T2-weighted Magnetic Resonance Imaging Images Research Articles

Role of Artificial intelligence model in prediction of low back pain using T2 weighted MRI of Lumbar spine

Background Low back pain (LBP), the primary cause of disability, is the most common musculoskeletal disorder globally and the primary cause of disability. Magnetic resonance imaging (MRI) studies are inconclusive and less sensitive for identifying and classifying patients with LBP. Hence, this study aimed to investigate the role of artificial intelligence (AI) models in the prediction of LBP using T2 weighted MRI image of the lumbar spine. Methods This was a prospective case-control study. A total of 200 MRI patients (100 cases and controls each) referred for lumbar spine and whole spine screening were included. The scans were performed using 3.0 Tesla MRI (United Imaging Healthcare). T2 weighted images of the lumbar spine were segmented to extract radiomic features. Machine learning (ML) models, such as random forest, decision tree, logistic regression, K-nearest neighbors, adaboost, and deep learning methods (DL), such as ResNet and GoogleNet, were used, and performance measures were calculated. Results Our study showed that Random forest and AdaBoost are the most reliable ML models for predicting LBP. Random forest showed high performance with area under curve (AUC) values from 0.83 to 0.88 across all lumbar vertebrae and L2-L3, L3-L4, and L4-L5 intervertebral discs (IVDs), with AUCs of 0.88 the highest at L5-S1 IVD (0.92). Adaboost demonstrated high performance at the L2-L5 vertebrae with AUC values of 0.82 to 0.90, with the highest AUC (0.97) at the L5-S1 IVD. Among the DL models, GoogleNet outperformed the other models at 30 epochs with an accuracy of 0.85, followed by ResNet 18 (30 epochs) with an accuracy of 0.84. Conclusion The study demonstrated that ML and DL models can effectively predict LBP from MRI T2 weighted image of the lumbar spine. ML and DL models could also enhance the diagnostic accuracy of LBP, potentially leading to better patient management and outcomes.

Application of deep learning reconstruction in abdominal magnetic resonance cholangiopancreatography for image quality improvement and acquisition time reduction

PurposeTo compare deep learning (DL)-based and conventional reconstruction through subjective and objective analysis and ascertain whether DL-based reconstruction improves the quality and acquisition speed of clinical abdominal magnetic resonance imaging (MRI). MethodsThe 124 patients who underwent abdominal MRI between January and July 2021 were retrospectively studied. For each patient, two-dimensional axial T2-weighted single-shot fast spin-echo MRI images with or without fat saturation were reconstructed using DL-based and conventional methods. The subjective image quality scores and objective metrics, including signal-to-noise ratios (SNRs) and contrast-to-noise ratios (CNRs) of the images were analysed. An explorative analysis was performed to compare 20 patients’ MRI images with site routine settings, high-resolution settings and high-speed settings. Paired t tests and Wilcoxon signed-rank tests were used for subjective and objective comparisons. ResultsA total of 144 patients were evaluated (mean age, 62.2 ± 14.1 years; 83 men). The MRI images reconstructed using DL-based methods had higher SNRs and CNRs than did those reconstructed using conventional methods (all p < 0.01). The subjective scores of the images reconstructed using DL-based methods were higher than those of the images reconstructed using conventional methods (p < 0.01), with significantly lower variation (p < 0.01). Exploratory analysis revealed that the DL-based reconstructions with thin slice thickness and higher temporal resolution had the highest image quality and were associated with the shortest scan times. ConclusionsDL-based reconstruction methods can be used to improve the quality with higher stability and accelerate the acquisition of abdominal MRI.

The Utility of Magnetic Resonance Imaging-based Vertebral Bone Quality Scores as a Predictor of Cage Subsidence Following Transforaminal and Posterior Lumbar Interbody Fusion.

Retrospective study. The objectives were to determine whether vertebral bone quality (VBQ) scores are associated with interbody cage subsidence following transforaminal (TLIF) and posterior (PLIF) lumbar interbody fusions and whether there is a clinically sensitive threshold for subsidence. Interbody cage subsidence following lumbar fusion is a complication that can generate poor surgical outcomes. Prior research has correlated cage subsidence with bone mineral density. VBQ scores derived from magnetic resonance imaging (MRI) have been proposed as a tool for measuring bone mineral density, offering a potential new and convenient preoperative risk assessment tool for subsidence. The study involved patients undergoing single-level PLIF or TLIF between 2007 and 2022. Exclusions were for nondegenerative diagnoses, multilevel/revision surgeries, inadequate radiographs, missing immediate postoperative radiographs, and preoperative MRI studies older than 1 year. VBQ was calculated at L1-L4 from preoperative T1-weighted MRI images. Subsidence was assessed by changes in disc height (DH; >2mm difference) and segmental lordosis (SL; >5 degrees difference) between immediate weight-bearing postoperative and latest postoperative lateral radiographs. Statistical analysis included descriptive and inferential statistics. Subsidence was observed in 27% (SL parameter) and 47% (DH parameter) of 51 total patients. VBQ scores were significantly associated with cage subsidence based on both SL (odds ratio = 7.750, P = 0.012; correlation coefficient = 0.382, P = 0.006) and DH (odds ratio = 4.074, P = 0.026; correlation coefficient = 0.258, P = 0.057) in the combined TLIF/PLIF cohorts. In the cohort of 36 patients undergoing TLIF, a VBQ of 2.70 yielded 100.0% sensitivity and 46.2% specificity in detecting subsidence with SL measurement (area under the curve = 0.812, P < 0.001) and 86.7% sensitivity and 47.6% specificity with the DH measurement (area under the curve = 0.692, P = 0.033). We found that MRI-based VBQ scores are effective predictors of cage subsidence following TLIF surgery. A VBQ score of 2.70 demonstrated a reliable model and high sensitivity for doing so, identifying a potential clinical threshold for preoperative subsidence risk assessment. Level III.

Role of Artificial intelligence model in prediction of low back pain using T2 weighted MRI of Lumbar spine

Background Low back pain (LBP), the primary cause of disability, is the most common musculoskeletal disorder globally and the primary cause of disability. Magnetic resonance imaging (MRI) studies are inconclusive and less sensitive for identifying and classifying patients with LBP. Hence, this study aimed to investigate the role of artificial intelligence (AI) models in the prediction of LBP using T2 weighted MRI image of the lumbar spine. Methods This was a prospective case-control study. A total of 200 MRI patients (100 cases and controls each) referred for lumbar spine and whole spine screening were included. The scans were performed using 3.0 Tesla MRI (United Imaging Healthcare). T2 weighted images of the lumbar spine were segmented to extract radiomic features. Machine learning (ML) models, such as random forest, decision tree, logistic regression, K-nearest neighbors, adaboost, and deep learning methods (DL), such as ResNet and GoogleNet, were used, and performance measures were calculated. Results Our study showed that Random forest and AdaBoost are the most reliable ML models for predicting LBP. Random forest showed high performance with area under curve (AUC) values from 0.83 to 0.88 across all lumbar vertebrae and L2-L3, L3-L4, and L4-L5 intervertebral discs (IVDs), with AUCs of 0.88 the highest at L5-S1 IVD (0.92). Adaboost demonstrated high performance at the L2-L5 vertebrae with AUC values of 0.82 to 0.90, with the highest AUC (0.97) at the L5-S1 IVD. Among the DL models, GoogleNet outperformed the other models at 30 epochs with an accuracy of 0.85, followed by ResNet 18 (30 epochs) with an accuracy of 0.84. Conclusion The study demonstrated that ML and DL models can effectively predict LBP from MRI T2 weighted image of the lumbar spine. ML and DL models could also enhance the diagnostic accuracy of LBP, potentially leading to better patient management and outcomes.

A parasitic leiomyoma of the sigmoid mesentery with schwannoma-like image findings

BackgroundParasitic leiomyoma (PL) consists of uterine fibroids separate from the uterus that grow in extrauterine tissues such as the peritoneum and mesenterium. The diagnosis of PL requires a thorough medical history of laparoscopic myomectomies using a morcellator and the identification of typical magnetic resonance imaging (MRI) findings as uterine fibroids. Imaging diagnosis of PL is occasionally difficult when PL degenerates in various ways, owing to atypical findings on computed tomography (CT) and MRI.Case presentationA 29-year-old woman with a history of laparoscopic myomectomy visited a local hospital with lower abdominal pain. A mesenteric tumor on the sigmoid mesentery was suspected on MRI, and she was referred to our hospital. CT scan showed strong early contrast uptake in the center of the tumor, and MRI T2-weighted images showed high signals at the tumor margins and low signals in the center, suggesting a schwannoma. PL was also part of the differential diagnosis because of the patient’s history of laparoscopic myomectomy. With a preoperative diagnosis of a sigmoid colon mesenteric tumor undeniably of malignant origin, laparoscopic resection of the sigmoid mesenteric tumor was performed. Histopathological examination revealed it to be a PL.ConclusionsWe report a case of PL of the sigmoid mesentery with schwannoma-like findings on imaging that was treated laparoscopically. PL is sometimes difficult to distinguish from schwannomas because of the variety of imaging findings, such as uterine fibroids. PL should be considered in the differential diagnosis of mesenteric tumors following laparoscopic myomectomies, even if it does not show typical imaging findings, such as uterine fibroids.

Difference in Laterality of the Dorsal Striatum in Schizoaffective Disorder.

Recent research has demonstrated that the dorsal striatum is directly associated with the integration of cognitive, sensory-motor, and motivational/emotional data. Disruptions in the corticostriatal circuit have been implicated in the pathophysiology of psychosis. The dorsal striatum was reported to show lateralized pathology in psychotic disorders. In this study, we aimed to analyze the laterality of the dorsal striatum with texture analysis of T2-weighted magnetic resonance imaging (MRI) images from schizoaffective disorder (SAD) patients. Twenty SAD patients, met the inclusion criteria and had available cranial MRI data were assigned as the patient group. Twenty healthy individuals were determined as the control group. Texture analysis values were obtained from striatum region of interests (ROI) generated from T2-weighted MRI images. Data are presented as mean and standard deviation. The suitability of the data for normal distribution was analyzed with the Kolmogorov-Smirnov test. Analysis of variance (ANOVA) test (Post Hoc TUKEY) was employed to compare the group data based on test findings. There was no significant difference between the groups in terms of gender and age. There were differences in the values of texture analysis parameters of both caudate and putamen nuclei in comparison to controls. We identified differences in the left dorsal striatum nuclei in SAD. The differences in the putamen were more and more pronounced than in the caudate. Texture analyses suggest that the left dorsal striatum nuclei may be different in SAD patients. Further studies are needed to determine the pathophysiology of SAD and how it may affect disease treatment.

MRI T1-Weighted Fat Segmentation Based on Active Contour for Osteosarcoma Patients

Osteosarcoma (OS) is the most common and frequent primary bone cancer in children and adolescents. This type of bone cancer often developed at the long bones’ extremities close to the metaphyseal growth plates. To address these issues, OS quantitative analysis is performed using Magnetic Resonance Imaging (MRI) to plan surgical procedures and track the effectiveness of treatment. Fat suppression is commonly employed to eliminate the fat signal from T1-Weighted MRI sequence. The suppressed fat signal in T1-Weighted is important and often used to identify abnormalities in other MRI sequences. In clinical, MRI images must be interpreted by a radiologist. Manually outlining the tumour location is laborious and subjective. Image processing involves segmentation to separate information from the desired target region of the image. Active Contour (AC) is an algorithm often used in medical images for segmentation. Nevertheless, the process of mask initialization can be challenging and requires careful attention. This study proposes a method to define the mask initialization so that the AC has a starting point to extract fat in T1-Weighted for the purpose of fat suppression. From the results obtained, AC with the proposed mask initialization method successfully segmented the fat in T1-Weighted MRI images with accuracy, precision, recall and F1-score of 0.92, 0.86, 0.88 and 0.89 respectively.

Integrating MRI-based radiomics and clinicopathological features for preoperative prognostication of early-stage cervical adenocarcinoma patients: in comparison to deep learning approach

ObjectivesThe roles of magnetic resonance imaging (MRI) -based radiomics approach and deep learning approach in cervical adenocarcinoma (AC) have not been explored. Herein, we aim to develop prognosis-predictive models based on MRI-radiomics and clinical features for AC patients.MethodsClinical and pathological information from one hundred and ninety-seven patients with cervical AC was collected and analyzed. For each patient, 107 radiomics features were extracted from T2-weighted MRI images. Feature selection was performed using Spearman correlation and random forest (RF) algorithms, and predictive models were built using support vector machine (SVM) technique. Deep learning models were also trained with T2-weighted MRI images and clinicopathological features through Convolutional Neural Network (CNN). Kaplan-Meier curve was analyzed using significant features. In addition, information from another group of 56 AC patients was used for the independent validation.ResultsA total of 107 radiomics features and 6 clinicopathological features (age, FIGO stage, differentiation, invasion depth, lymphovascular space invasion (LVSI), and lymph node metastasis (LNM) were included in the analysis. When predicting the 3-year, 4-year, and 5-year DFS, the model trained solely on radiomics features achieved AUC values of 0.659 (95%CI: 0.620–0.716), 0.791 (95%CI: 0.603–0.922), and 0.853 (95%CI: 0.745–0.912), respectively. However, the combined model, incorporating both radiomics and clinicopathological features, outperformed the radiomics model with AUC values of 0.934 (95%CI: 0.885–0.981), 0.937 (95%CI: 0.867–0.995), and 0.916 (95%CI: 0.857–0.970), respectively. For deep learning models, the MRI-based models achieved an AUC of 0.857, 0.777 and 0.828 for 3-year DFS, 4-year DFS and 5-year DFS prediction, respectively. And the combined deep learning models got a improved performance, the AUCs were 0.903. 0.862 and 0.969. In the independent test set, the combined model achieved an AUC of 0.873, 0.858 and 0.914 for 3-year DFS, 4-year DFS and 5-year DFS prediction, respectively.ConclusionsWe demonstrated the prognostic value of integrating MRI-based radiomics and clinicopathological features in cervical adenocarcinoma. Both radiomics and deep learning models showed improved predictive performance when combined with clinical data, emphasizing the importance of a multimodal approach in patient management.

An Implementation of Effective CNN Model for AD Detection

This paper focuses on detecting Alzheimer’s Disease (AD). The most usual form of dementia is Alzheimer's disease, which causes permanent cause memory cell damage. Alzheimer's disease, a neurodegenerative disease, increases slowly over time. For this matter, early detection of Alzheimer's disease is important. The purpose of this work is using Magnetic Resonance Imaging (MRI) to diagnose AD. A Convolution Neural Network (CNN) model, Reset, and VGG the pre-trained learning models are used. Performing analysis and validation of layers affects the effectiveness of the model. T1-weighted MRI images are taken for preprocessing from ADNI. The Dataset images are taken from the Alzheimer's Disease Neuroimaging Initiative (ADNI). 3D MRI scans into 2D image slices shows the optimization method in the training process while achieving 96% and 94% accuracy in VGG 16 and ResNet 18 respectively. This study aims to classify AD from brain 3D MRI images and obtain better results.

Unexpected change in hydrogel spacer volume during external-beam radiation therapy.

To reduce the rectal radiation dose during local radiation therapy of prostate cancer, a hydrogel spacer is typically implanted between the prostate and rectum. However, the spacer volume can change during external beam radiation therapy (EBRT). Therefore, we used magnetic resonance imaging (MRI) to determine changes in the spacer volume during EBRT and analyzed the data to identify patient factors influencing this change. A hydrogel spacer was implanted in each enrolled patient diagnosed with prostate cancer (n = 22, age = 69-86years) for EBRT with a total dose of 70Gy over 35 fractions. T2-weighted MRI images were acquired before (median = 8days) and during EBRT, when the radiation dose of 48Gy (median) was given at 55 days (median) after implantation. MRI images were used to determine the spacer volume as well as the maximum and minimum distances between the prostate and anterior wall of the rectum at the middle height of the prostate. Scatterplots were created to determine whether correlations existed between changes in the spacer volume and these two distances, while uni- and multivariate analyses were conducted to determine if the spacer volume change was influenced by the following patient factors: age, body mass index, estimated glomerular filtration rate, and visceral fat areas at the umbilical and femoral head positions. The spacer volume increased in all 22 patients, with the smaller spacer volume before EBRT increasing by a larger amount during EBRT. This increase in the spacer volume was unaffected by other patient factors. However, it correlated with the change in the maximum distance between the prostate and anterior wall of the rectum. To avoid adverse changes in the rectal radiation dose during EBRT, hydrogel spacer volume should be monitored, especially if the pre-EBRT volume is small.

Photoelectric active copper/iron porphyrin-based metal–organic framework embedded with ferric oxide nanocrystals for photoelectrochemical aptasensing miRNA-141 and in vivo tumor magnetic resonance imaging

A heterostructure of bimetallic copper/ferric metal–organic framework embedded with ferric oxide (denoted as Fe2O3@CuFe-MOF) nanocrystals was simultaneously employed as the bioplatform for constructing the photoelectrochemical (PEC) aptasensor to detect miRNA-141 and an excellent contrast agent for in vivo diagnosis cancer via magnetic resonance imaging (MRI). Fe2O3@CuFe-MOF was synthesized by linking 5,10,15,20-tetra(4-pyridyl)porphyrin (TPyP) with Cu2+ and Fe2+ ions, while partial Fe2+ ions were oxidized to Fe2O3 nanocrystals and embraced within CuFe-MOF (denoted as Fe2O3@CuFe-TPyP-MOF). Comparing with Cu-TPyP-MOF and Fe2O3@Fe-TPyP-MOF, Fe2O3@CuFe-TPyP-MOF exhibited enhanced separation ability of photocarriers, promoted electron transfer, larger specific surface area, and richer functionality, rendering boosted photoelectric conversion efficiency and superior PEC performance. The Fe2O3@CuFe-TPyP-MOF-based PEC aptasensor thus illustrated an ultralow limit of detection (0.8 fM) toward miRNA-141 within a wide range from 1.0 fM to 1.0 nM, accompanying with other outperformed aptasensing ability. In addition, due to the paramagnetic property of Cu2+ and Fe2+ ions, Fe2O3@CuFe-TPyP-MOF also afforded the promoted T1-weighted MRI imaging ability, showing the potential for the diagnosis of prostate tumor in vivo. This work provides a robust photoelectronic active MOF-based heterostructure for the sensitive detection and early cancer diagnosis.

MRI-based pedicle bone quality score: correlation to quantitative computed tomography bone mineral density and its role in quantitative assessment of osteoporosis

BACKGROUND CONTEXTBone quality in the pedicle region generally determines screw pullout strength, insertion torque, and vertebral body loading characteristics. Dual-energy X-ray absorptiometry (DEXA), as the gold standard for evaluating bone mineral density (BMD), cannot measure the BMD of specific parts, such as pedicle, and DEXA is limited in many ways. Recent studies have shown a correlation between the magnetic resonance imaging (MRI)-based vertebral bone quality (VBQ) score and BMD measured using DEXA or quantitative computed tomography (QCT). However, no studies have been reported on the MRI-based pedicle bone quality (PBQ) score. Moreover, few studies have investigated the relationship between MRI-based PBQ and osteoporosis. PURPOSETo create a new site-specific MRI-based PBQ assessment method and assess its diagnostic capacity in patients with normal BMD and osteopenia/osteoporosis. STUDY DESIGN/SETTINGA retrospective study. PATIENT SAMPLEA total of 156 patients underwent lumbar fusion surgery for chronic low back pain at our hospital between 2021 and 2022, with lumbar QCT and T1-weighted MRI performed before surgery. OUTCOME MEASURESCorrelation of the PBQ score with QCT BMD, and the association between the PBQ score and presence of osteopenia/osteoporosis. METHODSBMD of the lumbar was calculated as the mean BMD of the L1 and L2 vertebral bodies on the basis of asynchronous QCT measurements. The PBQ score, which is the average of the bone quality values of both pedicles on the basis of site-specific T1-weighted sagittal MRI images, was calculated by dividing the median signal intensity of the L1–L4 pedicles by the signal intensity of the cerebrospinal fluid at the L3 level. The interobserver reliability of the PBQ score was assessed using the intraclass correlation coefficient (ICC). A receiver operating characteristic curve was drawn, and the area under the curve (AUC) was calculated to assess the predictive performance of PBQ for osteoporosis. The PBQ score was compared with QCT BMD, as the gold standard, using Pearson correlation analysis. RESULTSIn total, 156 patients participated in this study, including 51 in the Normal BMD group and 105 in the osteopenia/osteoporosis group. The PBQ score in the osteopenia/osteoporosis group was significantly higher than that in the normal BMD group (3.19±0.55 vs. 2.84±0.51, p<.001). The VBQ and PBQ scores were calculated by 2 authors and were in good agreement (intraclass correlation coefficient=0.949 and 0.929, respectively). Pearson's test showed a significant negative correlation between PBQ and QCT BMD (r=−0.4887, p<.001). The optimal cutoff PBQ score to differentiate patients with osteopenia/osteoporosis from those with normal BMD was 3.160, with a sensitivity of 66.7%, specificity of 72.5%, and AUC of 0.776. The PBQ score correlated more strongly with QCT BMD (r=−0.4887) than VBQ (r=−0.4078). CONCLUSIONSIn this study, we propose a novel, MRI-based pedicle-specific bone quality score. This is the first study to investigate the relationship between the PBQ score and QCT BMD. The PBQ score showed diagnostic utility, differentiating between patients with osteopenia/osteoporosis and those with normal BMD (AUC=0.776), and the PBQ score correlated more strongly with QCT BMD than VBQ.

Synthesis, MTT assay, 99m-Technetium radiolabeling, biodistribution evaluation of radiotracer and in vitro magnetic resonance imaging study of P,N-doped graphene quantum dots as a new multipurpose imaging nano-agent

Abstract In this study, a new nano-structure, N,P-doped graphene quantum dots (N,P-GQDs), were synthesized as multipurpose imaging agent for performing scintigraphy and magnetic resonance imaging (MRI). Some standard characterization methods were used to identify the nano-structure. In vitro cytotoxicity evaluation using MTT assay revealed that N,P-GQDs nanoparticles had no significant cytotoxicity after 24 and 48 h against normal (MCF-10A) and cancerous (MCF 7) human breast cell line in concentration up to 200 μg/mL. The N,P-GQDs were radiolabeled with Technetium-99m as 99mTc-(N,P-GQDs) and the radiochemical purity was assayed by ITLC concluding RCP ≥ 95 %. The passing of 99mTc-(N,P-GQDs) through 0.1 µm filter demonstrated that 70.8 % of particles were &lt;0.1 µm. In order to perform scintigraphy, the 99mTc-(N,P-GQDs) were injected to female healthy Wistar rats. The results showed that the radio-complex was captured and eliminated just by kidneys. Moreover, in vitro T1-weighted phantom MRI imaging showed that the N,P-GQDs have proper relaxivity in comparison to Dotarem® as a clinically available contrast agent. The results showed that the N,P-GQDs have potential to be considered as a novel and encouraging agent for both molecular MRI and nuclear medicine imagings.

Comparison of MRI Sequences to Predict IDH Mutation Status in Gliomas Using Radiomics-Based Machine Learning.

Regarding the 2021 World Health Organization (WHO) classification of central nervous system (CNS) tumors, the isocitrate dehydrogenase (IDH) mutation status is one of the most important factors for CNS tumor classification. The aim of our study is to analyze which of the commonly used magnetic resonance imaging (MRI) sequences is best suited to obtain this information non-invasively using radiomics-based machine learning models. We developed machine learning models based on different MRI sequences and determined which of the MRI sequences analyzed yields the highest discriminatory power in predicting the IDH mutation status. In our retrospective IRB-approved study, we used the MRI images of 106 patients with histologically confirmed gliomas. The MRI images were acquired using the T1 sequence with and without administration of a contrast agent, the T2 sequence, and the Fluid-Attenuated Inversion Recovery (FLAIR) sequence. To objectively compare performance in predicting the IDH mutation status as a function of the MRI sequence used, we included only patients in our study cohort for whom MRI images of all four sequences were available. Seventy-one of the patients had an IDH mutation, and the remaining 35 patients did not have an IDH mutation (IDH wild-type). For each of the four MRI sequences used, 107 radiomic features were extracted from the corresponding MRI images by hand-delineated regions of interest. Data partitioning into training data and independent test data was repeated 100 times to avoid random effects associated with the data partitioning. Feature preselection and subsequent model development were performed using Random Forest, Lasso regression, LDA, and Naïve Bayes. The performance of all models was determined with independent test data. Among the different approaches we examined, the T1-weighted contrast-enhanced sequence was found to be the most suitable for predicting IDH mutations status using radiomics-based machine learning models. Using contrast-enhanced T1-weighted MRI images, our seven-feature model developed with Lasso regression achieved a mean area under the curve (AUC) of 0.846, a mean accuracy of 0.792, a mean sensitivity of 0.847, and a mean specificity of 0.681. The administration of contrast agents resulted in a significant increase in the achieved discriminatory power. Our analyses show that for the prediction of the IDH mutation status using radiomics-based machine learning models, among the MRI images acquired with the commonly used MRI sequences, the contrast-enhanced T1-weighted images are the most suitable.

Clinical analysis of uterine parameters evaluated by preoperative magnetic resonance imaging in patients treated by hysteroscopic approach with previous cesarean scar defect-related abnormal uterine bleeding: a retrospective cohort study.

As the cesarean delivery rate continues to rise globally, the treatment of previous cesarean scar defects (PCSD) remains challenging. This study aimed to analyze the variables that may influence the clinical cure rate of patients with PCSD-related abnormal uterine bleeding (AUB) as determined by preoperative magnetic resonance imaging (MRI) following hysteroscopic therapy. Women who underwent hysteroscopic surgery for PCSD-related AUB at the Gynecology Department of Third Xiangya Hospital of Central South University from 2018 to 2022 were recruited to this retrospective cohort investigation. A total of 147 patients were enrolled in this study and underwent follow-up over 6 months. The significance of clinical characteristics linked to the clinical cure rate of AUB was examined by logistic regression. There were 64 clinically cured (43.5%) and 83 non-clinically cured (56.5%) patients in the study. There were no significant differences in the age, menstrual duration, gravidity, parity, number of cesarean sections, time since the previous cesarean section, uterus position, width, depth, and thickness of the remaining muscle layer of the defect by MRI T2-weighted images (T2WI) before hysteroscopic surgery between the 2 groups. MRI T2WI of the myometrial thickness adjacent to the defect [P=0.038, odds ratio (OR) =2.095, 95% confidence interval (CI): 1.047-4.261] and the distance from the defect to the external cervical os (P=0.021, OR =2.254, 95% CI: 1.136-4.540) before hysteroscopic surgery are risk factors for the clinical cure rate. The myometrial thickness adjacent to the defect and the distance from the defect to the external cervical os in preoperative MRI are risk factors for clinical cure rate in patients with PCSD-related AUB after hysteroscopic treatment, which is helpful for evaluating the prognosis of disease.

Pixelwise Gradient Model with GAN for Virtual Contrast Enhancement in MRI Imaging.

Background: The development of advanced computational models for medical imaging is crucial for improving diagnostic accuracy in healthcare. This paper introduces a novel approach for virtual contrast enhancement (VCE) in magnetic resonance imaging (MRI), particularly focusing on nasopharyngeal cancer (NPC). Methods: The proposed model, Pixelwise Gradient Model with GAN for Virtual Contrast Enhancement (PGMGVCE), makes use of pixelwise gradient methods with Generative Adversarial Networks (GANs) to enhance T1-weighted (T1-w) and T2-weighted (T2-w) MRI images. This approach combines the benefits of both modalities to simulate the effects of gadolinium-based contrast agents, thereby reducing associated risks. Various modifications of PGMGVCE, including changing hyperparameters, using normalization methods (z-score, Sigmoid and Tanh) and training the model with T1-w or T2-w images only, were tested to optimize the model's performance. Results: PGMGVCE demonstrated a similar accuracy to the existing model in terms of mean absolute error (MAE) (8.56 ± 0.45 for Li's model; 8.72 ± 0.48 for PGMGVCE), mean square error (MSE) (12.43 ± 0.67 for Li's model; 12.81 ± 0.73 for PGMGVCE) and structural similarity index (SSIM) (0.71 ± 0.08 for Li's model; 0.73 ± 0.12 for PGMGVCE). However, it showed improvements in texture representation, as indicated by total mean square variation per mean intensity (TMSVPMI) (0.124 ± 0.022 for ground truth; 0.079 ± 0.024 for Li's model; 0.120 ± 0.027 for PGMGVCE), total absolute variation per mean intensity (TAVPMI) (0.159 ± 0.031 for ground truth; 0.100 ± 0.032 for Li's model; 0.153 ± 0.029 for PGMGVCE), Tenengrad function per mean intensity (TFPMI) (1.222 ± 0.241 for ground truth; 0.981 ± 0.213 for Li's model; 1.194 ± 0.223 for PGMGVCE) and variance function per mean intensity (VFPMI) (0.0811 ± 0.005 for ground truth; 0.0667 ± 0.006 for Li's model; 0.0761 ± 0.006 for PGMGVCE). Conclusions: PGMGVCE presents an innovative and safe approach to VCE in MRI, demonstrating the power of deep learning in enhancing medical imaging. This model paves the way for more accurate and risk-free diagnostic tools in medical imaging.

MRI-based prostate cancer classification using 3D efficient capsule network.

Prostate cancer (PCa) is the most common cancer in men and the second leading cause of male cancer-related death. Gleason score (GS) is the primary driver of PCa risk-stratification and medical decision-making, but can only be assessed at present via biopsy under anesthesia. Magnetic resonance imaging (MRI) is a promising non-invasive method to further characterize PCa, providing additional anatomical and functional information. Meanwhile, the diagnostic power of MRI is limited by qualitative or, at best, semi-quantitative interpretation criteria, leading to inter-reader variability. Computer-aided diagnosis employing quantitative MRI analysis has yielded promising results in non-invasive prediction of GS. However, convolutional neural networks (CNNs) do not implicitly impose a frame of reference to the objects. Thus, CNNs do not encode the positional information properly, limiting method robustness against simple image variations such as flipping, scaling, or rotation. Capsule network (CapsNet) has been proposed to address this limitation and achieves promising results in this domain. In this study, we develop a 3D Efficient CapsNet to stratify GS-derived PCa risk using T2-weighted (T2W) MRI images. In our method, we used 3D CNN modules to extract spatial features and primary capsule layers to encode vector features. We then propose to integrate fully-connected capsule layers (FC Caps) to create a deeper hierarchy for PCa grading prediction. FC Caps comprises a secondary capsule layer which routes active primary capsules and a final capsule layer which outputs PCa risk. To account for data imbalance, we propose a novel dynamic weighted margin loss. We evaluate our method on a public PCa T2W MRI dataset from the Cancer Imaging Archive containing data from 976 patients. Two groups of experiments were performed: (1) we first identified high-risk disease by classifying low + medium risk versus high risk; (2) we then stratified disease in one-versus-one fashion: low versus high risk, medium versus high risk, and low versus medium risk. Five-fold cross validation was performed. Our model achieved an area under receiver operating characteristic curve (AUC) of 0.83 and 0.64 F1-score for low versus high grade, 0.79 AUC and 0.75 F1-score for low + medium versus high grade, 0.75 AUC and 0.69 F1-score for medium versus high grade and 0.59 AUC and 0.57 F1-score for low versus medium grade. Our method outperformed state-of-the-art radiomics-based classification and deep learning methods with the highest metrics for each experiment. Our divide-and-conquer strategy achieved weighted Cohen's Kappa score of 0.41, suggesting moderate agreement with ground truth PCa risks. In this study, we proposed a novel 3D Efficient CapsNet for PCa risk stratification and demonstrated its feasibility. This developed tool provided a non-invasive approach to assess PCa risk from T2W MR images, which might have potential to personalize the treatment of PCa and reduce the number of unnecessary biopsies.

High-resolution MRI synthesis using a data-driven framework with denoising diffusion probabilistic modeling

Objective. High-resolution magnetic resonance imaging (MRI) can enhance lesion diagnosis, prognosis, and delineation. However, gradient power and hardware limitations prohibit recording thin slices or sub-1 mm resolution. Furthermore, long scan time is not clinically acceptable. Conventional high-resolution images generated using statistical or analytical methods include the limitation of capturing complex, high-dimensional image data with intricate patterns and structures. This study aims to harness cutting-edge diffusion probabilistic deep learning techniques to create a framework for generating high-resolution MRI from low-resolution counterparts, improving the uncertainty of denoising diffusion probabilistic models (DDPM). Approach. DDPM includes two processes. The forward process employs a Markov chain to systematically introduce Gaussian noise to low-resolution MRI images. In the reverse process, a U-Net model is trained to denoise the forward process images and produce high-resolution images conditioned on the features of their low-resolution counterparts. The proposed framework was demonstrated using T2-weighted MRI images from institutional prostate patients and brain patients collected in the Brain Tumor Segmentation Challenge 2020 (BraTS2020). Main results. For the prostate dataset, the bicubic interpolation model (Bicubic), conditional generative-adversarial network (CGAN), and our proposed DDPM framework improved the noise quality measure from low-resolution images by 4.4%, 5.7%, and 12.8%, respectively. Our method enhanced the signal-to-noise ratios by 11.7%, surpassing Bicubic (9.8%) and CGAN (8.1%). In the BraTS2020 dataset, the proposed framework and Bicubic enhanced peak signal-to-noise ratio from resolution-degraded images by 9.1% and 5.8%. The multi-scale structural similarity indexes were 0.970 ± 0.019, 0.968 ± 0.022, and 0.967 ± 0.023 for the proposed method, CGAN, and Bicubic, respectively. Significance. This study explores a deep learning-based diffusion probabilistic framework for improving MR image resolution. Such a framework can be used to improve clinical workflow by obtaining high-resolution images without penalty of the long scan time. Future investigation will likely focus on prospectively testing the efficacy of this framework with different clinical indications.

The Success of Deep Learning Modalities in Evaluating Modic Changes

BackgroundModic changes are pathologies that are common in the population and cause low back pain. The aim of the study is to analyze the modic changes detected in magnetic resonance imaging (MRI) using deep learning modalities. MethodsThe sagittal T1, sagittal and axial T2-weighted lumbar MRI images of 307 patients, of which 125 were female and 182 were male, aged 19-86 years, who underwent MRI examination between 2016-2021 were analyzed. Modic changes (MC) were categorized and marked according to signal changes. Our study consists of two independent stages: classification and segmentation. The categorized data was first classified using convolutional neural network (CNN) architectures such as DenseNet-121, DenseNet-169, and VGG-19. In the next stage, masks were removed by segmentation using U-Net, which is the CNN architecture, with image processing programs on the marked images. ResultsDuring the classification stage, the success rates for modic type 1, type 2, and type 3 changes were 98%, 96%, 100% in DenseNet-121, 100%, 94%, 100% in DenseNet-169, and 98%, 92%, 97% in VGG-19, respectively. At the segmentation phase, the success rate was 71% with the U-Net architecture. ConclusionEvaluation of MRI findings of MC in the etiology of lower back pain with deep learning architectures can significantly reduce the workload of the radiologist by providing ease of diagnosis.

Association between paraspinal muscle fat infiltration and regional kyphosis angle in thoracolumbar fracture patients: a retrospective study

This study aims to evaluate the impact of percutaneous pedicle screw fixation (PPSF) and open pedicle screw fixation (OPSF) on the postoperative paraspinal muscle fat infiltration (FI) rate in patients with thoracolumbar fractures through magnetic resonance imaging (MRI), and explore the association between paraspinal muscle FI rate and regional kyphosis angle. We retrospectively analyzed clinical data from 35 patients who underwent either PPSF or OPSF for thoracolumbar fractures, examining data at preoperative, 1-month postoperative, and 9-months postoperative time points, which included Visual Analog Scale (VAS), Oswestry Disability Index (ODI), and regional kyphosis angle. We obtained preoperative and 9-month postoperative paraspinal muscle FI rates using T2-weighted MRI images and ImageJ software. We analyzed the correlation of FI rates with VAS, ODI, as well as the correction loss percentage of regional kyphosis angle. The analysis revealed a positive correlation between postoperative FI rate increase and correction loss percentage of regional kyphosis angle (r = 0.696, p < 0.001). The increase in paraspinal muscle FI rate was positively correlated with 9-month postoperative ODI (r = 0.763, p < 0.001). These findings indicate that an increase in postoperative paraspinal muscle FI rate may result in more significant correction loss of regional kyphosis angle and can lead to increased functional impairment in patients.