T2-weighted Images Research Articles

FONDUE: Robust resolution-invariant denoising of MR Images using Nested UNets

Abstract Recent human magnetic resonance imaging (MRI) studies continually push the boundaries of spatial resolution as a means to enhance levels of neuroanatomical detail and increase the accuracy and sensitivity of derived brain morphometry measures. However, acquisitions required to achieve these resolutions have a higher noise floor, potentially impacting segmentation and morphometric analysis results. This study proposes a novel, fast, robust, and resolution-invariant deep learning method to denoise structural human brain MRIs. We explore denoising of T1-weighted (T1w) brain images from varying field strengths (1.5T to 7T), voxel sizes (1.2mm to 250µm), scanner vendors (Siemens, GE, and Phillips), and diseased and healthy participants from a wide age range (young adults to aging individuals). Our proposed Fast-Optimized Network for Denoising through residual Unified Ensembles (FONDUE) method demonstrated stable denoising capabilities across multiple resolutions with performance on par or superior to the state-of-the-art methods while being several orders of magnitude faster at low relative cost when using a dedicated Graphics Processing Unit (GPU). FONDUE achieved the best performance on at least one of the four denoising-performance metrics on all the test datasets used, showing its generalization capabilities and stability. Due to its high-quality performance, robustness, fast execution times, and relatively low-GPU memory requirements, as well as its open-source public availability, FONDUE can be widely used for structural MRI denoising, especially in large-cohort studies. We have made the FONDUE repository and all training and evaluation scripts as well as the trained weights available at https://github.com/waadgo/FONDUE.

Deep learning-based segmentation of kidneys and renal cysts on T2-weighted MRI from patients with autosomal dominant polycystic kidney disease

BackgroundOur aim was to train and test a deep learning-based algorithm for automatically segmenting kidneys and renal cysts in patients with autosomal dominant polycystic kidney disease (ADPKD).MethodsWe retrospectively selected all ADPKD patients who underwent renal MRI with coronal T2-weighted imaging at our institution from 2008 to 2022. The 20 most recent examinations constituted the test dataset, to mimic pseudoprospective enrolment. The remaining ones constituted the training dataset to which eight normal renal MRIs were added. Kidneys and cysts ground truth segmentations were performed on coronal T2-weighted images by a junior radiologist supervised by an experienced radiologist. Kidneys and cysts of the 20 test MRIs were segmented by the algorithm and three independent human raters. Segmentations were compared using overlap metrics. The total kidney volume (TKV), total cystic volume (TCV), and cystic index (TCV divided by TKV) were compared using Bland–Altman analysis.ResultsWe included 164 ADPKD patients. Dice similarity coefficients ranged from 85.9% to 87.4% between the algorithms and the raters’ segmentations and from 84.2% to 86.2% across raters’ segmentations. For TCV assessment, the biases ± standard deviations (SD) were 3–19 ± 137–151 mL between the algorithm and the raters, and 22–45 ± 49–57 mL across raters. The algorithm underestimated TKV and TCV in two outliers with TCV > 2800 mL. For cystic index assessment, the biases ± SD were 2.5–6.9% ± 6.7–8.3% between the algorithm and the raters, and 2.1–9.4 ± 7.4–11.6% across raters.ConclusionThe algorithm’s performance fell within the range of inter-rater variability, but large TKV and TCV were underestimated.Relevance statementAccurate automated segmentation of the renal cysts will enable the large-scale evaluation of the prognostic value of TCV and cystic index in ADPKD patients. If these biomarkers are prognostic, then automated segmentation will facilitate their use in daily routine.Key PointsCystic volume is an emerging biomarker in ADPKD.The algorithm’s performance in segmenting kidneys and cysts fell within interrater variability.The segmentation of very large cysts, under-represented in the training dataset, needs improvement.Graphical

Does the high-intensity zone of lumbar intervertebral disc at magnetic resonance imaging have diagnostic value for discogenic low back pain? A meta-analysis

ObjectiveThe correlation between high-intensity zone (HIZ) of lumbar disc magnetic resonance imaging (MRI) and discogenic low back pain (DLBP) is currently controversial, this study aimed to systematically evaluate the correlation between HIZ of lumbar disc MRI and positive discography, as well as its diagnostic value for DLBP.MethodDatabases were searched to include research literature on high intensity zone (HIZ) related to discography and DLBP diagnosis. HIZ is a separate small, confined area of high signal located at the posterior border of the annulus fibrosus on MRI T2-weighted images of the lumbar spine, which is separated from the nucleus pulposus but has a higher signal than the nucleus pulposus. Studies on the correlation of HIZ with discography and DLBP diagnosis were searched in the Pubmed, EMBASE, Cochrane Central, Science Direct, China Knowledge Network, Wanfang Database, and China Biomedical Literature Databases, Scopus from January 1992 to June 2024. The outcomes were diagnostic values of HIZ for DLBP. The risk assessment was performed by Deeks’ funnel methods in the Stata 17.0 software after 2 investigators independently screened the literature, extracted information and evaluated the risk of bias of the included studies.ResultsA total of 25 studies including 5889 patients were included. meta-analysis showed that the sensitivity of HIZ for the diagnosis of DLBP was (0.49, 95% CI [0.37,0.61]) and specificity was (0.89, 95% CI [0.85,0.93]); the positive likelihood ratio was (4.52, 95% CI [3.28,6.25]) and the negative likelihood ratio was (0.58, 95% CI [0.46,0.71]). The diagnostic ratio was (7.87, 95% CI [5.05,12.26]).ConclusionThe available evidence suggests that HIZ has acceptable sensitivity and high specificity in the diagnosis of DLBP. Due to the limitation of the number and quality of included studies, the above conclusions need to be validated by more high-quality studies.

SPINEPS—automatic whole spine segmentation of T2-weighted MR images using a two-phase approach to multi-class semantic and instance segmentation

Abstract Objectives Introducing SPINEPS, a deep learning method for semantic and instance segmentation of 14 spinal structures (ten vertebra substructures, intervertebral discs, spinal cord, spinal canal, and sacrum) in whole-body sagittal T2-weighted turbo spin echo images. Material and methods This local ethics committee-approved study utilized a public dataset (train/test 179/39 subjects, 137 female), a German National Cohort (NAKO) subset (train/test 1412/65 subjects, mean age 53, 694 female), and an in-house dataset (test 10 subjects, mean age 70, 5 female). SPINEPS is a semantic segmentation model, followed by a sliding window approach utilizing a second model to create instance masks from the semantic ones. Segmentation evaluation metrics included the Dice score and average symmetrical surface distance (ASSD). Statistical significance was assessed using the Wilcoxon signed-rank test. Results On the public dataset, SPINEPS outperformed a nnUNet baseline on every structure and metric (e.g., an average over vertebra instances: dice 0.933 vs 0.911, p < 0.001, ASSD 0.21 vs 0.435, p < 0.001). SPINEPS trained on automated annotations of the NAKO achieves an average global Dice score of 0.918 on the combined NAKO and in-house test split. Adding the training data from the public dataset outperforms this (average instance-wise Dice score over the vertebra substructures 0.803 vs 0.778, average global Dice score 0.931 vs 0.918). Conclusion SPINEPS offers segmentation of 14 spinal structures in T2w sagittal images. It provides a semantic mask and an instance mask separating the vertebrae and intervertebral discs. This is the first publicly available algorithm to enable this segmentation. Key Points QuestionNo publicly available automatic approach can yield semantic and instance segmentation masks for the whole spine (including posterior elements) in T2-weighted sagittal TSE images. FindingsSegmenting semantically first and then instance-wise outperforms a baseline trained directly on instance segmentation. The developed model produces high-resolution MRI segmentations for the whole spine. Clinical relevanceThis study introduces an automatic approach to whole spine segmentation, including posterior elements, in arbitrary fields of view T2w sagittal MR images, enabling easy biomarker extraction, automatic localization of pathologies and degenerative diseases, and quantifying analyses as downstream research.

Spatial structural abnormality maps associated with cognitive and physical performance in relapsing-remitting multiple sclerosis.

We aimed to characterize the brain abnormalities that are associated with the cognitive and physical performance of patients with relapsing-remitting multiple sclerosis (RRMS) using a deep learning algorithm. Three-dimensional (3D) nnU-Net was employed to calculate a novel spatial abnormality map by T1-weighted images and 281 RRMS patients (Dataset-1, male/female = 101/180, median age [range] = 35.0 [17.0, 65.0] years) were categorized into subtypes. Comparison of clinical and MRI features between RRMS subtypes was conducted by Kruskal-Wallis test. Kaplan-Meier analysis was conducted to investigate disability progression in RRMS subtypes. Additional validation using two other RRMS datasets (Dataset-2, n = 33 and Dataset-3, n = 56) was conducted. Five RRMS subtypes were identified: (1) a Frontal-I subtype showing preserved cognitive performance and mild physical disability, and low risk of disability worsening; (2) a Frontal-II subtype showing low cognitive scores and severe physical disability with significant brain volume loss, and a high propensity for disability worsening; (3) a temporal-cerebellar subtype demonstrating lowest cognitive scores and severest physical disability among all subtypes but remaining relatively stable during follow-up; (4) an occipital subtype demonstrating similar clinical and imaging characteristics as the Frontal-II subtype, except a large number of relapses at baseline and preserved cognitive performance; and (5) a subcortical subtype showing preserved cognitive performance and low physical disability but a similar prognosis as the occipital and Frontal-II subtypes. Additional validation confirmed the above findings. Spatial abnormality maps can explain heterogeneity in cognitive and physical performance in RRMS and may contribute to stratified management. Question Can a deep learning algorithm characterize the brain abnormalities associated with the cognitive and physical performance of patients with RRMS? Findings Five RRMS subtypes were identified by the algorithm that demonstrated variable cognitive and physical performance. Clinical relevance The spatial abnormality maps derived RRMS subtypes had distinct cognitive and physical performances, which have a potential for individually tailored management.

Ovarian–Adnexal Reporting and Data System (O-RADS) MRI Scoring: Diagnostic Accuracy, Interobserver Agreement, and Applicability to Machine Learning

Abstract Objectives To evaluate the interobserver agreement and diagnostic accuracy of ovarian–adnexal reporting and data system magnetic resonance imaging (O-RADS MRI) and applicability to machine learning. Material and methods Dynamic contrast-enhanced pelvic MRI examinations 471 lesions were retrospectively analyzed and assessed by three radiologists according to O-RADS MRI criteria. Radiomic data were extracted from T2, and post-contrast fat-suppressed T1-weighted images. Using these data, an artificial neural network (ANN), support vector machine, random forest, and naive Bayes models were constructed. Results Among all readers, the lowest agreement was found for the O-RADS 4 group (kappa: 0.669 (95% confidence interval [CI] 0.634–0.733)), followed by the O-RADS 5 group (kappa: 0.709 (95% CI 0.678–0.754)). O-RADS 4 predicted a malignancy with an area under the curve (AUC) value of 74.3% (95% CI 0.701–0.782), and O-RADS 5 with an AUC of 95.5% (95% CI 0.932–0.972),(p < 0.001). Among the machine learning models, ANN achieved the highest success, distinguishing O-RADS groups with an AUC of 0.948, a precision of 0.861, and a recall of 0.824. Conclusion The interobserver agreement and diagnostic sensitivity of the O-RADS MRI in assigning O-RADS 4–5 were not perfect, indicating a need for structural improvement. Integrating artificial intelligence into MRI protocols may enhance their performance. Advances in knowledge Machine learning can achieve high accuracy in the correct classification of O-RADS MRI. Malignancy prediction rates were 74% for O-RADS 4 and 95% for O-RADS 5.

Saturation transfer (CEST and MT) MRI for characterization of U-87 MG glioma in the rat.

The focus of this work was to identify the optimal magnetic resonance imaging (MRI) contrast between orthotopic U-87 MG tumours and normal appearing brain with the eventual goal of treatment response monitoring. U-87 MG human glioblastoma cells were injected into the brain of RNU nude rats (n= 9). The rats were imaged at 7T at three timepoints for all animals: 3-5, 7-9, and 11-13 days after implantation. Whole-brain T1-weighted (before and after gadolinium contrast agent injection), diffusion, and fluid-attenuated inversion recovery scans were performed. In addition, single-slice saturation-transfer-weighted chemical exchange saturation transfer (CEST), magnetization transfer (MT), and water saturation shift referencing (WASSR) contrast Z-spectra and T1 and T2 maps were also acquired. The MT and WASSR Z-spectra and T1 map were fitted to a two-pool quantitative MT model to estimate the T2 of the free and macromolecular-bound water molecules, the relative macromolecular pool size (M0, MT), and the magnetization exchange rate from the macromolecular pool to the free pool (RMT). The T1-corrected apparent exchange-dependent relaxation (AREX) metric to isolate the CEST contributions was also calculated. The lesion on M0, MT and AREX maps with a B1 of 2μT best matched the hyperintensity on the post-contrast T1-weighted image. There was also good separation in Z-spectra between the lesion and contralateral cortex in the 2-μT CEST and 3- and 5-μT MT Z-spectra at all time points. A pairwise Wilcoxon signed-rank tests with Holm-Bonferroni adjustment on MRI parameters was performed and the differences between enhancing lesion and contralateral cortex for the MT ratio with 2μT saturation at 3.6ppm frequency offset (corresponding to the amide chemical group) and M0, MT were both strongly significant (p< 0.001) at all time points. This work has identified that differences between enhancing lesion and contralateral cortex are strongest in MTR with B1=2μT at 3.6ppm and relative macromolecular pool size (M0, MT) images over entire period of 3-13 days after cancer cell implantation.

Assessing myocardial tissue in the acute setting with Photon-Counting CT (PCCT) in comparison with Cardiac Magnetic Resonance (CMR)

Abstract Background Cardiac magnetic resonance (CMR) characterization of myocardial tissue is routinely used in clinical practice. Compared to traditional CT, Photon counting CT (PCCT) uses photon-counting detectors offering higher spatial resolution, elimination of electronic noise and improved contrast-to-noise ratio. A comparison between CMR and PCCT for myocardial characterizazion in an urgent setting, however, has not yet been explored. Purpose The aim of this study was to compare the ability to characterize myocardial tissue between PCCT and CMR in patients presenting with acute chest pain in the emergency department (ED). Methods This single-center prospective study enrolled all consecutive patients presenting to the ED of our hospital from November 2023 to January 2024 with chest pain, ECG alterations and troponin rise consistent with myocardial injury. Patients needing urgent coronary angiography according to guidelines were excluded. PCCT was performed urgently for triple rule-out, but also for tissue characterization through post-iodine-contrast administration images. Within 24 hours, all patients underwent a CMR protocol including T2-weighted and late gadolinium enhancement (LGE) images for tissue characterization. Results Six male patients with a mean age of 20 ± 10 years were enrolled. All clinical, PCCT, and CMR parameters are presented in Table 1. All patients presented with diffuse ST-segment elevation and exhibited T2-weighted images (for edema detection) and LGE distribution (subepicardial and intramyocardial) consistent with acute myocarditis. Mean radiation exposure for PCCT was 0.8±0.1 mSv. A strong linear correlation emerged between PCCT-derived delayed enhancement (DE) volume and CMR-derived late gadolinium enhancement (LGE) volume (r = 0.970; p value 0.001) (Table 1). Similar results were achieved when comparing the number of DE and LGE myocardial segments (r = 0.871; p value 0.026). Significant correlations also emerged between DE, LGE, and high-sensitive troponin (hsTN) at presentation (respectively: PCCT-DE vs hsTN r = 0.951, p value 0.004; CMR-LGE vs hsTN r = 0.918, p value 0.010). CMR showed a much higher signal-to-noise ratio (SNR) for the delayed enhanced myocardium and contrast-to-noise (CNR) ratio between the delayed enhanced and normal myocardium with respect to PCCT in post-contrast images (respectively 7.4 ± 3.9 vs 27.2 ± 17.7, p value 0.024 and 5.5 ± 3.7 vs 38.8 ± 30.2, p value 0.023). In T2-weighted images both SNR for the hyperintense myocardium (30.3 [20.4-122.4]) and CNR between the hyperintense and normal myocardium (26.7 [13.3-60.2]) resulted to be higher compared to PCCT (p value 0.027 and 0.045, respectively) (Table 1). Conclusions PCCT appears to be reliable when compared to CMR for detecting myocardial edema through the acquisition of post-contrast images in the acute setting with an acceptable CNR.

Comparison of image quality and lesion conspicuity between conventional and deep learning reconstruction in gadoxetic acid-enhanced liver MRI

ObjectiveTo compare the image quality and lesion conspicuity of conventional vs deep learning (DL)-based reconstructed three-dimensional T1-weighted images in gadoxetic acid-enhanced liver magnetic resonance imaging (MRI).MethodsThis prospective study (NCT05182099) enrolled participants scheduled for gadoxetic acid-enhanced liver MRI due to suspected focal liver lesions (FLLs) who provided signed informed consent. A liver MRI was conducted using a 3-T scanner. T1-weighted images were reconstructed using both conventional and DL-based (AIRTM Recon DL 3D) reconstruction algorithms. Three radiologists independently reviewed the image quality and lesion conspicuity on a 5-point scale.ResultsFifty participants (male = 36, mean age 62 ± 11 years) were included for image analysis. The DL-based reconstruction showed significantly higher image quality than conventional images in all phases (3.71–4.40 vs 3.37–3.99, p < 0.001 for all), as well as significantly less noise and ringing artifacts than conventional images (p < 0.05 for all), while also showing significantly altered image texture (p < 0.001 for all). Lesion conspicuity was significantly higher in DL-reconstructed images than in conventional images in the arterial phase (2.15 [95% confidence interval: 1.78, 2.52] vs 2.03 [1.65, 2.40], p = 0.036), but no significant difference was observed in the portal venous phase and hepatobiliary phase (p > 0.05 for all). There was no significant difference in the figure-of-merit (0.728 in DL vs 0.709 in conventional image, p = 0.474).ConclusionDL reconstruction provided higher-quality three-dimensional T1-weighted imaging than conventional reconstruction in gadoxetic acid-enhanced liver MRI.Critical relevance statementDL reconstruction of 3D T1-weighted images improves image quality and arterial phase lesion conspicuity in gadoxetic acid-enhanced liver MRI compared to conventional reconstruction.Key PointsDL reconstruction is feasible for 3D T1-weighted images across different spatial resolutions and phases.DL reconstruction showed superior image quality with reduced noise and ringing artifacts.Hepatic anatomic structures were more conspicuous on DL-reconstructed images.Graphical

Causal relationships between cortical brain structural alterations and migraine subtypes: a bidirectional Mendelian randomization study of 2,347 neuroimaging phenotypes

BackgroundPrevious studies have shown that migraines are associated with brain structural changes. However, the causal relationships between these changes and migraine, as well as its subtypes, migraine with aura (MA) and migraine without aura (MO), remain largely unclear.MethodsWe utilized genome-wide association study (GWAS) summary statistics from European cohorts for 2,347 cortical structural magnetic resonance imaging (MRI) phenotypes, derived from both T1-weighted and diffusion tensor imaging scans (n = 36,663), with migraine and its subtypes (n = 147,970–375,752). Cortical phenotypes included both macrostructural (e.g., cortical thickness, surface area) and microstructural (e.g., fractional anisotropy, mean diffusivity) features. Genetic correlations were first assessed to identify significant associations, followed by bidirectional Mendelian randomization (MR) analyses to determine causal relationships between these brain phenotypes and migraine, as well as its subtypes (MA and MO). Sensitivity analyses were applied to ensure the robustness of the results.ResultsGenetic correlation analysis identified 510 significant associations between cortical structural phenotypes and migraine across 401 distinct traits. Forward MR analysis revealed nine significant causal effects of cortical structural changes on migraine risk. Specifically, increased cortical thickness and local gyrification index in specific cortical regions were associated with a decreased risk of overall migraine, MA, and MO, while intracellular volume fraction and orientation diffusion index in specific regions increased the risk of MA and MO. Reverse MR analysis demonstrated that MA causally increased mean diffusivity in the insular and frontal opercular cortex. Sensitivity analyses confirmed the robustness of these findings, with no evidence of horizontal pleiotropy or heterogeneity.ConclusionThis study identifies causal relationships between cortical neuroimaging phenotypes and migraine, highlighting potential biomarkers for migraine diagnosis, treatment, and prevention.

Clinical significance of MRI-measured olfactory bulb height as an imaging biomarker of idiopathic Parkinson's disease.

This study aimed to determine whether the olfactory bulb height (OBH) measured using magnetic resonance imaging (MRI) has clinical utility as an imaging biomarker in the evaluation of patients with idiopathic Parkinson's disease (iPD) through its correlation with movement impairment. This retrospective study included cognitively intact patients with suspected parkinsonism. All participants underwent T2-weighted imaging to measure OBH. Logistic regression was used to determine whether OBH was an independent risk factor for distinguishing iPD patients from disease controls, and its relation with clinical parameters related to motor impairment, including clinical laterality, modified Hoehn and Yahr (HY) stage, and Unified Parkinson's Disease Rating Scale (UPDRS) III score, was investigated. Based on the final clinical diagnosis, 79 patients with iPD and 16 disease controls were included. The mean OBH was significantly smaller in iPD than in disease controls (p < 0.0001). OBH was a significant independent predictor of iPD, with a cutoff of 1.52 mm. In the comparison among the ipsilateral, contralateral side of iPD with clinical laterality, and disease control group, the OBH of the disease control group was significantly larger than both the ipsilateral and contralateral sides (P < 0.05). However, there was no significant difference in OBH between the ipsilateral and contralateral sides (p > 0.05). OBH according to HY stage was significantly smaller in HY stage 2-3 groups than in the disease control group (p < 0.001). The correlation analysis between UPDRS III and OBH showed a mild negative correlation (r = -0.32, p = 0.013). MRI-measured OBH is decreased in iPD regardless of age and sex and may be correlated with the progression of motor symptoms in the iPD.

Neonatal Brain MRI: Periventricular Germinal Matrix Mimicking Hypoxic-ischemic White Matter Injuries.

As pregnancy progresses, the germinal matrix volume decreases. Residual periventricular germinal matrix may be mistaken for hypoxic-ischemic white matter injury. This study aims to determine the prevalence and imaging characteristics of these findings. This retrospective study analyzed brain MRIs of newborns from 2012-2023, performed within the first week of life. MRIs were done for suspected hypoxic-ischemic injuries, post-natal neurological symptoms, and evaluation of prenatally diagnosed structural anomalies. Image analysis targeted the remnants of the frontal periventricular germinal matrix, assessing its imaging characteristics, including diffusion, T1, and T2 signal characteristics, and laterality. Frontal migrating cell bands were also assessed. Seventy newborns were included (mean gestational age at delivery was 38.3 ± 2.1 weeks, mean scan age 5.1 ± 1.9 days). Frontal periventricular gray matter was detected in 39 newborns (90% bilateral) on T2-weighted images, negatively correlated with gestational age (r = -0.31, p = 0.013); none showed decreased ADC or shortened T1 signal compared with the basal ganglia. Frontal periventricular bands were found in 37 newborns (97.3% bilateral), strongly correlating with periventricular gray matter (r = 0.71, p < 0.001). No correlation was found between clinical hypoxic-ischemic injuries and these features. The presence of frontal periventricular gray matter observed in early neonatal MRIs, without decreased ADC values or shortened T1 signal, is developmental, reflecting a late maturation phase. Careful interpretation of MRI characteristics, including diffusion, T1, and T2 signal intensities, is necessary before attributing these findings to hypoxic-ischemic white matter injury.

Prediction of hepatocellular carcinoma response to radiation segmentectomy using an MRI-based machine learning approach.

To evaluate the value of pre-treatment MRI-based radiomics in patients with hepatocellular carcinoma (HCC) for the prediction of response to Yttrium 90 radiation segmentectomy. This retrospective study included 154 patients (38 female; mean age 66.8years) who underwent contrast-enhanced MRI prior to radiation segmentectomy. Radiomics features were manually extracted on volumes of interest on post-contrast T1-weighted images at the portal venous phase (PVP). Tumor-based response assessment was evaluated 6months post-treatment using mRECIST. A logistic regression model was used to predict binary response outcome [complete response at 6months with no-re-treatment (response group) against the rest (non-response group, including partial response, progressive disease, stable disease and complete response after re-treatment within 6months after radiation segmentectomy) using baseline clinical parameters and radiomics features. We accessed the value of different sets of predictors using cross-validation technique. AUCs were compared using DeLong tests. A total 168 HCCs (mean size 2.9 ± 1.7cm) were analyzed in 154 patients. The response group consisted of 113 HCCs and the non-response group of 55 HCCs. Baseline clinical parameters (AUC 0.531; sensitivity, 0.781; specificity, 0.279; positive predictive value (PPV), 0.345; negative predictive value (NPV), 0.724) and AFP (AUC 0.632; sensitivity, 0.833; specificity, 0.466; PPV, 0.432; NPV, 0.851) showed poor performance for response prediction. The model using a combination of radiomics features and clinical parameters/AFP showed the best performance (AUC 0.736; sensitivity, 0.706; specificity, 0.662; PPV 0.504; NPV, 0.822), significantly better than the clinical model (p < 0.001) or AFP alone (p < 0.001). The combination of radiomics features from pre-treatment MRI with clinical parameters and AFP showed fair performance for predicting HCC response to radiation segmentectomy, better than that of AFP. These results need further validation.

Impact of non-contrast-enhanced imaging input sequences on the generation of virtual contrast-enhanced breast MRI scans using neural network.

To investigate how different combinations of T1-weighted (T1w), T2-weighted (T2w), and diffusion-weighted imaging (DWI) impact the performance of virtual contrast-enhanced (vCE) breast MRI. The IRB-approved, retrospective study included 1064 multiparametric breast MRI scans (age: 52 ± 12 years) obtained from 2017 to 2020 (single site, two 3-T MRI). Eleven independent neural networks were trained to derive vCE images from varying input combinations of T1w, T2w, and multi-b-value DWI sequences (b-value = 50-1500 s/mm2). Three readers evaluated the vCE images with regard to qualitative scores of diagnostic image quality, image sharpness, satisfaction with contrast/signal-to-noise ratio, and lesion/non-mass enhancement conspicuity. Quantitative metrics (SSIM, PSNR, NRMSE, and median symmetrical accuracy) were analyzed and statistically compared between the input combinations for the full breast volume and both enhancing and non-enhancing target findings. The independent test set consisted of 187 cases. The quantitative metrics significantly improved in target findings when multi-b-value DWI sequences were included during vCE training (p < 0.05). Non-significant effects (p > 0.05) were observed for the quantitative metrics on the full breast volume when comparing input combinations including T1w. Using T1w and DWI acquisitions during vCE training is necessary to achieve high satisfaction with contrast/SNR and good conspicuity of the enhancing findings. The input combination of T1w, T2w, and DWI sequences with three b-values showed the best qualitative performance. vCE breast MRI performance is significantly influenced by input sequences. Quantitative metrics and visual quality of vCE images significantly benefit when multi b-value DWI is added to morphologic T1w-/T2w sequences as input for model training. Question How do different MRI sequences impact the performance of virtual contrast-enhanced (vCE) breast MRI? Findings The input combination of T1-weighted, T2-weighted, and diffusion-weighted imaging sequences with three b-values showed the best qualitative performance. Clinical relevance While in the future neural networks providing virtual contrast-enhanced images might further improve accessibility to breast MRI, the significant influence of input data needs to be considered during translational research.

Thoracic hypertrophied ligamentum flavum: A retrospective study of pre-operative factors and surgical outcome

Objectives: Thoracic hypertrophied ligamentum flavum (HLF) is a rare cause of thoracic myelopathy which requires surgical treatment. The study aims to analyze the pre-operative factors and surgical outcomes in patients with thoracic myelopathy secondary to HLF. Materials and Methods: This is a retrospective study of patients who underwent surgery for thoracic myelopathy due to HLF at a single center from December 2015 to November 2023. We included patient’s demographic data, clinical symptoms, radiological details, operative details, and outcome. We used Nurick’s grading system for outcome evaluation in pre-operative, post-operative, and follow-up. Relationship of various pre-operative parameters was evaluated with surgical outcomes on binary scale (favorable group [Nurick’s grade 1 and 2] and unfavorable outcome [Nurick’s grade 3–5]) using univariate and multivariate analysis. Results: A total of 57 patients were included in the study. On analyzing various prognostic factors with respect to favorable and unfavorable outcomes using univariate analysis, four factors came out to be statistically significant which were segments involved (multi vs. single segment involvement) (P &lt; 0.05), walking difficulty (P &lt; 0.05), intramedullary signal changes on T2-weighted magnetic resonance imaging (MRI) (P &lt; 0.001), and pre-operative Nurick’s grade (P &lt; 0.001). On multivariate analysis, only one factor, pre-operative Nurick’s grade came out statistically significant. Conclusion: Various factors are important in predicting the outcome of a patient with thoracic myelopathy secondary to HLF. The most important of which is pre-operative Nurick’s grade. Other factors that also affect the outcome are the presence of multisegmented disease and intramedullary T2 signal changes on MRI.

Imaging manifestations and misdiagnosis of liposclerosing myxofibroma fumor

To focus on the imaging features of liposclerosing myxofibrous tumor (LSMFT) and diagnostic challenges to enhance clinical recognition and differential diagnosis accuracy. Retrospective analysis was conducted on the imaging and pathological data from 21 cases diagnosed with LSMFT between January 2014 and November 2022, including 14 males and 7 females (aged from 21 to 73 years; the course of disease ranged from 4 to 48 months). Patient demographics, clinical presentations, and imaging modalities including X-ray, CT, and MRI were reviewed. Pathological findings were correlated with imaging features to delineate diagnostic criteria and identify causes of misdiagnosis. All 21 patients represented primary lesions, located in 14 bilateral femurs [9 on the right and 5 on the left; 12 proximal femurs (intertrochanteric and peripheral) and 2 distal femurs], 2 proximal tibia, 2 proximal humerus, 1 proximal radius, and 1 ilium and 1 calcaneus. On X-ray and CT, all lesions showed cartographical or quasi-circular osteolytic destruction with distinct and sclerotic margins, continuous bone cortex, absence of periosteal reaction and surrounding soft tissue mass. Calcification, bone ridge and varying degrees of fat components were observed in 16 lesions, displaying mixed density;5 lesions showed ground-glass density with minimal bone ridge. On MRI, the signal intensity of the lesion was heterogeneous. T1-weighted imaging showed iso-to slightly high signal intensity, while T2-weighted imaging demonstrated unevenly high signal intensity. Fat-suppressed sequences depicted significantly elevated signal intensity within lesions, with post-contrast enhancement showing uneven patterns. 5 Cases were initially misdiagnosed as fibrous dysplasia. LSMFT is an uncommon benign bone tumor, typically localized in the proximal femur but occasionally found in other skeletal sites. Understanding its distinct imaging characteristics is crucial for accurate diagnosis. Typical cases exhibit identifiable imaging patterns, whereas atypical presentations may lead to misdiagnosis as fibrous dysplasia.

Clinical-radiomics nomogram based on the fat-suppressed T2 sequence for differentiating luminal and non-luminal breast cancer.

To establish and validate a new clinical-radiomics nomogram based on the fat-suppressed T2 sequence for differentiating luminal and non-luminal breast cancer. A total of 593 breast cancer patients who underwent preoperative breast MRI from Jan 2017 to Dec 2020 were enrolled, which were randomly divided into the training (n=474) and test sets (n=119) at the ratio of 8:2. Intratumoral region (ITR) of interest were manually delineated, and peritumoral regions of 3mm and 5mm (PTR-3 mm and PTR-5 mm) were automatically obtained by dilating the ITR. Intratumoral and peritumoral radiomics features were extracted from the fat-suppressed T2-weighted images, including first-order statistical features, shape features, texture features, and filtered features. The Mann-Whitney U Test, Z score normalization, K-best method, and least absolute shrinkage and selection operator (LASSO) algorithm were applied to select key features to construct radscores based on ITR, PTR-3 mm, PTR-5 mm, ITR+PTR-3 mm and ITR+ PTR-5 mm. Risk factors were selected by univariate and multivariate logistic regressions and were used to construct a clinical model and a clinical-radiomics model that presented as a nomogram. The performance of models was assessed by sensitivity, specificity, accuracy, the area under the curve (AUC) of receiver operating characteristic (ROC), calibration curves, and decision curve analysis (DCA). ITR+PTR-3 mm radsore and histological grade were selected as risk factors. A clinical-radiomics model was constructed by adding ITR+PTR-3mm radscore to the clinical factor, which was presented as a nomogram. The clinical-radiomics nomogram showed the highest AUC (0.873), sensitivity (72.3%), specificity (78.9%) and accuracy (77.0%) in the training set and the highest AUC (0.851), sensitivity (71.4%), specificity (79.8%) and accuracy (77.3%) in the test set. DCA showed that the clinical-radiomics nomogram had the greatest net clinical benefit compared to the other models. The clinical-radiomics nomogram showed promising clinical application value in differentiating luminal and non-luminal breast cancer.

Synthetic double inversion recovery imaging for rectal cancer T staging evaluation: imaging quality and added value to T2-weighted imaging

ObjectiveTo assess the image quality of synthetic double inversion recovery (SyDIR) imaging and enhance the value of T2-weighted imaging (T2WI) in evaluating T stage for rectal cancer patients.MethodsA total of 112 pathologically confirmed rectal cancer patients were retrospectively selected after undergoing MRI, including synthetic MRI. The image quality of T2WI and SyDIR imaging was compared based on signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), overall picture quality, presence of motion artifacts, lesion edge sharpness, and conspicuity. The concordance between MRI and pathological staging results, using T2WI alone and the combination of T2WI and SyDIR for junior and senior radiologists, was assessed using the Kappa test. The area under the receiver operating characteristic curve (AUC) was used to assess the diagnostic efficacy of extramural infiltration in rectal cancer patients.ResultsNo significant differences in imaging quality were observed between conventional T2WI and SyDIR (p = 0.07–0.53). The combination of T2WI and SyDIR notably improved the staging concordance between MRI and pathology for both junior (kappa value from 0.547 to 0.780) and senior radiologists (kappa value from 0.738 to 0.834). In addition, the integration of T2WI and SyDIR increased the AUC for diagnosing extramural infiltration for both junior (from 0.842 to 0.918) and senior radiologists (from 0.917 to 0.938).ConclusionThe combination of T2WI and SyDIR increased the consistency of T staging between MRI and pathology, as well as the diagnostic performance of extramural infiltration, which would benefit treatment selection.Critical relevance statementSyDIR sequence provides additional diagnostic value for T2WI in the T staging of rectal cancer, improving the agreement of T staging between MRI and pathology, as well as the diagnostic performance of extramural infiltration.Key PointsSynthetic double inversion recovery (SyDIR) and T2WI have comparable image quality.SyDIR provides rectal cancer anatomical features for extramural infiltration detections.The combination of T2WI and SyDIR improves the accuracy of T staging in rectal cancer.Graphical

Clinical, magnetic resonance imaging, histopathological features, treatment options and outcome of spinal ependymoma in dogs: 8 cases (2011-2022).

This study aims to report on the clinical magnetic resonance imaging, histological features, treatment options and outcomes of spinal ependymomas in dogs. Retrospective evaluation of medical records from dogs histologically confirmed spinal ependymomas with clinical presentations, magnetic resonance imaging findings, histological aspects, treatment options and outcomes. Eight dogs presented with acute to subacute onset of para- or tetraparesis. Magnetic resonance imaging findings included intramedullary oval-shaped space-occupying lesions that appeared hyperintense on T2-weighted images isointense on T1-weighted images and exhibited marked homogeneous or ring contrast enhancement. A peculiar feature, previously described only in human ependymomas, was observed in three patients - a T2-weighted hypointense rim, termed hemosiderin cap sign. Haematomyelia with necrotic foci was observed in one dog. Surgery, when performed, enabling a definitive intra-vitam diagnosis. Histological examination revealed that rosettes and pseudo-rosettes as disposition of neoplastic cells were the most common features reported. Furthermore, cerebrospinal fluid metastases were identified in one case. Clinical and histopathological findings in our case series were consistent with those previously reported in the literature. Magnetic resonance imaging features were fairly typical and highly suggestive of spinal ependymomas. The hemosiderin cap sign may aid in the presumptive intra-vitam diagnosis of these rare spinal tumours. Additionally, we described cerebrospinal fluid spread of neoplastic cells and subsequent multifocal or metastasis presentations. Surgery offered a dual benefit by facilitating intra-vitam diagnosis and, in some cases, extending survival time.

The mediation role of gray matter volume in the relationship between childhood maltreatment and psychological resilience in adolescents with first-episode major depressive disorder

Previous studies have revealed morphologic alterations in patients with major depressive disorder (MDD) with experiences of childhood trauma. However, the underlying neural mechanisms remain largely unknown. This study aims to explore the brain structural changes and their possible mediation role in the relationship between childhood maltreatment and psychological resilience in drug-naïve adolescents with first-episode MDD. A total of 57 adolescents with first-episode MDD and 36 healthy controls (HCs) completed the T1-weighted magnetic resonance imaging scan. The adverse childhood experiences and current psychological resilience were assessed using the Childhood Trauma Questionnaire-Short Form and the Connor Davidson Resilience Scale, respectively. The voxel-based morphometry approach was applied to examine changes in the gray matter volume (GMV). Compared with the HCs, adolescents with MDD had significantly reduced GMV volumes in the left fusiform gyrus, right orbitofrontal gyrus, right superior temporal gyrus, right calcarine cortex, right middle frontal gyrus, left angular gyrus, right precuneus, right posterior cingulate gyrus, and right posterior central gyrus, as well as significantly increased GMV volumes in the left lenticular putamen and right lenticular pallidum. The GMV of the right calcarine cortex was found to be negatively correlated with the severity of emotional abuse and positively correlated with the level of psychological resilience. Moreover, the GMV of the right calcarine cortex might partially mediate the relationship between childhood maltreatment and psychological resilience. The present study provided further evidence for structural impairments in adolescents with MDD. Our findings also confirmed the important role of depression-related GMV changes in childhood growth experiences and psychological resilience characteristics during adolescent brain maturation.