Imaging segmentation mechanism for rectal tumors using improved U-Net.

OBJECTIVE: In 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. METHODS: The 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. RESULTS: We 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. CONCLUSION: Due 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. SIGNIFICANCE: The proposed method can help doctors segment rectal tumors, thereby ensuring good diagnostic quality and accuracy.

Discrimination models with radiomics features derived from cardiovascular magnetic resonance images for distinguishing hypertensive heart disease from hypertrophic cardiomyopathy.

Background: Discriminating hypertrophic cardiomyopathy (HCM) and hypertensive heart disease (HHD) is challenging, because both are characterized by left ventricular hypertrophy (LVH). Radiomics might be effective to differentiate HHD from HCM. Therefore, this study aimed to investigate discriminators and build discrimination models between HHD and HCM using multiparametric cardiac magnetic resonance (CMR) findings and radiomics score (radscore) derived from late gadolinium enhancement (LGE) and cine images. Methods: In this single center, retrospective study, 421 HCM patients [median and interquartile range (IQR), 50.0 (38.0-59.0) years; male, 70.5%] from January 2017 to September 2021 and 200 HHD patients [median and IQR, 44.5 (35.0-57.0) years; male, 88.5%] from September 2015 to July 2022 were consecutively included and randomly stratified into a training group and a validation group at a ratio of 6:4. Multiparametric CMR findings were obtained using cvi42 software and radiomics features using Python software. After dimensional reduction, the radscore was calculated by summing the remaining radiomics features weighted by their coefficients. Multiparametric CMR findings and radscore that were statistically significant in univariate logistic regression were used to build combined discrimination models via multivariate logistic regression. Results: After multivariate logistic regression, the maximal left ventricular end diastolic wall thickness (LVEDWT), left ventricular ejection fraction (LVEF), presence of LGE, cine radscore and LGE radscore were identified as significant characteristics and used to build a combined discrimination model. This model achieved an area under the receiver operator characteristic curve (AUC) of 0.979 (0.968-0.990) in the training group and 0.981 (0.967-0.995) in the validation group, significantly better than the model using multiparametric CMR findings alone (P<0.001). Conclusions: Radiomics features derived from cardiac cine and LGE images can effectively discriminate HHD from HCM.

A Lightweight Convolutional Neural Network Based on Dynamic Level-Set Loss Function for Spine MR Image Segmentation.

BACKGROUND: Spine MR image segmentation is important foundation for computer-aided diagnostic (CAD) algorithms of spine disorders. Convolutional neural networks segment effectively, but require high computational costs. PURPOSE: To design a lightweight model based on dynamic level-set loss function for high segmentation performance. STUDY TYPE: Retrospective. POPULATION: Four hundred forty-eight subjects (3163 images) from two separate datasets. Dataset-1: 276 subjects/994 images (53.26% female, mean age 49.02 ± 14.09), all for disc degeneration screening, 188 had disc degeneration, 67 had herniated disc. Dataset-2: public dataset with 172 subjects/2169 images, 142 patients with vertebral degeneration, 163 patients with disc degeneration. FIELD STRENGTH/SEQUENCE: T2 weighted turbo spin echo sequences at 3T. ASSESSMENT: Dynamic Level-set Net (DLS-Net) was compared with four mainstream (including U-net++) and four lightweight models, and manual label made by five radiologists (vertebrae, discs, spinal fluid) used as segmentation evaluation standard. Five-fold cross-validation are used for all experiments. Based on segmentation, a CAD algorithm of lumbar disc was designed for assessing DLS-Net's practicality, and the text annotation (normal, bulging, or herniated) from medical history data were used as evaluation standard. STATISTICAL TESTS: All segmentation models were evaluated with DSC, accuracy, precision, and AUC. The pixel numbers of segmented results were compared with manual label using paired t-tests, with P < 0.05 indicating significance. The CAD algorithm was evaluated with accuracy of lumbar disc diagnosis. RESULTS: With only 1.48% parameters of U-net++, DLS-Net achieved similar accuracy in both datasets (Dataset-1: DSC 0.88 vs. 0.89, AUC 0.94 vs. 0.94; Dataset-2: DSC 0.86 vs. 0.86, AUC 0.93 vs. 0.93). The segmentation results of DLS-Net showed no significant differences with manual labels in pixel numbers for discs (Dataset-1: 1603.30 vs. 1588.77, P = 0.22; Dataset-2: 863.61 vs. 886.4, P = 0.14) and vertebrae (Dataset-1: 3984.28 vs. 3961.94, P = 0.38; Dataset-2: 4806.91 vs. 4732.85, P = 0.21). Based on DLS-Net's segmentation results, the CAD algorithm achieved higher accuracy than using non-cropped MR images (87.47% vs. 61.82%). DATA CONCLUSION: The proposed DLS-Net has fewer parameters but achieves similar accuracy to U-net++, helps CAD algorithm achieve higher accuracy, which facilitates wider application. EVIDENCE LEVEL: 2 TECHNICAL EFFICACY: Stage 1.

Collaborative Learning for Annotation-Efficient Volumetric MR Image Segmentation.

BACKGROUND: Deep learning has presented great potential in accurate MR image segmentation when enough labeled data are provided for network optimization. However, manually annotating three-dimensional (3D) MR images is tedious and time-consuming, requiring experts with rich domain knowledge and experience. PURPOSE: To build a deep learning method exploring sparse annotations, namely only a single two-dimensional slice label for each 3D training MR image. STUDY TYPE: Retrospective. POPULATION: Three-dimensional MR images of 150 subjects from two publicly available datasets were included. Among them, 50 (1377 image slices) are for prostate segmentation. The other 100 (8800 image slices) are for left atrium segmentation. Five-fold cross-validation experiments were carried out utilizing the first dataset. For the second dataset, 80 subjects were used for training and 20 were used for testing. FIELD STRENGTH/SEQUENCE: 1.5 T and 3.0 T; axial T2-weighted and late gadolinium-enhanced, 3D respiratory navigated, inversion recovery prepared gradient echo pulse sequence. ASSESSMENT: A collaborative learning method by integrating the strengths of semi-supervised and self-supervised learning schemes was developed. The method was trained using labeled central slices and unlabeled noncentral slices. Segmentation performance on testing set was reported quantitatively and qualitatively. STATISTICAL TESTS: Quantitative evaluation metrics including boundary intersection-over-union (B-IoU), Dice similarity coefficient, average symmetric surface distance, and relative absolute volume difference were calculated. Paired t test was performed, and P < 0.05 was considered statistically significant. RESULTS: Compared to fully supervised training with only the labeled central slice, mean teacher, uncertainty-aware mean teacher, deep co-training, interpolation consistency training (ICT), and ambiguity-consensus mean teacher, the proposed method achieved a substantial improvement in segmentation accuracy, increasing the mean B-IoU significantly by more than 10.0% for prostate segmentation (proposed method B-IoU: 70.3% ± 7.6% vs. ICT B-IoU: 60.3% ± 11.2%) and by more than 6.0% for left atrium segmentation (proposed method B-IoU: 66.1% ± 6.8% vs. ICT B-IoU: 60.1% ± 7.1%). DATA CONCLUSIONS: A collaborative learning method trained using sparse annotations can segment prostate and left atrium with high accuracy. LEVEL OF EVIDENCE: 0 TECHNICAL EFFICACY: Stage 1.

Sparse annotation learning for dense volumetric MR image segmentation with uncertainty estimation.

Objective.Training neural networks for pixel-wise or voxel-wise image segmentation is a challenging task that requires a considerable amount of training samples with highly accurate and densely delineated ground truth maps. This challenge becomes especially prominent in the medical imaging domain, where obtaining reliable annotations for training samples is a difficult, time-consuming, and expert-dependent process. Therefore, developing models that can perform well under the conditions of limited annotated training data is desirable.Approach.In this study, we propose an innovative framework called the extremely sparse annotation neural network (ESA-Net) that learns with only the single central slice label for 3D volumetric segmentation which explores both intra-slice pixel dependencies and inter-slice image correlations with uncertainty estimation. Specifically, ESA-Net consists of four specially designed distinct components: (1) an intra-slice pixel dependency-guided pseudo-label generation module that exploits uncertainty in network predictions while generating pseudo-labels for unlabeled slices with temporal ensembling; (2) an inter-slice image correlation-constrained pseudo-label propagation module which propagates labels from the labeled central slice to unlabeled slices by self-supervised registration with rotation ensembling; (3) a pseudo-label fusion module that fuses the two sets of generated pseudo-labels with voxel-wise uncertainty guidance; and (4) a final segmentation network optimization module to make final predictions with scoring-based label quantification.Main results.Extensive experimental validations have been performed on two popular yet challenging magnetic resonance image segmentation tasks and compared to five state-of-the-art methods.Significance.Results demonstrate that our proposed ESA-Net can consistently achieve better segmentation performances even under the extremely sparse annotation setting, highlighting its effectiveness in exploiting information from unlabeled data.

Could the Cerebral Involvement of COVID-19 Disease be Related to Microstructural Changes that are not Reflected in Conventional MRI Images?

In patients with COVID-19, neurodegeneration may develop before clinical symptoms appear. Diffusion-weighted (DW) MRI is an important technique for analyzing microstructural changes such as gliosis. In this study, a quantitative evaluation of microstructural changes in the brain with apparent diffusion coefficient (ADC) values in patients presenting with a headache after the COVID-19 disease was analyzed and compared. DW MR images of patients of 20 COVID-19 patients (13 females, 7 males) who required imaging due to headache; 20 controls (16 females, 4 males) were retrospectively reevaluated. ADC measurements were taken from 16 regions of the brain, including right and left symmetrical in patients with COVID-19 infections and control groups. All regions of interest (ROIs) were taken from the hypothalamus, parahippocampus, thalamus, corpus striatum, cingulate gyrus, occipital gyrus, dentate nucleus, and medulla oblongata posterior. ADC values in the dentate nucleus right (784.6 ± 75.7 vs. 717.25 ± 50.75), dentate nucleus left (768.05 ± 69.76 vs. 711.40 ± 52.99), right thalamus (731.15 ± 38.14 vs. 701.60 ± 43.65), left thalamus (744.05 ± 39.00 vs. 702.85 ± 28.88), right parahippocampus (789.10 ± 56.35 vs. 754.75 ± 33.78), right corpus striatum (710.00 ± 39.81 vs. 681.55 ± 39.84) were significantly higher than those in the control group. No significant changes were observed in other areas. A significant increase in ADC values at many levels in the brain in patients with COVID-19 disease and headache was observed. Thus, this study indicates that cerebral involvement in COVID-19 disease may be related to microstructural changes that are not reflected in conventional MRI images.

Bayesian statistical modeling to predict observer-specific optimal windowing parameters in magnetic resonance imaging.

Magnetic resonance (MR) images require a process known as windowing for optimizing the display conditions. However, the conventional windowing process often fails to achieve the preferred display conditions for observers due to various factors. This study proposes a novel framework for predicting the preferred windowing parameters for each observer using Bayesian statistical modeling. MR images obtained from 1000 patients were divided into training and test sets at a 7:3 ratio. The image intensity and windowing parameters were standardized using previously reported methods. Bayesian statistical modeling was utilized to predict the windowing parameters preferred by three MR imaging (MRI) operators. The performance of the proposed framework was evaluated by assessing the mean relative error (MRE), mean absolute error (MAE), and Pearson's correlation coefficient (ρ) of the test set. In addition, the naive method, which presumes that the average value of the windowing parameters for each acquisition sequence and body region in the training set is optimal, was also used for comparison. Three MRI operators and three radiologists conducted visual assessments. The mean MRE, MAE, and ρ values for the window level and width (WL/WW) in the proposed framework were 12.6 and 13.9, 42.9 and 85.4, and 0.98 and 0.98, respectively. These results outperformed those obtained using the naive method. The visual assessments revealed no significant differences between the original and predicted display conditions, indicating that the proposed framework accurately predicts individualized windowing parameters with the additional advantages of robustness and ease of use. Thus, the proposed framework can effectively predict the windowing parameters preferred by each observer.

Data-driven electrical conductivity brain imaging using 3 T MRI.

Magnetic resonance electrical properties tomography (MR-EPT) is a non-invasive measurement technique that derives the electrical properties (EPs, e.g., conductivity or permittivity) of tissues in the radiofrequency range (64 MHz for 1.5 T and 128 MHz for 3 T MR systems). Clinical studies have shown the potential of tissue conductivity as a biomarker. To date, model-based conductivity reconstructions rely on numerical assumptions and approximations, leading to inaccuracies in the reconstructed maps. To address such limitations, we propose an artificial neural network (ANN)-based non-linear conductivity estimator trained on simulated data for conductivity brain imaging. Network training was performed on 201 synthesized T2-weighted spin-echo (SE) data obtained from the finite-difference time-domain (FDTD) electromagnetic (EM) simulation. The dataset was composed of an approximated T2-w SE magnitude and transceive phase information. The proposed method was tested three in-silico and in-vivo on two volunteers and three patients' data. For comparison purposes, various conventional phase-based EPT reconstruction methods were used that ignore B 1 + magnitude information, such as Savitzky-Golay kernel combined with Gaussian filter (S-G Kernel), phase-based convection-reaction EPT (cr-EPT), magnitude-weighted polynomial-fitting phase-based EPT (Poly-Fit), and integral-based phase-based EPT (Integral-based). From the in-silico experiments, quantitative analysis showed that the proposed method provides more accurate and improved quality (e.g., high structural preservation) conductivity maps compared to conventional reconstruction methods. Representatively, in the healthy brain in-silico phantom experiment, the proposed method yielded mean conductivity values of 1.97 ± 0.20 S/m for CSF, 0.33 ± 0.04 S/m for WM, and 0.52 ± 0.08 S/m for GM, which were closer to the ground-truth conductivity (2.00, 0.30, 0.50 S/m) than the integral-based method (2.56 ± 2.31, 0.39 ± 0.12, 0.68 ± 0.33 S/m). In-vivo ANN-based conductivity reconstructions were also of improved quality compared to conventional reconstructions and demonstrated network generalizability and robustness to in-vivo data and pathologies. The reported in-vivo brain conductivity values were in agreement with literatures. In addition, the proposed method was observed for various SNR levels (SNR levels = 10, 20, 40, and 58) and repeatability conditions (the eight acquisitions with the number of signal averages = 1). The preliminary investigations on brain tumor patient datasets suggest that the network trained on simulated dataset can generalize to unforeseen in-vivo pathologies, thus demonstrating its potential for clinical applications.

Dual adversarial models with cross-coordination consistency constraint for domain adaption in brain tumor segmentation.

The brain tumor segmentation task with different domains remains a major challenge because tumors of different grades and severities may show different distributions, limiting the ability of a single segmentation model to label such tumors. Semi-supervised models (e.g., mean teacher) are strong unsupervised domain-adaptation learners. However, one of the main drawbacks of using a mean teacher is that given a large number of iterations, the teacher model weights converge to those of the student model, and any biased and unstable predictions are carried over to the student. In this article, we proposed a novel unsupervised domain-adaptation framework for the brain tumor segmentation task, which uses dual student and adversarial training techniques to effectively tackle domain shift with MR images. In this study, the adversarial strategy and consistency constraint for each student can align the feature representation on the source and target domains. Furthermore, we introduced the cross-coordination constraint for the target domain data to constrain the models to produce more confident predictions. We validated our framework on the cross-subtype and cross-modality tasks in brain tumor segmentation and achieved better performance than the current unsupervised domain-adaptation and semi-supervised frameworks.

Bidirectional feature matching based on deep pairwise contrastive learning for multiparametric MRI image synthesis.

Objective.Multi-parametric MR image synthesis is an effective approach for several clinical applications where specific modalities may be unavailable to reach a diagnosis. While technical and practical conditions limit the acquisition of new modalities for a patient, multimodal image synthesis combines multiple modalities to synthesize the desired modality.Approach.In this paper, we propose a new multi-parametric magnetic resonance imaging (MRI) synthesis model, which generates the target MRI modality from two other available modalities, in pathological MR images. We first adopt a contrastive learning approach that trains an encoder network to extract a suitable feature representation of the target space. Secondly, we build a synthesis network that generates the target image from a common feature space that approximately matches the contrastive learned space of the target modality. We incorporate a bidirectional feature learning strategy that learns a multimodal feature matching function, in two opposite directions, to transform the augmented multichannel input in the learned target space. Overall, our training synthesis loss is expressed as the combination of the reconstruction loss and a bidirectional triplet loss, using a pair of features.Main results.Compared to other state-of-the-art methods, the proposed model achieved an average improvement rate of 3.9% and 3.6% on the IXI and BraTS'18 datasets respectively. On the tumor BraTS'18 dataset, our model records the highest Dice score of 0.793(0.04) for preserving the synthesized tumor regions in the segmented images.Significance.Validation of the proposed model on two public datasets confirms the efficiency of the model to generate different MR contrasts, and preserve tumor areas in the synthesized images. In addition, the model is flexible to generate head and neck CT image from MR acquisitions. In future work, we plan to validate the model using interventional iMRI contrasts for MR-guided neurosurgery applications, and also for radiotherapy applications. Clinical measurements will be collected during surgery to evaluate the model's performance.

An in-vivo study of the combined therapeutic effects of pulsed non-thermal focused ultrasound and radiation for prostate cancer.

PURPOSE: The purpose of this study was to investigate the in vivo combined effects of pulsed focused ultrasound (pFUS) and radiation (RT) for prostate cancer treatment. MATERIALS AND METHODS: An animal prostate tumor model was developed by implanting human LNCaP tumor cells in the prostates of nude mice. Tumor-bearing mice were treated with pFUS, RT or both (pFUS + RT) and compared with a control group. Non-thermal pFUS treatment was delivered by keeping the body temperature below 42 °C as measured real-time by MR thermometry and using a pFUS protocol (1 MHz, 25 W focused ultrasound; 1 Hz pulse rate with a 10% duty cycle for 60 sec for each sonication). Each tumor was covered entirely using 4-8 sonication spots. RT treatment with a dose of 2 Gy was delivered using an external beam (6 MV photon energy with dose rate 300MU/min). Following the treatment, mice were scanned weekly with MRI for tumor volume measurement. RESULTS: The results showed that the tumor volume in the control group increased exponentially to 142 ± 6%, 205 ± 12%, 286 ± 22% and 410 ± 33% at 1, 2, 3 and 4 weeks after treatment, respectively. In contrast, the pFUS group was 29% (p < 0.05), 24% (p < 0.05), 8% and 9% smaller, the RT group was 7%, 10%, 12% and 18% smaller, and the pFUS + RT group was 32%, 39%, 41% and 44% (all with p < 0.05) smaller than the control group at 1, 2, 3, and 4 weeks post treatment, respectively. Tumors treated by pFUS showed an early response (i.e. the first 2 weeks), while the RT group showed a late response. The combined pFUS + RT treatment showed consistent response throughout the post-treatment weeks. CONCLUSIONS: These results suggest that RT combined with non-thermal pFUS can significantly delay the tumor growth. The mechanism of tumor cell killing between pFUS and RT may be different. Pulsed FUS shows early tumor growth delay, while RT contributes to the late effect on tumor growth delay. The addition of pFUS to RT significantly enhanced the therapeutic effect for prostate cancer treatment.

An Improved Soft Subspace Clustering Algorithm Based on Particle Swarm Optimization for MR Image Segmentation.

Soft subspace clustering (SSC), which analyzes high-dimensional data and applies various weights to each cluster class to assess the membership degree of each cluster to the space, has shown promising results in recent years. This method of clustering assigns distinct weights to each cluster class. By introducing spatial information, enhanced SSC algorithms improve the degree to which intraclass compactness and interclass separation are achieved. However, these algorithms are sensitive to noisy data and have a tendency to fall into local optima. In addition, the segmentation accuracy is poor because of the influence of noisy data. In this study, an SSC approach that is based on particle swarm optimization is suggested with the intention of reducing the interference caused by noisy data. The particle swarm optimization method is used to locate the best possible clustering center. Second, increasing the amount of geographical membership makes it possible to utilize the spatial information to quantify the link between different clusters in a more precise manner. In conclusion, the extended noise clustering method is implemented in order to maximize the weight. Additionally, the constraint condition of the weight is changed from the equality constraint to the boundary constraint in order to reduce the impact of noise. The methodology presented in this research works to reduce the amount of sensitivity the SSC algorithm has to noisy data. It is possible to demonstrate the efficacy of this algorithm by using photos with noise already present or by introducing noise to existing photographs. The revised SSC approach based on particle swarm optimization (PSO) is demonstrated to have superior segmentation accuracy through a number of trials; as a result, this work gives a novel method for the segmentation of noisy images.

An optimized segmentation convolutional neural network with dynamic energy loss function for 3D reconstruction of lumbar spine MR images.

3D reconstruction for lumbar spine based on segmentation of Magnetic Resonance (MR) images is meaningful for diagnosis of degenerative lumbar spine diseases. However, spine MR images with unbalanced pixel distribution often cause the segmentation performance of Convolutional Neural Network (CNN) reduced. Designing a composite loss function for CNN is an effective way to enhance the segmentation capacity, yet composition loss values with fixed weight may still cause underfitting in CNN training. In this study, we designed a composite loss function with a dynamic weight, called Dynamic Energy Loss, for spine MR images segmentation. In our loss function, the weight percentage of different loss values could be dynamically adjusted during training, thus CNN could fast converge in earlier training stage and focus on detail learning in the later stage. Two datasets were used in control experiments, and the U-net CNN model with our proposed loss function achieved superior performance with Dice similarity coefficient values of 0.9484 and 0.8284 respectively, which were also verified by the Pearson correlation, Bland-Altman, and intra-class correlation coefficient analysis. Furthermore, to improve the 3D reconstruction based on the segmentation results, we proposed a filling algorithm to generate contextually related slices by computing the pixel-level difference between adjacent slices of segmented images, which could enhance the structural information of tissues between slices, and improve the performance of 3D lumbar spine model rendering. Our methods could help radiologists to build a 3D lumbar spine graphical model accurately for diagnosis while reducing burden of manual image reading.

Mn-doped Ti-based MOFs for magnetic resonance imaging-guided synergistic microwave thermal and microwave dynamic therapy of liver cancer.

Currently, precise ablation of tumors without damaging the surrounding normal tissue is still an urgent problem for clinical microwave therapy of liver cancer. Herein, we synthesized Mn-doped Ti MOFs (Mn-Ti MOFs) nanosheets by in-situ doping method and applied them for microwave therapy. Infrared thermal imaging results indicate Mn-Ti MOFs can rapidly increase the temperature of normal saline, attributing to the porous structure improving microwave-induced ion collision frequency. Moreover, Mn-Ti MOFs show higher 1O2 output than Ti MOFs under 2 W of low-power microwave irradiation due to the narrower band-gap after Mn doping. At the same time, Mn endows the MOFs with a desirable T1 contrast of magnetic resonance imaging (r2/r1 = 2.315). Further, results on HepG2 tumor-bearing mice prove that microwave-triggered Mn-Ti MOFs nearly eradicate the tumors after 14 days of treatment. Our study offers a promising sensitizer for synergistic microwave thermal and microwave dynamic therapy of liver cancer.

LSW-Net: Lightweight Deep Neural Network Based on Small-World properties for Spine MR Image Segmentation.

BACKGROUND: Segmenting spinal tissues from MR images is important for automatic image analysis. Deep neural network-based segmentation methods are efficient, yet have high computational costs. PURPOSE: To design a lightweight model based on small-world properties (LSW-Net) to segment spinal MR images, suitable for low-computing-power embedded devices. STUDY TYPE: Retrospective. POPULATION: A total of 386 subjects (2948 images) from two independent sources. Dataset I: 214 subjects/779 images, all for disk degeneration screening, 147 had disk degeneration, 52 had herniated disc. Dataset II: 172 subjects/2169 images, 142 patients with vertebral degeneration, 163 patients with disc degeneration. 70% images in each dataset for training, 20% for validation, and 10% for testing. FIELD STRENGTH/SEQUENCE: T1- and T2-weighted turbo spin echo sequences at 3 T. ASSESSMENT: Segmentation performance of LSW-Net was compared with four mainstream (including U-net and U-net++) and five lightweight models using five radiologists' manual segmentations (vertebrae, disks, spinal fluid) as reference standard. LSW-Net was also deployed on NVIDIA Jetson nano to compare the pixels number in segmented vertebrae and disks. STATISTICAL TESTS: All models were evaluated with accuracy, precision, Dice similarity coefficient (DSC), and area under the receiver operating characteristic (AUC). Pixel numbers segmented by LSW-Net on the embedded device were compared with manual segmentation using paired t-tests, with P < 0.05 indicating significance. RESULTS: LSW-Net had 98.5% fewer parameters than U-net but achieved similar accuracy in both datasets (dataset I: DSC 0.84 vs. 0.87, AUC 0.92 vs. 0.94; dataset II: DSC 0.82 vs. 0.82, AUC 0.88 vs. 0.88). LSW-Net showed no significant differences in pixel numbers for vertebrae (dataset I: 5893.49 vs. 5752.61, P = 0.21; dataset II: 5073.42 vs. 5137.12, P = 0.56) and disks (dataset I: 1513.07 vs. 1535.69, P = 0.42; dataset II: 1049.74 vs. 1087.88, P = 0.24) segmentation on an embedded device compared to manual segmentation. DATA CONCLUSION: Proposed LSW-Net achieves high accuracy with fewer parameters than U-net and can be deployed on embedded device, facilitating wider application. EVIDENCE LEVEL: 2. TECHNICAL EFFICACY: 1.

Evaluation of Atlas-based auto-segmentation of liver in MR images for Yttrium-90 selective internal radiation therapy.

PURPOSE: The aim was to explore the feasibility of applying an atlas-based auto-segmentation tool, MIM Atlas Segment, for liver delineation in MR images in Y-90 selective internal radiation therapy (SIRT). MATERIALS AND METHODS: MR images of 41 liver patients treated with resin Y-90 SIRT were included: 20 patients' images were used to create an atlas, and the other 21 patients' images were used for testing. Auto-segmentation of liver in the MR images was performed with MIM Atlas Segment, and various settings for the auto-segmentation (i.e., with and without normalized deformable registration, single atlas-match and multi-atlas match, and multi-atlas match using different finalization methods) were tested. Auto-segmented liver contours were compared with physician manually-delineated contours, using Dice similarity coefficient (DSC) and mean distance to agreement (MDA). Ratio of volume (RV) and ratio of activity (RA) were calculated to further evaluate the auto-segmentation results. RESULTS: Auto-segmentations with normalized deformable registration generated better contours than those without normalized deformable registration. With normalized deformable registration, 3-atlas match using Majority Vote (MV) method generated better results than single-atlas match and 3-atlas match using STAPLE method, and generated similar results as 5-atlas match using MV method or STAPLE method. The average DSC, MDA, and RV of the contours generated with normalized deformable registration are 0.80-0.83, 0.60-0.67, and 0.91-1.00 cm, respectively. The average RA are 1.00-1.01, which indicate that the activities calculated using the auto-segmented liver contours are close to the accurate activities. CONCLUSION: The atlas-based auto-segmentation can be applied to generate initial liver contours in MR images for resin Y-90 SIRT, which can be used for activity calculations after physicians review.

Safety and efficacy of single insertion accelerated MR-image guided brachytherapy following chemo-radiation in locally advanced cervix cancer: modifying our EMBRACE during the COVID pandemic.

BACKGROUND: Utero-vaginal brachytherapy (BT) is an irreplaceable care component for the curative treatment of locally advanced cervix cancer (LACC). Magnetic Resonance Imaging (MRI)-image guided adaptive BT (IGABT) using the GYN-GEC-ESTRO EMBRACE guidelines is the international care standard. Usually following chemo-radiation therapy (CRT), IGABT has high proven utility in LACC but requires significant health system resources. Timely access was disrupted by the COVID-19 pandemic which challenged us to re-design our established IGABT care pathway. METHODS: From April 2020 consecutive patients with LACC were enrolled after CRT in a single arm exploratory non-inferiority study of a modified IGABT (mIGABT) protocol. This delivered an iso-effective IGABT dose (39.3 Gy: EQD2: α/ß10Gy concept) over a 24-h period during a single overnight hospitalisation. RESULTS: Fourteen LACC patients received mIGABT from April 2020 to March 2022. Median age was 62.5 years (37-82 years). LACC histology was primary squamous (9/14) or adeno-carcinoma (5/14). International Federation of Gynaecology and Obstetrics (FIGO) 2018 stages ranged from IB1/2 (N = 3), IIA1/IIB (5), IIIB (2), IIIC1/2 (4) with mean ± standard deviation (SD) gross tumour volume-at-diagnosis (GTV_D) of 37.7 cc ± 71.6 cc. All patients achieved complete metabolic, clinical, and cytologic cancer response with CRT and IGABT. High-risk HPV was cleared by 6-months. Complete MRI-defined cancer response before mIGABT (GTV_Fx1) was seen in 77% of cases (10/13). Only two women developed metastatic disease and one died at 12-months; 13 patients were alive without cancer at mean 20.3 ± 7.2 months follow-up. Actuarial 2-year overall survival was 93%. Compared with our pre-COVID IGABT program, overall mIGABT cost-saving in this cohort was USD 22,866. Prescribed dose covered at least 90% (D90) of the entire cervix and any residual cancer at time of BT (HRCTV_D90: high-risk clinical target volume) with 3-fractions of 8.5 Gy delivered over 24-h (22.8 ± 1.7 h). Total treatment time including CRT was 38 days. The mIGABT schedule was well tolerated and the entire cohort met EMBRACE recommended (EQD2: α/ß10Gy) combined HRCTV_D90 coverage of 87.5 ± 3.7 Gy. Similarly, organ-at-risk (OAR) median: interquartile range D2cc constraints (EQD2: α/ß3Gy) were EMBRACE compliant: bladder (65.9 Gy: 58.4-72.5 Gy), rectum (59.1 Gy: 55.7-61.8 Gy), and sigmoid colon (54.6 Gy: 50.3-58.9 Gy). ICRU recto-vaginal point dose was significantly higher (75.7 Gy) in our only case of severe (G4) pelvic toxicity. CONCLUSIONS: This study demonstrated the utility of mIGABT and VMAT CRT in a small cohort with LACC. Loco-regional control was achieved in all cases with minimal emergent toxicity. Single insertion mIGABT was logistically efficient, cost-saving, and patient-centric during the COVID-19 pandemic.

Double total variation (DTV) regularization and Improved adaptive moment estimation (IADAM) optimization method for fast MR image reconstruction.

BACKGROUND AND OBJECTIVE: Compressed sensing has been extensively studied as an advanced technique for fast MR image reconstruction. Current reconstruction algorithms often use total variation as the regularization term. Traditional total variation can easily lead to a staircase effect because it only pays attention to the variational information of the horizontal and vertical subbands. METHODS: In this paper, we propose a novel algorithm to reduce the staircase effect by increasing the variational information of the two diagonal subbands, which named Double Total Variation (DTV). We optimize the conjugate gradient algorithm by Improved Adaptive Moment Estimation (IADAM) as the solution algorithm. RESULTS: MR images of three body parts (head, knee and ankle) were used for simulations under different acceleration factor conditions. The conjugate gradient and fast conjugate gradient series algorithms were selected for comparison experiments. The results showed that the improved adaptive moment estimation conjugate gradient combined with DTV achieves the best reconstruction performance, therefore proved the superiority of DTV. After that, 64 different MR images of the three body parts were further simulated and the results demonstrated the general superiority from the proposed algorithm. CONCLUSIONS: The results of this study support that the proposed method may facilitate the development of the research field of image reconstruction algorithms and provide ideas for other algorithmic improvements.

Automatic Detection of Meniscus Tears Using Backbone Convolutional Neural Networks on Knee MRI.

BACKGROUND: Timely diagnosis of meniscus injuries is key for preventing knee joint dysfunction and improving patient outcomes because it decreases morbidity and facilitates treatment planning. PURPOSE: To train and evaluate a deep learning model for automated detection of meniscus tears on knee magnetic resonance imaging (MRI). STUDY TYPE: Bicentric retrospective study. SUBJECTS: In total, 584 knee MRI studies, divided among training (n = 234), testing (n = 200), and external validation (n = 150) data sets, were used in this study. The public data set MRNet was used as a second external validation data set to evaluate the performance of the model. SEQUENCE: A 3 T, coronal, and sagittal images from T1-weighted proton density (PD) fast spin-echo (FSE) with fat saturation and T2-weighted FSE with fat saturation sequences. ASSESSMENT: The detection system for meniscus tear was based on the improved YOLOv4 model with Darknet-53 as the backbone. The performance of the model was also compared with that of three radiologists of varying levels of experience. The determination of the presence of a meniscus tear from surgery reports was used as the ground truth for the images. STATISTICAL TESTS: Sensitivity, specificity, prevalence, positive predictive value, negative predictive value, accuracy, and receiver operating characteristic curve were used to evaluate the performance of the detection model. Two-way analysis of variance, Wilcoxon signed-rank test, and Tukey's multiple tests were used to evaluate differences in performance between the model and radiologists. RESULTS: The overall accuracies for detecting meniscus tears using our model on the internal testing, internal validation, and external validation data sets were 95.4%, 95.8%, and 78.8%, respectively. One radiologist had significantly lower performance than our model in detecting meniscal tears (accuracy: 0.9025 ± 0.093 vs. 0.9580 ± 0.025). DATA CONCLUSION: The proposed model had high sensitivity, specificity, and accuracy for detecting meniscus tears on knee MRIs. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.

Magnetic Resonance Imaging Findings in Patients with Tarsal Tunnel Syndrome.

Tarsal tunnel syndrome (TTS) is a common entrapment syndrome whose diagnosis can be difficult. We compared preoperative magnetic resonance imaging (MRI) and operative findings in 23 consecutive TTS patients (28 sides) whose mean age was 74.5 years. The 1.5T MRI sequence was 3D T2* fat suppression. We compared the MRI findings with surgical records and intraoperative videos to evaluate them. MRI- and surgical findings revealed that a ganglion was involved on one side (3.6%), and the other 27 sides were diagnosed with idiopathic TTS. MRI visualized the nerve compression point on 23 sides (82.1%) but failed to reveal details required for surgical planning. During surgery of the other five sides (17.9%), three involved varices, and on one side each, there was connective tissue entrapment or nerve compression due to small vascular branch strangulation. MRI studies were useful for nerve compression due to a mass lesion or idiopathic factors. Although MRI revealed the compression site, it failed to identify the specific involvement of varices and small vessel branches and the presence of connective tissue entrapment.