and of our 1.5T protocols by 9 minutes 5 seconds (37.0%), while simultaneously maintaining or improving spatial resolution. This clinical report describes our experience optimizing and implementing commercially available 3D T1-weighted MR imaging pulse sequence protocols for clinical face and neck MR imaging examinations using illustrative cases. We provide protocol details to allow others to replicate our implementations, and we report challenges we faced along with our solutions.

Impact of bias field correction on 0.35 T pelvic MR images: evaluation on generative adversarial network-based OARs' auto-segmentation and visual grading assessment.

Magnetic resonance imaging (MRI)-guided radiotherapy enables adaptive treatment plans based on daily anatomical changes and accurate organ visualization. However, the bias field artifact can compromise image quality, affecting diagnostic accuracy and quantitative analyses. This study aims to assess the impact of bias field correction on 0.35 T pelvis MRIs by evaluating clinical anatomy visualization and generative adversarial network (GAN) auto-segmentation performance. 3D simulation MRIs from 60 prostate cancer patients treated on MR-Linac (0.35 T) were collected and preprocessed with the N4ITK algorithm for bias field correction. A 3D GAN architecture was trained, validated, and tested on 40, 10, and 10 patients, respectively, to auto-segment the organs at risk (OARs) rectum and bladder. The GAN was trained and evaluated either with the original or the bias-corrected MRIs. The Dice similarity coefficient (DSC) and 95th percentile Hausdorff distance (HD95th) were computed for the segmented volumes of each patient. The Wilcoxon signed-rank test assessed the statistical difference of the metrics within OARs, both with and without bias field correction. Five radiation oncologists blindly scored 22 randomly chosen patients in terms of overall image quality and visibility of boundaries (prostate, rectum, bladder, seminal vesicles) of the original and bias-corrected MRIs. Bennett's S score and Fleiss' kappa were used to assess the pairwise interrater agreement and the interrater agreement among all the observers, respectively. In the test set, the GAN trained and evaluated on original and bias-corrected MRIs showed DSC/HD95th of 0.92/5.63mm and 0.92/5.91mm for the bladder and 0.84/10.61mm and 0.83/9.71mm for the rectum. No statistical differences in the distribution of the evaluation metrics were found neither for the bladder (DSC: p = 0.07; HD95th: p = 0.35) nor for the rectum (DSC: p = 0.32; HD95th: p = 0.63). From the clinical visual grading assessment, the bias-corrected MRI resulted mostly in either no change or an improvement of the image quality and visualization of the organs' boundaries compared with the original MRI. The bias field correction did not improve the anatomy visualization from a clinical point of view and the OARs' auto-segmentation outputs generated by the GAN.

The use of individual-based FDG-PET volume of interest in predicting conversion from mild cognitive impairment to dementia

BackgroundBased on a longitudinal cohort design, the aim of this study was to investigate whether individual-based 18F fluorodeoxyglucose positron emission tomography (18F-FDG-PET) regional signals can predict dementia conversion in patients with mild cognitive impairment (MCI).MethodsWe included 44 MCI converters (MCI-C), 38 non-converters (MCI-NC), 42 patients with Alzheimer’s disease with dementia, and 40 cognitively normal controls. Data from annual cognitive measurements, 3D T1 magnetic resonance imaging (MRI) scans, and 18F-FDG-PET scans were used for outcome analysis. An individual-based FDG-PET approach was applied using seven volumes of interest (VOIs), Z transformed using a normal FDG-PET template. Hypometabolism was defined as a Z score < -2 of regional standard uptake value ratio. For the longitudinal cognitive test scores, generalized estimating equations were used. A linear mixed-effects model was used to compare the temporal impact of cortical hypometabolism and cortical thickness degeneration.ResultsThe clinical follow-up period was 6.6 ± 3.8 years (range 3.1 to 16.0 years). The trend of cognitive decline could differentiate MCI-C from MCI-NC after 3 years of follow-up. In the baseline 18F-FDG-PET scan of the patients with MCI, medial temporal lobe (MTL; 94.7% sensitivity, 80.5% specificity) and posterior cingulate cortex (PCC; 89.5% sensitivity, 73.1% specificity) hypometabolism predicted conversion with high accuracy. 18F-FDG-PET hypometabolism preceded dementia conversion at an interval of 3.70 ± 1.68 years and was earlier than volumetric changes, with the exception of the MTL.ConclusionsOur finding supports the use of individual-based 18F-FDG-PET analysis to predict MCI conversion to dementia. Reduced FDG-PET metabolism in the MTL and PCC were strongly associated with future cognitive decline in the MCI-C group. Changes in 18F-FDG-PET occurred 1 to 8 years prior to conversion to dementia. Progressive hypometabolism in the PCC, precuneus and lateral temporal lobe, but not MTL, preceded MRI findings at the MCI stage.

Volumetric assessment of ulnar nerves in cubital tunnel syndrome with 3D modeling of the MRI and its relationship with electrodiagnostic findings

BackgroundThickened nerve cross-sectional areas (CSA) have been investigated in compressive neuropathy, but the longitudinal extent of nerve swelling has yet to be evaluated. We did a volumetric assessment of the ulnar nerve in cubital tunnel syndrome (CuTS) with three-dimensional (3D) magnetic resonance imaging (MRI) modeling and investigated this relationship with clinical and electrodiagnostic parameters. MethodsWe compared 40 CuTS patient elbow MRIs to 46 patient elbow MRIs with lateral elbow epicondylitis as controls. The ulnar nerve was modeled with Mimics software and was assessed qualitatively and quantitatively. The CSA and ulnar nerve volumes were recorded, and the area under the receiver operating characteristic (ROC) curve was calculated for diagnostic performance. We analyzed clinical and electrodiagnostic parameters to investigate their relationship with the 3D ulnar nerve parameters. ResultsFor the diagnosis of CuTS, the area under the curve value was 0.915 for the largest CSA and 0.910 for the volume in the ROC curve. The optimal cut-off was 14.53 mm2 and 529 mm3 respectively. When electrodiagnostic parameters were investigated, the 3D ulnar nerve volume was significantly inversely associated with motor conduction velocity, although there was no association between the largest CSA and any of the electrodiagnostic parameters. ConclusionsThe 3D ulnar nerve volume, which is an integration or multilevel measurement of CSAs, showed diagnostic usefulness similar to CSA, but it correlated better with conduction velocity, indicating demyelination or early-to-moderate nerve damage in CuTS.

Deep learning-based automatic pipeline for 3D needle localization on intra-procedural 3D MRI.

Accurate and rapid needle localization on 3D magnetic resonance imaging (MRI) is critical for MRI-guided percutaneous interventions. The current workflow requires manual needle localization on 3D MRI, which is time-consuming and cumbersome. Automatic methods using 2D deep learning networks for needle segmentation require manual image plane localization, while 3D networks are challenged by the need for sufficient training datasets. This work aimed to develop an automatic deep learning-based pipeline for accurate and rapid 3D needle localization on in vivo intra-procedural 3D MRI using a limited training dataset. The proposed automatic pipeline adopted Shifted Window (Swin) Transformers and employed a coarse-to-fine segmentation strategy: (1) initial 3D needle feature segmentation with 3D Swin UNEt TRansfomer (UNETR); (2) generation of a 2D reformatted image containing the needle feature; (3) fine 2D needle feature segmentation with 2D Swin Transformer and calculation of 3D needle tip position and axis orientation. Pre-training and data augmentation were performed to improve network training. The pipeline was evaluated via cross-validation with 49 in vivo intra-procedural 3D MR images from preclinical pig experiments. The needle tip and axis localization errors were compared with human intra-reader variation using the Wilcoxon signed rank test, with p < 0.05 considered significant. The average end-to-end computational time for the pipeline was 6s per 3D volume. The median Dice scores of the 3D Swin UNETR and 2D Swin Transformer in the pipeline were 0.80 and 0.93, respectively. The median 3D needle tip and axis localization errors were 1.48mm (1.09 pixels) and 0.98°, respectively. Needle tip localization errors were significantly smaller than human intra-reader variation (median 1.70mm; p < 0.01). The proposed automatic pipeline achieved rapid pixel-level 3D needle localization on intra-procedural 3D MRI without requiring a large 3D training dataset and has the potential to assist MRI-guided percutaneous interventions.

Constantly optimized mean teacher for semi-supervised 3D MRI image segmentation.

The mean teacher model and its variants, as important methods in semi-supervised learning, have demonstrated promising performance in magnetic resonance imaging (MRI) data segmentation. However, the superior performance of teacher model through exponential moving average (EMA) is limited by the unreliability of unlabeled image, resulting in potentially unreliable predictions. In this paper, we propose a framework to optimized the teacher model with reliable expert-annotated data while preserving the advantages of EMA. To avoid the tight coupling that results from EMA, we leverage data augmentations to provide two distinct perspectives for the teacher and student models. The teacher model adopts weak data augmentation to provide supervision for the student model and optimizes itself with real annotations, while the student uses strong data augmentation to avoid overfitting on noise information. In addition, double softmax helps the model resist noise and continue learning meaningful information from the images, which is a key component in the proposed model. Extensive experiments show that the proposed method exhibits competitive performance on the Left Atrium segmentation MRI dataset (LA) and the Brain Tumor Segmentation MRI dataset (BraTS2019). For the LA dataset, we achieved a dice of 91.02% using only 20% labeled data, which is close to the dice of 91.14% obtained by the supervised approach using 100% labeled data. For the BraTs2019 dataset, the proposed method achieved 1.02% and 1.92% improvement on 5% and 10% labeled data, respectively, compared to the best baseline method on this dataset. This study demonstrates that the proposed model can be a potential candidate for medical image segmentation in semi-supervised learning scenario.