A novel deep learning method for large-scale analysis of bone marrow adiposity using UK Biobank Dixon MRI data

Background

Bone marrow adipose tissue (BMAT) represents > 10% fat mass in healthy humans and can be measured by magnetic resonance imaging (MRI) as the bone marrow fat fraction (BMFF). Human MRI studies have identified several diseases associated with BMFF but have been relatively small scale. Population-scale studies therefore have huge potential to reveal BMAT's true clinical relevance. The UK Biobank (UKBB) is undertaking MRI of 100,000 participants, providing the ideal opportunity for such advances.

Objective

To establish deep learning for high-throughput multi-site BMFF analysis from UKBB MRI data.

Materials and methods

We studied males and females aged 60-69. Bone marrow (BM) segmentation was automated using a new lightweight attention-based 3D U-Net convolutional neural network that improved segmentation of small structures from large volumetric data. Using manual segmentations from 61-64 subjects, the models were trained to segment four BM regions of interest: the spine (thoracic and lumbar vertebrae), femoral head, total hip and femoral diaphysis. Models were tested using a further 10-12 datasets per region and validated using datasets from 729 UKBB participants. BMFF was then quantified and pathophysiological characteristics assessed, including site- and sex-dependent differences and the relationships with age, BMI, bone mineral density, peripheral adiposity, and osteoporosis.

Results

Model accuracy matched or exceeded that for conventional U-Nets, yielding Dice scores of 91.2% (spine), 94.5% (femoral head), 91.2% (total hip) and 86.6% (femoral diaphysis). One case of severe scoliosis prevented segmentation of the spine, while one case of Non-Hodgkin Lymphoma prevented segmentation of the spine, femoral head and total hip because of T2 signal depletion; however, successful segmentation was not disrupted by any other pathophysiological variables. The resulting BMFF measurements confirmed expected relationships between BMFF and age, sex and bone density, and identified new site- and sex-specific characteristics.

Conclusions

We have established a new deep learning method for accurate segmentation of small structures from large volumetric data, allowing high-throughput multi-site BMFF measurement in the UKBB. Our findings reveal new pathophysiological insights, highlighting the potential of BMFF as a novel clinical biomarker. Applying our method across the full UKBB cohort will help to reveal the impact of BMAT on human health and disease.

Jointly constrained group sparse connectivity representation improves early diagnosis of Alzheimer's disease on routinely acquired T1-weighted imaging-based brain network

Background

Radiomics-based morphological brain networks (radMBN) constructed from routinely acquired structural MRI (sMRI) data have gained attention in Alzheimer's disease (AD). However, the radMBN suffers from limited characterization of AD because sMRI only characterizes anatomical changes and is not a direct measure of neuronal pathology or brain activity.

Purpose

To establish a group sparse representation of the radMBN under a joint constraint of group-level white matter fiber connectivity and individual-level sMRI regional similarity (JCGS-radMBN).

Methods

Two publicly available datasets were adopted, including 120 subjects from ADNI with both T1-weighted image (T1WI) and diffusion MRI (dMRI) for JCGS-radMBN construction, 818 subjects from ADNI and 200 subjects solely with T1WI from AIBL for validation in early AD diagnosis. Specifically, the JCGS-radMBN was conducted by jointly estimating non-zero connections among subjects, with the regularization term constrained by group-level white matter fiber connectivity and individual-level sMRI regional similarity. Then, a triplet graph convolutional network was adopted for early AD diagnosis. The discriminative brain connections were identified using a two-sample t-test, and the neurobiological interpretation was validated by correlating the discriminative brain connections with cognitive scores.

Results

The JCGS-radMBN exhibited superior classification performance over five brain network construction methods. For the typical NC vs. AD classification, the JCGS-radMBN increased by 1-30% in accuracy over the alternatives on ADNI and AIBL. The discriminative brain connections exhibited a strong connectivity to hippocampus, parahippocampal gyrus, and basal ganglia, and had significant correlation with MMSE scores.

Conclusion

The proposed JCGS-radMBN facilitated the AD characterization of brain network established on routinely acquired imaging modality of sMRI.

Supplementary Information

The online version of this article (10.1007/s13755-023-00269-0) contains supplementary material, which is available to authorized users.

Pathways explaining racial/ethnic and socio-economic disparities in brain white matter integrity outcomes in the UK Biobank study

Pathways explaining racial/ethnic and socio-economic status (SES) disparities in white matter integrity (WMI) reflecting brain health, remain underexplored, particularly in the UK population. We examined racial/ethnic and SES disparities in diffusion tensor brain magnetic resonance imaging (dMRI) markers, namely global and tract-specific mean fractional anisotropy (FA), and tested total, direct and indirect effects through lifestyle, health-related and cognition factors using a structural equations modeling approach among 36,184 UK Biobank participants aged 40-70 y at baseline assessment (47% men). Multiple linear regression models were conducted, testing independent associations of race/ethnicity, socio-economic and other downstream factors in relation to global mean FA, while stratifying by Alzheimer's Disease polygenic Risk Score (AD PRS) tertiles. Race (Non-White vs. White) and lower SES predicted poorer WMI (i.e. lower global mean FA) at follow-up, with racial/ethnic disparities in FAmean involving multiple pathways and SES playing a central role in those pathways. Mediational patterns differed across tract-specific FA outcomes, with SES-FAmean total effect being partially mediated (41% of total effect = indirect effect). Furthermore, the association of poor cognition with FAmean was markedly stronger in the two uppermost AD PRS tertiles compared to the lower tertile (T2 and T3: β±SE: -0.0009 ± 0.0001 vs. T1: β±SE: -0.0005 ± 0.0001, P < 0.001), independently of potentially confounding factors. Race and lower SES were generally important determinants of adverse WMI outcomes, with partial mediation of socio-economic disparities in global mean FA through lifestyle, health-related and cognition factors. The association of poor cognition with lower global mean FA was stronger at higher AD polygenic risk.

Multiparametric quantification of T1 and T2 relaxation time of bone metastasis in comparison with red or fatty bone marrow using magnetic resonance fingerprinting

OBJECTIVES

To assess the T1 and T2 values of bone marrow lesions in spine and pelvis derived from magnetic resonance fingerprinting (MRF) and to evaluate the differences in values among bone metastasis, red marrow and fatty marrow.

METHODS

Sixty patients who underwent lumbar spine and pelvic MRI with magnetic resonance fingerprinting were retrospectively included. Among eligible patients, those with bone metastasis, benign red marrow deposition and normal fatty marrow were identified. Two radiologists independently measured the T1 and T2 values from metastatic bone lesions, fatty marrow, and red marrow deposition on three-dimensional-magnetic resonance fingerprinting. Intergroup comparison and interobserver agreement were analyzed.

RESULTS

T1 relaxation time was significantly higher in osteoblastic metastasis than in red marrow (1674.6 ± 436.3 vs 858.7 ± 319.5, p < .001). Intraclass correlation coefficients for T1 and T2 values were 0.96 (p < 0.001) and 0.83 (p < 0.001), respectively. T2 relaxation time of osteoblastic metastasis and red marrow deposition had no evidence of a difference (osteoblastic metastasis, 57.9 ± 25.0 vs red marrow, 58.0 ± 34.4, p = 0.45), as were the average T2 values of osteolytic metastasis and red marrow deposition (osteolytic metastasis, 45.3 ± 15.1 vs red marrow, 58.0 ± 34.4, p = 0.63).

CONCLUSIONS

We report the feasibility of three-dimensional-magnetic resonance fingerprinting based quantification of bone marrow to differentiate bone metastasis from red marrow. Simultaneous T1 and T2 quantification of metastasis and red marrow deposition was possible in spine and pelvis and showed significant different values with excellent inter-reader agreement.

ADVANCE IN KNOWLEDGE

T1 values from three-dimensional-magnetic resonance fingerprinting might be a useful quantifier for evaluating bone marrow lesions.

Alterations in cortical thickness of frontoparietal regions in patients with social anxiety disorder

Although functional changes of the frontal and (para)limbic area for emotional hyper-reactivity and emotional dysregulation are well documented in social anxiety disorder (SAD), prior studies on structural changes have shown mixed results. This study aimed to identify differences in cortical thickness between SAD and healthy controls (CON). Thirty-five patients with SAD and forty-two matched CON underwent structural magnetic resonance imaging. A vertex-based whole brain and regional analyses were conducted for between-group comparison. The whole-brain analysis revealed increased cortical thickness in the left insula, left superior parietal lobule, left superior temporal gyrus, and left frontopolar cortex in patients with SAD compared to CON, as well as decreased thickness in the left superior/middle frontal gyrus and left fusiform gyrus in patients (after multiple-correction). The results from the ROI analysis did not align with these findings at the statistically significant level after multiple corrections. Changes in cortical thickness were not correlated with social anxiety symptoms. While consistent results were not obtained from different analysis methods, the results from the whole-brain analysis suggest that patients with SAD exhibit distinct neural deficits in areas involved in salience, attention, and socioemotional processing.

Evaluation of renal tubular function by multiparametric functional MRI in early diabetes

Purpose To evaluate the tubular function in an alloxan-induced type 1 diabetes mellitus (DM) rabbit model measured by renal oxygenation (R2*), oxygen extraction fraction (OEF), and renal blood flow (RBF) using blood oxygenation level dependent, asymmetric spin echo, and arterial spin labeling MRI. Methods Twenty-six rabbits were randomized into the 3-day DM group (n = 13) and the 7-day DM group (n = 13). We performed pairs of multiparametric MRIs (before and after furosemide injection) at baseline and 3/7 days post-DM, and scored pathological kidney injury. We performed statistical analyses using non-parametric, chi-square, and Spearman correlation tests. Results At baseline, medullary R2* significantly decreased by 24.97% and 16.74% in the outer and inner stripes of the outer medulla (OS and IS, p = 0.006 and 0.003, respectively) after furosemide administration. While the corresponding OEF decreased by 15.91% for OS and 16.67% for IS (both p = 0.003), and no significant change in medullary RBF was observed (p > 0.05). In the 3-day DM group, the decrease of medullary R2* and OEF post-furosemide became unremarkable, suggesting tubular dysfunction. We noticed similar changes in the 7-day DM group. Correlation analysis showed pathological tubular injury score significantly correlated with medullary ∆R2* (post-furosemide - pre-furosemide difference, r = 0.82 for OS and 0.82 for IS) and ∆OEF (r = 0.82 for OS and 0.82 for IS) (p < 0.001, respectively). Conclusion: The combination of medullary OEF and R2* in response to furosemide could detect renal tubular dysfunction in early DM.

A novel MRI-based diagnostic model for predicting placenta accreta spectrum

Objective To develop and evaluate a diagnostic model based on MRI signs for predicting placenta accreta spectrum. Materials and Methods A total of 155 pregnant women were included in this study, randomly divided into 104 cases in the training set and 51 cases in the validation set. There were 93 Non-PAS cases, and 62 cases in the PAS group. The training set included 62 Non-PAS cases and 42 PAS cases. Clinical factors and MRI signs were collected for univariate analysis. Then, binary logistic regression analysis was used to develop independent diagnostic models with clinical relevant risk factors or MRI signs, as well as those combining clinical risk factors and MRI signs. The ROC curve analysis was used to evaluate the diagnostic performance of each diagnostic model. Finally, the validation was performed with the validation set. Results In the training set, four clinical factors (gestity, parity, uterine surgery history, placental position) and 11 MRI features (T2-dark bands, placental bulge, T2 hypointense interface loss, myometrial thinning, bladder wall interruption, focal exophytic mass, abnormal placental bed vascularization, placental heterogeneity, asymmetric placental thickening/shape, placental ischemic infarction, abnormal intraplacental vascularity) were considered as risk factors for PAS. The AUC of the clinical diagnostic model, MRI diagnostic model, and clinical + MRI model of PAS were 0.779, 0.854, and 0.874, respectively. In the validation set, the AUC of the clinical diagnostic model, MRI diagnostic model, and clinical + MRI model of PAS were 0.655, 0.728, and 0.735, respectively. Conclusion Diagnosis model based on MRI features in this study can well predict placenta accreta spectrum.

Quantifying blood-brain barrier permeability in patients with ischemic stroke using non-contrast MRI

Increased blood-brain barrier permeability (BBBP) after ischemic stroke predisposes patients to hemorrhagic conversion. While altered BBBP can impact patient recovery, it is not routinely assessed during the workup of acute ischemic stroke (AIS). We study the effectiveness of the non-contrast MRI sequences diffusion-prepared pseudocontinuous arterial spin labeling (DP-pCASL) and Neurite Orientation Dispersion and Density Imaging (NODDI) in assessing BBBP and correlating to tissue microstructure after ischemic insult. Twelve patients with AIS were prospectively enrolled to undergo our multimodal MR imaging, which generated the DP-pCASL-derived cerebral blood flow (CBF), arterial transit time (ATT), and water exchange rate (kw) and the NODDI-derived b0, mean diffusivity (MD), orientation dispersion index (ODI), intracellular volume fraction (ICVF), and isotropic volume fraction (ISO) parametric maps. The mean age of the patients was 70.2 ± 14.8 with an average NIHSS of 13.0 (7.3-19.8). MR imaging was performed on average at 53.7 (27.8-93.3) hours from stroke symptom onset. The water exchange rate (kw) of the infarcted area and its contralateral territory were 89.7 min-1 (66.7-121.9) and 89.9 min-1 (65.9-106.0) respectively (p = 0.887). Multivariable linear regression analysis showed that b0, ODI, ISO and mechanical thrombectomy were significant predictors of kw. DP-pCASL and NODDI are promising non-contrast sequences for the routine assessment of BBBP.

Multisequence MRI-based radiomics signature as potential biomarkers for differentiating KRAS mutations in non-small cell lung cancer with brain metastases

Background

Kirsten rat sarcoma virus (KRAS) has evolved from a genotype with predictive value to a therapeutic target recently. The study aimed to establish non-invasive radiomics models based on MRI to discriminate KRAS from epidermal growth factor receptor (EGFR) or anaplastic lymphoma kinase (ALK) mutations in lung cancer patients with brain metastases (BM), then further explore the optimal sequence for prediction.

Methods

This retrospective study involved 317 patients (218 patients in training cohort and 99 patients in testing cohort) who had confirmed of KRAS, EGFR or ALK mutations. Radiomics features were separately extracted from T2WI, T2 fluid-attenuated inversion recovery (T2-FLAIR), diffusion weighted imaging (DWI) and contrast-enhanced T1-weighted imaging (T1-CE) sequences. The maximal information coefficient and recursive feature elimination method were used to select informative features. Then we built four radiomics models for differentiating KRAS from EGFR or ALK using random forest classifier. ROC curves were used to validate the capability of the models.

Results

The four radiomics models for discriminating KRAS from EGFR all worked well, especially DWI and T2WI models (AUCs: 0.942, 0.942 in training cohort, 0.949, 0.954 in testing cohort). When KRAS compared to ALK, DWI and T2-FLAIR models showed excellent performance in two cohorts (AUCs: 0.947, 0.917 in training cohort, 0.850, 0.824 in testing cohort).

Conclusions

Radiomics classifiers integrating MRI have potential to discriminate KRAS from EGFR or ALK, which are helpful to guide treatment and facilitate the discovery of new approaches capable of achieving this long-sought goal of cure in lung cancer patients with KRAS.

Radiomics and machine learning based on preoperative MRI for predicting extrahepatic metastasis in hepatocellular carcinoma patients treated with transarterial chemoembolization

Purpose

To develop and validate a radiomics machine learning (Rad-ML) model based on preoperative MRI to predict extrahepatic metastasis (EHM) in hepatocellular carcinoma (HCC) patients receiving transarterial chemoembolization (TACE) treatment.

Methods

A total of 355 HCC patients who received multiple TACE procedures were split at random into a training set and a test set at a 7:3 ratio. Radiomic features were calculated from tumor and peritumor in arterial phase and portal venous phase, and were identified using intraclass correlation coefficient, maximal relevance and minimum redundancy, and least absolute shrinkage and selection operator techniques. Cox regression analysis was employed to determine the clinical model. The best-performing algorithm among eight machine learning methods was used to construct the Rad-ML model. A nomogram combining clinical and Rad-ML parameters was used to develop a combined model. Model performance was evaluated using C-index, decision curve analysis, calibration plot, and survival analysis.

Results

In clinical model, elevated neutrophil to lymphocyte ratio and alpha-fetoprotein were associated with faster EHM. The XGBoost-based Rad-ML model demonstrated the best predictive performance for EHM. When compared to the clinical model, both the Rad-ML model and the combination model performed better (C-indexes of 0.61, 0.85, and 0.86 in the training set, and 0.62, 0.82, and 0.83 in the test set, respectively). However, the combined model's and the Rad-ML model's prediction performance did not differ significantly. The most influential feature was peritumoral waveletHLL_firstorder_Minimum in AP, which exhibited an inverse relationship with EHM risk.

Conclusions

Our study suggests that the preoperative MRI-based Rad-ML model is a valuable tool to predict EHM in HCC patients treated with TACE.

Advanced protein-embedded bimetallic nanocomposite optimized for in vivo fluorescence and magnetic resonance bimodal imaging

HYPOTHESIS

The development of bimodal imaging probes represents a hot topic of current research. Herein, we deal with developing an innovative bimodal contrast agent enabling fluorescence imaging (FI)/magnetic resonance imaging (MRI) and, simultaneously, consisting of biocompatible nanostructures. Optimized synthesis of advanced protein-embedded bimetallic (APEBM) nanocomposite containing luminescent gold nanoclusters (AuNC) and superparamagnetic iron oxide nanoparticles (SPION), suitable for in vivo dual-modal FI/MR imaging is reported.

EXPERIMENTS

The APEBM nanocomposite was prepared by a specific sequential one-pot green synthetic approach that is optimized to increase metals (Au, Fe) content and, consequently, the imaging ability of the resulting nanostructures. The protein matrix, represented by serum albumin, was intentionally chosen, and used since it creates an efficient protein corona for both types of optically/magnetically-susceptible nanostructures (AuNC, SPION) and ensures biocompatibility of the resulting APEBM nanocomposite although it contains elevated metal concentrations (approx. 1 mg·mL-1 of Au, around 0.3 mg·mL-1 of Fe). In vitro and in vivo imaging was performed.

FINDINGS

Successful in vivo FI and MRI recorded in healthy mice corroborated the applicability of the APEBM nanocomposite and, simultaneously, served as a proof of concept concerning the potential future exploitation of this new FI/MRI bimodal contrast agent in preclinical and clinical practice.

A fully automatic parenchyma extraction method for MRI T2* relaxometry of iron loaded liver in transfusion-dependent patients

PURPOSE

To develop a fully automatic parenchyma extraction method for the T2* relaxometry of iron overload liver.

METHODS

A retrospective multicenter collection of liver MR examinations from 177 transfusion-dependent patients was conducted. The proposed method extended a semiautomatic parenchyma extraction algorithm to a fully automatic approach by introducing a modified TransUNet on the R2* (1/T2*) map for liver segmentation. Axial liver slices from 129 patients at 1.5 T were allocated to training (85%) and internal test (15%) sets. Two external test sets separately included 1.5 T data from 20 patients and 3.0 T data from 28 patients. The final T2* measurement was obtained by fitting the average signal of the extracted liver parenchyma. The agreement between T2* measurements using fully and semiautomatic parenchyma extraction methods was assessed using coefficient of variation (CoV) and Bland-Altman plots.

RESULTS

Dice of the deep network-based liver segmentation was 0.970 ± 0.019 on the internal dataset, 0.960 ± 0.035 on the external 1.5 T dataset, and 0.958 ± 0.014 on the external 3.0 T dataset. The mean difference bias between T2* measurements of the fully and semiautomatic methods were separately 0.12 (95% CI: -0.37, 0.61) ms, 0.04 (95% CI: -1.0, 1.1) ms, and 0.01 (95% CI: -0.25, 0.23) ms on the three test datasets. The CoVs between the two methods were 4.2%, 4.8% and 2.0% on the internal test set and two external test sets.

CONCLUSIONS

The developed fully automatic parenchyma extraction approach provides an efficient and operator-independent T2* measurement for assessing hepatic iron content in clinical practice.

Volumetric brain MRI signatures of heart failure with preserved ejection fraction in the setting of dementia

Heart failure with preserved ejection fraction (HFpEF) is an important, emerging risk factor for dementia, but it is not clear whether HFpEF contributes to a specific pattern of neuroanatomical changes in dementia. A major challenge to studying this is the relative paucity of datasets of patients with dementia, with/without HFpEF, and relevant neuroimaging. We sought to demonstrate the feasibility of using modern data mining tools to create and analyze clinical imaging datasets and identify the neuroanatomical signature of HFpEF-associated dementia. We leveraged the bioinformatics tools at Vanderbilt University Medical Center to identify patients with a diagnosis of dementia with and without comorbid HFpEF using the electronic health record. We identified high resolution, clinically-acquired neuroimaging data on 30 dementia patients with HFpEF (age 76.9 ± 8.12 years, 61% female) as well as 301 age- and sex-matched patients with dementia but without HFpEF to serve as comparators (age 76.2 ± 8.52 years, 60% female). We used automated image processing pipelines to parcellate the brain into 132 structures and quantify their volume. We found six regions with significant atrophy associated with HFpEF: accumbens area, amygdala, posterior insula, anterior orbital gyrus, angular gyrus, and cerebellar white matter. There were no regions with atrophy inversely associated with HFpEF. Patients with dementia and HFpEF have a distinct neuroimaging signature compared to patients with dementia only. Five of the six regions identified in are in the temporo-parietal region of the brain. Future studies should investigate mechanisms of injury associated with cerebrovascular disease leading to subsequent brain atrophy.

Multicompartmental cystic trigeminal schwannoma as an uncommon differential diagnosis of cerebellopontine angle tumors

Trigeminal schwannoma is the second most common intracranial schwannoma yet accounts for less than 0.5% intracranial tumors [1]. Cystic degeneration is uncommon. We would like to report a pathologically proven multicompartmental cystic trigeminal schwannoma in a young adult presenting with chronic headache. A literature review on the imaging features of trigeminal schwannoma is performed to assist radiologists in accurate disease localization and prioritizing differential diagnosis in challenging cases. Confident preoperative radiological diagnosis would directly affect management strategies.

Multi modality fusion transformer with spatio-temporal feature aggregation module for psychiatric disorder diagnosis

Bipolar disorder (BD) is characterized by recurrent episodes of depression and mild mania. In this paper, to address the common issue of insufficient accuracy in existing methods and meet the requirements of clinical diagnosis, we propose a framework called Spatio-temporal Feature Fusion Transformer (STF2Former). It improves on our previous work - MFFormer by introducing a Spatio-temporal Feature Aggregation Module (STFAM) to learn the temporal and spatial features of rs-fMRI data. It promotes intra-modality attention and information fusion across different modalities. Specifically, this method decouples the temporal and spatial dimensions and designs two feature extraction modules for extracting temporal and spatial information separately. Extensive experiments demonstrate the effectiveness of our proposed STFAM in extracting features from rs-fMRI, and prove that our STF2Former can significantly outperform MFFormer and achieve much better results among other state-of-the-art methods.

Computed tomography: Diagnostic detection of complete pericardial agenesis: A case report

Congenital complete absence of the pericardium is a rare condition, often difficult to diagnose due to its incidental discovery or nonspecific clinical manifestations. Instrumental investigations commonly used as initial approaches, such as chest radiography and electrocardiogram, are often insufficient. Echocardiography is an imaging technique that is used for the initial evaluation of pericardial diseases. However, echocardiography does not offer a physiological anatomical delineation of the pericardium and can be affected by operator dependency, acoustic and nontraditional imaging windows. Therefore, accurate imaging techniques such as computed tomography (CT) or magnetic resonance imaging (MRI) are required for correct diagnosis. We present a case of a symptomatic patient with complete pericardial agenesis diagnosed on angio-CT. This case can contribute to highlighting the importance of CT as a comprehensive imaging method in diagnosis, despite MRI being the gold standard in pericardial disease assessment.

Peripheral tear of the triangular fibrocartilage complex: diagnostic accuracy of magnetic resonance imaging and diagnostic performance of the primary and secondary signs

OBJECTIVE

This study is to assess the diagnostic performance of magnetic resonance imaging (MRI) findings for type 1B triangular fibrocartilage complex (TFCC) tear of the wrist.

MATERIALS AND METHODS

This study retrospectively enrolled 78 patients to examine the diagnostic performance of preoperative MRI examinations in patients with type 1B TFCC tears. Thirty-nine participants had confirmed type 1B TFCC tear. The control group included 39 patients who were randomly selected from 1157 patients who underwent MRI for wrist pain. Both groups underwent a review of 19 MRI findings by two independent radiologists, and the correlation between each diagnostic finding and type 1B TFCC tear was assessed using the chi-squared test. The 19 MRI findings comprised eight primary signs of abnormalities in the distal or proximal lamina, in conjunction with 11 secondary signs suggestive of abnormalities in the surrounding structures.

RESULTS

The TFCC tear group demonstrated a significantly higher incidence of two primary MRI signs, i.e., fiber discontinuity and signal alteration in the distal lamina, as observed by both readers (R1, 74.4% vs. 38.5%, p = 0.003, and 87.2% vs. 43.6%, p < 0.001; R2, 74.4% vs. 35.9%, p = 0.001, and 87.2% vs. 53.8%, p < 0.003, respectively). Reader 2 identified a higher prevalence of two additional primary MRI signs: fiber discontinuity and signal alteration in the proximal lamina (all p < 0.05). None of the 11 secondary MRI signs demonstrated statistically significant associations with type 1B TFCC.

CONCLUSION

MRI manifestations of fiber discontinuity and signal alteration in the distal lamina may provide predictive markers for type 1B TFCC wrist tear.

Fast and accurate 3-D spine MRI segmentation using FastCleverSeg

Accurate and efficient segmenting of vertebral bodies, muscles, and discs is crucial for analyzing various spinal diseases. However, traditional methods are either laborious and time-consuming (manual segmentation) or require extensive training data (fully automatic segmentation). FastCleverSeg, our proposed semi-automatic segmentation approach, addresses those limitations by significantly reducing user interaction while maintaining high accuracy. First, we reduce user interaction by requiring the manual annotation of only two or three slices. Next, we automatically Estimate the Annotation on Intermediary Slices (EANIS) using traditional computer vision/graphics concepts. Finally, our proposed method leverages improved voxel weight balancing to achieve fast and precise volumetric segmentation in the segmentation process. Experimental evaluations on our assembled diverse MRI databases comprising 179 patients (60 male, 119 female), demonstrate a remarkable 25 ms (30 ms standard deviation) processing time and a significant reduction in user interaction compared to existing approaches. Importantly, FastCleverSeg maintains or surpasses the segmentation quality of competing methods, achieving a Dice score of 94%. This invaluable tool empowers physicians to efficiently generate reliable ground truths, expediting the segmentation process and paving the way for future integration with deep learning approaches. In turn, this opens exciting possibilities for future fully automated spine segmentation.

Hippocampal neurometabolic and structural changes from pre-to post-COVID-19: A case-series study

BACKGROUND

Neurological complications of the COVID-19 infection may be caused in part by local neurochemical and structural abnormalities that could not be detected during routine medical examinations. We examined within subject neurometabolic and structural brain alterations from pre-to post-COVID-19 in the hippocampal region of three elderly individuals (aged 63-68 years) who had a COVID-19 infection with mild symptoms. Patients were participating in an interventional study in which they were closely monitored at the time they were diagnosed with COVID-19. Patients 1 and 2 just completed 18-20 resistance training sessions prior to their diagnosis. Patient 3 was assigned to a non-training condition in the same study.

METHODS

Whole brain magnetic resonance imaging (MRI) images and proton magnetic resonance spectroscopy (1H-MRS) of the left hippocampus were collected before and after infection. Structural and spectroscopic imaging measures post-COVID-19 were contrasted to the pre-COVID-19 measures and were compared with values for Minimal Detectable Change at 95% (MDC95) and 90% (MDC90) confidence from a group of six elderly (aged 60-79 years) without COVID-19 that participated in the same study.

RESULTS

After SARS-COV-2 infection, we observed a reduction of glutamate-glutamine (Glx) in Patients 1 and 2 (≥ 42.0%) and elevation of myo-inositol (mIns) and N-acetyl-aspartate (NAA) in Patient 3 (≥ 36.4%); all > MDC90. MRI findings showed increased (Patients 1 and 2) or unchanged (Patient 3) hippocampal volume.

CONCLUSIONS

Overall, findings from this exploratory study suggest that mild COVID-19 infection could be associated with development of local neuroinflammation and reduced glutamate levels in the hippocampus. Our 1H-MRS findings may have clinical value for explaining chronic neurological and psychological complaints in COVID-19 long-haulers.

Thoracic aorta injury detected by 4D flow MRI predicts subsequent main adverse cardiovascular events in breast cancer patients receiving anthracyclines: A longitudinal study

PURPOSE

To investigate longitudinal thoracic aorta injury using 3-dimensional phase-contrast magnetic resonance imaging (4D flow MRI) parameters and to evaluate their value for predicting the subsequent main adverse cardiovascular events (MACEs) in breast cancer patients receiving anthracyclines.

METHODS

Between July 2020 and July 2021, eighty-eight female participants with breast cancer scheduled to receive anthracyclines with or without trastuzumab prospectively enrolled. Each subjects underwent 4D flow MRI at baseline, 3 and 6 months in relation to baseline. The diameter, peak velocity (Vpeak), wall shear stress (WSS), pulse wave velocity (PWV), energy loss (EL) and pressure gradient (PG) of thoracic aorta were measured. The association between these parameters and subsequent MACEs was performed by Cox proportional hazard models.

RESULTS

Ten participants had subsequently MACEs. The Vpeak and PG gradually decreased and the WSS, PWV and EL progressively increased at 3 and 6 months compared with baseline. Adjusted multivariable analysis showed that the WSS of the proximal, mid- and distal ascending aorta [HR, 1.314 (95% confidence interval (CI): 1.003, 1.898)], [HR, 1.320 (95% CI: 1.002, 1.801)] and [HR, 1.322 (95% CI: 1.001, 1.805)] and PWV of ascending aorta [HR, 2.223 (95% CI: 1.010, 4.653)] at 3 months were associated with subsequent MACEs. Combined WSS and PWV of ascending aorta at 3 months yielded the highest AUC (0.912) for predicting subsequent MACEs.

CONCLUSION

Combined WSS and PWV of ascending aorta at 3 months is helpful for predicting the subsequent MACEs in breast cancer patients treated by anthracyclines.

Characterization of kidneys in patients with systemic sclerosis by multi-parametric magnetic resonance quantitative imaging

PURPOSE

To determine the usefulness of multiparametric magnetic resonance (MR) quantitative imaging in characterizing the kidneys in systemic sclerosis (SSc) patients.

MATERIAL AND METHODS

Forty-six SSc patients (47.9 ± 12.8 years, 40 females) and 22 age- and sex- matched healthy volunteers (46.1 ± 13.8 years, 20 females) were recruited and underwent renal MR imaging by acquiring blood oxygen level dependent and saturated multi-delay renal arterial spin labeling (SAMURAI) sequences. The T2* value, T1 value, renal blood flow (RBF), arterial bolus arrival time (aBAT), and tissue bolus arrival time (tBAT) of renal cortex were measured and compared among diffuse cutaneous SSc (dcSSc) and limited cutaneous SSc (lcSSc) groups and healthy controls using One-way ANOVA and analyzed by logistic regression.

RESULTS

Compared to healthy volunteers, SSc patients with normal estimated glomerular filtration rate (n = 40) had significantly lower T2* value (P = 0.026) in the left renal cortex, longer T1 value (right: P = 0.015; left: P = 0.023), lower RBF (right: P < 0.001; left: P < 0.001), and shorter tBAT (right: P < 0.001; left: P = 0.005) in both right and left renal cortex after adjusting for demographics. The dcSSc patients (n = 23) had significantly lower RBF in both right (226.7 ± 65.2 mL/100 g/min vs. 278.2 ± 73.5 mL/100 g/min, P = 0.022) and left (194.5 ± 71.5 mL/100 g/min vs. 252.7 ± 84.4 mL/100 g/min, P = 0.020) renal cortex compared to the lcSSc patients (n = 23) after adjusting for demographics, but the significance of the difference was attenuated after further adjusting for modified Rodnan skin score and digital ulcers.

CONCLUSION

Multi-parametric MR quantitative imaging, particularly multi-delay ASL perfusion imaging, is a useful technique for characterizing the kidneys and classification of SSc patients.

Autoimmune encephalitis: Early and late findings on serial MR imaging and correlation to treatment timepoints

Introduction

MRI is negative in a large percentage of autoimmune encephalitis cases or lacks findings specific to an antibody. Even rarer is literature correlating the evolution of imaging findings with treatment timepoints. We aim to characterize imaging findings in autoimmune encephalitis at presentation and on follow up correlated with treatment timepoints for this rare disease.

Methods

A full-text radiological information system search was performed for "autoimmune encephalitis" between January 2012 and June 2022. Patients with laboratory-identified autoantibodies were included. MRI findings were assessed in correlation to treatment timepoints by two readers in consensus. For statistical analysis, cell-surface vs intracellular antibody groups were assessed for the presence of early limbic, early extralimbic, late limbic, and late extralimbic findings using the χ2 test.

Results

Thirty-seven patients (female n = 18, median age 58.8 years; range 25.7 to 82.7 years) with 15 different autoantibodies were included in the study. Twenty-three (62%) patients were MRI-negative at time of presentation; 5 of these developed MRI findings on short-term follow up. Of the 19 patients with early MRI findings, 9 (47%) demonstrated improvement upon treatment initiation (7/9 cell-surface group). There was a significant difference (p = 0.046) between the MRI spectrum of cell-surface vs intracellular antibody syndromes as cell-surface antibody syndromes demonstrated more early classic findings of limbic encephalitis and intracellular antibody syndromes demonstrated more late extralimbic abnormalities.

Conclusion

MRI can be used to help narrow the differential diagnosis in autoimmune encephalitis and can be used as a monitoring tool for certain subtypes of this rare disease.

Subcortical structure alteration in patients with drug-induced parkinsonism: Evidence from neuroimaging

Parkinson's Disease (PD) and Drug-induced parkinsonism (DIP) are the most common subtypes of parkinsonism, yet no studies have reported that the subcortical volume alterations in DIP patients. This study aimed to identify specific alterations of subcortical structures volume in DIP patients, and investigate association between the subcortical structure modifications and clinical symptoms. We recruited 27 PD patients, 25 DIP patients and 30 healthy controls (HCs). The clinical symptom-related parameters (Unified Parkinson's Disease Rating Scale, UPDRS) were evaluated. Structural imaging was performed on a 3.0 T scanner, and volumes of subcortical structures were obtained using FreeSurfer software. Analysis of covariance (ANCOVA) and partial correlation analysis were performed. DIP group had significantly smaller volume of the thalamus, pallidum, hippocampus and amygdala compared to HCs. ROC curve analysis demonstrated that the highest area under curve (AUC) value was in the right pallidum (AUC = 0.831) for evaluating the diagnostic efficacy in DIP from HCs. Moreover, the volumes of the putamen, hippocampus and amygdala were negatively correlated with UPDRSII in the DIP patients. The volume of the amygdala was negatively correlated with UPDRSIII. The present study provides novel information regarding neuroanatomical alteration of subcortical nuclei in DIP patients, suggesting that these methods might provide the basis for early diagnosis and differential diagnosis of DIP.

Dot-in-circle sign in cervical actinomycotic mycetoma: An extremely rare case report

Actinomycosis is an unusual, chronic granulomatous infection caused by Actinomycetes spp. The organism also causes mycetoma, a neglected tropical disease in endemic regions. We present a very uncommon case of extensive actinomycosis of the soft tissues in the neck with perivertebral extension that showed the dot-in-circle sign on magnetic resonance imaging. A 29-year-old male patient presented with progressively enlarging hard posterior neck swelling of 4 years duration. Subsequently, magnetic resonance imaging showed the dot-in-circle sign in an avidly enhancing infiltrative lesion with extensive involvement of the neck and perivertebral soft tissue. The pathologic examination was in line with actinomycosis. The patient responded to high doses of combination parenteral and oral antibiotics.

Fenestration of the posterior cerebral artery P1-P2 junction: A case report

Fenestration is a rare anatomical variation in the posterior cerebral artery. To the best of our knowledge, there are few reports of fenestrations at the posterior cerebral artery P1-P2 junction. Herein, we report a case of fenestration of the posterior cerebral artery P1-P2 junction diagnosed by 3-T magnetic resonance imaging and magnetic resonance angiography. A 75-year-old woman visited our hospital because of headaches. Magnetic resonance imaging incidentally showed fenestration around the P1-P2 segment of the right posterior cerebral artery. Magnetic resonance angiography revealed a small fenestration at the right posterior cerebral artery P1-P2 junction. The vessel diameter of both limbs forming the fenestration was nearly equal. Careful imaging assessment is important to identify fenestration of the posterior cerebral artery P1-P2 junction. Both magnetic resonance angiography and magnetic resonance imaging were useful for diagnosis in this case.

Magnetic resonance elastography as a preoperative assessment for predicting intrahepatic recurrence in patients with hepatocellular carcinoma

PURPOSE

Magnetic resonance elastography (MRE) is a noninvasive tool for diagnosing hepatic fibrosis with high accuracy. We investigated the preoperative clinical and imaging predictors of intrahepatic recurrence after curative resection of hepatocellular carcinoma (HCC), and evaluated MRE as a predictor of intrahepatic recurrence.

METHODS

We retrospectively evaluated 80 patients who underwent preoperative contrast-enhanced magnetic resonance imaging (MRI) with two-dimensional MRE and curative resection for treatment-naïve HCC between May 2019 and December 2021. Liver stiffness (LS) was measured on the elastograms, and the optimal cutoff of LS for predicting intrahepatic recurrence was obtained using receiver operating characteristic (ROC) analysis. An LS above this cutoff was defined as MRE-recurrence. Preoperative imaging features of the tumor were assessed on MRI, including features in the Liver Imaging Reporting and Data System and microvascular invasion (MVI). Recurrence-free survival (RFS) rates were estimated using the Kaplan-Meier method, and differences were compared using the log-rank test. Using a Cox proportional hazards model, we conducted a multivariable analysis to investigate the factors affecting recurrence-free survival.

RESULTS

During a median follow-up period of 32 months (range, 4-52 months), thirteen patients (16.3%) developed intrahepatic recurrence. ROC analysis determined an LS cutoff of ≥4.35 kPa to define MRE-recurrence. The 4-year RFS rate was significantly higher in patients without MRE-recurrence than in those with MRE-recurrence (93.4% vs. 48.9%; p = 0.001). In multivariable analysis, MRE-recurrence (Hazard ratio [HR], 5.9; 95% confidence interval [CI], 1.5-23.1) and MVI (HR, 3.4; 95% CI, 1.0-11.3) were independent predictors of intrahepatic recurrence.

CONCLUSIONS

Patients without MRE-recurrence had significantly higher RFS rates than those with MRE-recurrence. MRE-recurrence and MVI were independent predictors of intrahepatic recurrence in patients after curative resection for HCC.

Prediction of Lymphovascular invasion status in breast cancer based on magnetic resonance imaging radiomics features

OBJECTIVE

This study intended to investigate the feasibility and effectiveness of using clinical magnetic resonance imaging (MRI) radiomics features to predict lymphovascular invasion (LVI) status in breast cancer (BC) patients.

METHODS

A total of 182 BC patients were retrospectively collected and randomly divided into a training set (n = 127) and a validation set (n = 55) in a 7:3 ratio. Based on pathological examination results, the training set was further divided into LVI group (n = 60) and non-LVI group (n = 67), and the validation set was divided into LVI group (n = 24) and non-LVI group (n = 31). General data and MRI examination indicators were compared. Multivariate logistic regression was utilized to analyze MRI radiomics features and clinically relevant indicators that were significant in the baseline information of patients in training set, independent risk factors were identified, and a logistic regression model was built. The accuracy of logistic model was validated using ROC curves in training and validation sets.

RESULTS

Age, pathohistological classification, tumor length, tumor width, presence or absence of Magnetic Resonance Spectroscopy (MRS) cho peak, presence or absence of spicule sign, peritumoral enhancement, and peritumoral edema were statistically significant (P < 0.05) between the two groups. Multivariate logistic regression analysis presented that spicule and peritumoral edema were independent risk factors for LVI in BC patients (P < 0.05). The ROC curve illustrated that AUC of the logistic regression model in the training set was 0.807 (95%CI: 0.730-0.885) and that in the validation set was 0.837 (95%CI: 0.731-0.944).

CONCLUSION

Radiomics features of spicule sign and peritumoral edema were independent risk factors for LVI in BC patients. A logistic regression model based on these factors, along with age, could accurately predict LVI occurrence in BC patients, providing data support for diagnosis and modeling of LVI in BC patients.

Non-invasive differentiation of hepatic steatosis and steatohepatitis in a mouse model using nitroxyl radical as an MRI-contrast agent

Drug-induced steatohepatitis is considered more serious than drug-induced hepatic steatosis, so that differentiating between the two is crucial in drug development. In addition, early detection of drug-induced steatohepatitis is considered important since recovery is possible with drug withdrawal. However, no method has been established to differentiate between the two. In the development of drug-induced steatohepatitis, reactive oxygen species (ROS) is excessively generated in the liver. It has been reported that ROS can be monitored with electron spin resonance (ESR) and dynamic nuclear polarization-magnetic resonance imaging (DNP-MRI) by using nitroxyl radicals, which are known to participate in various in vivo redox reactions. The decay/reduction rate, which is an index for monitoring nitroxyl radicals, has been reported to be increased in tissues with excessive ROS levels other than liver, but decreased in methionine choline deficient (MCD) diet-induced steatohepatitis with excess ROS. Therefore, looking to differentiate between drug-induced hepatic steatosis and steatohepatitis, we examined whether the reduction rate decreases in steatohepatitis other than the MCD-diet induced disease and whether the decrease could be detected by MRI. We used STAM™ mice in which hepatic steatosis and steatohepatitis developed sequentially under diabetic conditions. 3-carbamoyl-PROXYL (CmP), one of the nitroxyl radicals, was injected intravenously during the MRI procedure and the reduction rate was calculated. The reduction rate was significantly higher in early steatohepatitis than in hepatic steatosis and the control. Excess ROS in early steatohepatitis was detected by an immunohistochemical marker for ROS. Therefore, it was indicated that the increase or decrease in the reduction rate in steatohepatitis differs depending on the model, and early steatohepatitis could be noninvasively differentiated from hepatic steatosis using CmP in MRI. Since the change in direction of the reduction rate in steatohepatitis in clinical studies could be predicted by confirming the reduction rate in preclinical studies, the present method, which can be used consistently in clinical and preclinical studies, warrants consideration as a candidate monitoring method for differentiating between early drug-induced steatohepatitis and hepatic steatosis in drug development.

Paclitaxel combined with trastuzumab chemotherapy-related posterior reversible encephalopathy syndrome: A case report and literature review

Posterior reversible encephalopathy syndrome (PRES) in breast carcinoma is a rare disease in clinical practice that is often misdiagnosed and ignored. This study reported a case of a patient admitted to our hospital and discussed the clinical, imaging, and pathogenesis properties of the disease. We retrospectively analyzed the clinical data of this patient and reviewed the relevant literature. Imaging was used to diagnose PRES based on clinical findings, and clinical symptoms improved after discontinuation of the relevant drugs.

Which neuroimaging and fluid biomarkers method is better in theranostic of Alzheimer's disease? An umbrella review

Biomarkers are measured to evaluate physiological and pathological processes as well as responses to a therapeutic intervention. Biomarkers can be classified as diagnostic, prognostic, predictor, clinical, and therapeutic. In Alzheimer's disease (AD), multiple biomarkers have been reported so far. Nevertheless, finding a specific biomarker in AD remains a major challenge. Three databases, including PubMed, Web of Science, and Scopus were selected with the keywords of Alzheimer's disease, neuroimaging, biomarker, and blood. The results were finalized with 49 potential CSF/blood and 35 neuroimaging biomarkers. To distinguish normal from AD patients, amyloid-beta42 (Aβ42), plasma glial fibrillary acidic protein (GFAP), and neurofilament light (NFL) as potential biomarkers in cerebrospinal fluid (CSF) as well as the serum could be detected. Nevertheless, most of the biomarkers fairly change in the CSF during AD, listed as kallikrein 6, virus-like particles (VLP-1), galectin-3 (Gal-3), and synaptotagmin-1 (Syt-1). From the neuroimaging aspect, atrophy is an accepted biomarker for the neuropathologic progression of AD. In addition, Magnetic resonance spectroscopy (MRS), diffusion weighted imaging (DWI), diffusion tensor imaging (DTI), tractography (DTT), positron emission tomography (PET), and functional magnetic resonance imaging (fMRI), can be used to detect AD. Using neuroimaging and CSF/blood biomarkers, in combination with artificial intelligence, it is possible to obtain information on prognosis and follow-up on the different stages of AD. Hence physicians could select the suitable therapy to attenuate disease symptoms and follow up on the efficiency of the prescribed drug.

Axial T1-weighted imaging of the lumbar spine: a redundancy or an asset?

OBJECTIVE

To determine the diagnostic value of axial T1-weighted imaging for patients suffering from lower back pain.

MATERIALS AND METHODS

In this retrospective study, 100 consecutive lumbar spine MRIs obtained in patients with chronic low back pain were reviewed in two sessions: First, readers viewed core sequences (sagittal T1-weighted, STIR and T2-weighted, and axial T2-weighted) with axial T1-weighted sequences, and second, readers viewed cores sequences alone. Readers recorded the presence of disc degeneration, nerve root compromise, facet joint arthritis, and stenosis at each lumbar spine level as well as the presence of lipoma of filum terminale (LFT), spondylolisthesis, transitional vertebrae, and fractures. The McNemar, Wilcoxon signed-rank, and student T tests were utilized.

RESULTS

For 100 studies, 5 spine levels were evaluated (L1-L2 through L5-S1). There were cases of disc disease (444/500 bulges, 56/500 herniations), nerve root compromise (1/500 nerve enlargement, 36/500 contact only, 20/500 displacement or compression), facet arthritis (438/500), stenosis (58/500 central canal, 64/500 lateral recess, 137/500 neuroforaminal), 6/100 LFTs, and other abnormalities (58/500 spondylolisthesis, 10/100 transitional vertebrae, 10/500 fracture/spondylolysis). There was no difference in diagnostic performance between the interpretation sessions (with and without axial T1-weighted imaging) at any level (p > 0.05), although four small additional LFTs were identified with axial T1-weighted imaging availability.

CONCLUSION

There was no clinically significant difference in the interpretation of lumbar spine MRI viewed with and without axial T1-weighted imaging, suggesting that the axial T1-weighted sequence does not add diagnostic value to routine lumbar spine MRI.

Fast 5-minute shoulder MRI protocol with accelerated TSE-sequences and deep learning image reconstruction for the assessment of shoulder pain at 1.5 and 3 Tesla

Purpose

The objective of this study was to implement a 5-minute MRI protocol for the shoulder in routine clinical practice consisting of accelerated 2D turbo spin echo (TSE) sequences with deep learning (DL) reconstruction at 1.5 and 3 Tesla, and to compare the image quality and diagnostic performance to that of a standard 2D TSE protocol.

Methods

Patients undergoing shoulder MRI between October 2020 and June 2021 were prospectively enrolled. Each patient underwent two MRI examinations: first a standard, fully sampled TSE (TSES) protocol reconstructed with a standard reconstruction followed by a second fast, prospectively undersampled TSE protocol with a conventional parallel imaging undersampling pattern reconstructed with a DL reconstruction (TSEDL). Image quality and visualization of anatomic structures as well as diagnostic performance with respect to shoulder lesions were assessed using a 5-point Likert-scale (5 = best). Interchangeability analysis, Wilcoxon signed-rank test and kappa statistics were performed to compare the two protocols.

Results

A total of 30 participants was included (mean age 50±15 years; 15 men). Overall image quality was evaluated to be superior in TSEDL versus TSES (p<0.001). Noise and edge sharpness were evaluated to be significantly superior in TSEDL versus TSES (noise: p<0.001, edge sharpness: p<0.05). No difference was found concerning qualitative diagnostic confidence, assessability of anatomical structures (p>0.05), and quantitative diagnostic performance for shoulder lesions when comparing the two sequences.

Conclusions

A fast 5-minute TSEDL MRI protocol of the shoulder is feasible in routine clinical practice at 1.5 and 3 T, with interchangeable results concerning the diagnostic performance, allowing a reduction in scan time of more than 50% compared to the standard TSES protocol.

Evaluation of multi-parameter MRI in preoperative staging of endometrial carcinoma

Background

Endometrial carcinoma (EC) is a prevalent gynecological malignancy, necessitating accurate preoperative staging for effective treatment planning. This study explores the application value of multi-parameter MRI in diagnosing and staging endometrial cancer.

Methods

Seventy-six patients diagnosed with endometrial cancer underwent 3.0 T pelvic MRI within two weeks before surgery. Imaging data were analyzed based on FIGO clinical staging criteria. The study assessed the sensitivity, specificity, positive predictive value, and negative predictive value of MRI for each stage.

Results

Postoperative pathology confirmed 71 cases of endometrial adenocarcinoma, 3 serous adenocarcinoma, and 2 clear cell carcinomas. MRI staging showed a high consistency (Kappa value = 0.786) with postoperative pathology. The overall accuracy of MRI diagnosis was 86.8%. Sensitivity and specificity varied for each stage: IA (91.3%, 96.2%), IB (88.6%, 93.8%), II (97.4%, 89.2%), and III (84.2%, 100%).

Conclusion

While there was a slight misdiagnosis rate, the overall accuracy of preoperative MRI for endometrial cancer was high, aiding in precise diagnosis and clinical staging. MRI effectively identified myometrial infiltration, cervical involvement, paracentral extension, and lymph node metastasis. Further research with larger sample sizes is recommended for enhanced reliability.

Vanishing pelvic mass: Decidualized endometriosis during pregnancy

A 40-year-old woman without history of endometriosis was found to have 10 cm pelvic mass on the routine first trimester ultrasound. Magnetic resonance imaging (MRI) of the pelvis demonstrated a large solid mass abutting the rectum which raised the concern for malignancy. Transrectal biopsy of the mass was performed with histopathology result of decidualized endometriosis. Patient continued her pregnancy and had cesarean section at 39 weeks. Interestingly, no mass was found when obstetrician performed pelvic examination after delivery in the operative room. This case is a unique presentation of endometriosis during pregnancy in a patient with no prior history of endometriosis. Large size and abutment of the rectum by the decidualized endometriosis on MRI led to misinterpretation as malignancy. Our case highlights complexity of the deep infiltrative endometriosis (DIE) during pregnancy which can misguide the providers, lead to unnecessary procedures and unwanted complications.

Quantitative T2 mapping-based longitudinal assessment of brain injury and therapeutic rescue in the rat following acute organophosphate intoxication

Acute intoxication with organophosphate (OP) cholinesterase inhibitors poses a significant public health risk. While currently approved medical countermeasures can improve survival rates, they often fail to prevent chronic neurological damage. Therefore, there is need to develop effective therapies and quantitative metrics for assessing OP-induced brain injury and its rescue by these therapies. In this study we used a rat model of acute intoxication with the OP, diisopropylfluorophosphate (DFP), to test the hypothesis that T2 measures obtained from brain magnetic resonance imaging (MRI) scans provide quantitative metrics of brain injury and therapeutic efficacy. Adult male Sprague Dawley rats were imaged on a 7T MRI scanner at 3, 7 and 28 days post-exposure to DFP or vehicle (VEH) with or without treatment with the standard of care antiseizure drug, midazolam (MDZ); a novel antiseizure medication, allopregnanolone (ALLO); or combination therapy with MDZ and ALLO (DUO). Our results show that mean T2 values in DFP-exposed animals were: (1) higher than VEH in all volumes of interest (VOIs) at day 3; (2) decreased with time; and (3) decreased in the thalamus at day 28. Treatment with ALLO or DUO, but not MDZ alone, significantly decreased mean T2 values relative to untreated DFP animals in the piriform cortex at day 3. On day 28, the DUO group showed the most favorable T2 characteristics. This study supports the utility of T2 mapping for longitudinally monitoring brain injury and highlights the therapeutic potential of ALLO as an adjunct therapy to mitigate chronic morbidity associated with acute OP intoxication.

Characterization of Blood-Brain Barrier Opening Induced by Transcranial Histotripsy in Murine Brains

OBJECTIVE

Transcranial histotripsy has shown promise as a non-invasive neurosurgical tool, as it has the ability to treat a wide range of locations in the brain without overheating the skull. One important effect of histotripsy in the brain is the blood-brain barrier (BBB) opening (BBBO) at the ablation site, but there is a knowledge gap concerning the extent of histotripsy-induced BBBO. Here we describe induction of BBBO by transcranial histotripsy and use of magnetic resonance imaging (MRI) and histology to quantify changes in BBBO at the periphery of the histotripsy ablation zone over time in the healthy mouse brain.

METHODS

An eight-element, 1 MHz histotripsy transducer with a focal distance of 32.5 mm was used to treat the brains of 23 healthy female BL6 mice. T1-gadolinium (T1-Gd) MR images were acquired immediately following histotripsy treatment and during each of the subsequent 4 wk to quantify the size and intensity of BBB leakage.

RESULTS

The T1-Gd MRI results revealed that the hyperintense BBBO volume increased over the first week and subsided gradually over the following 3 wk. Histology revealed complete loss of tight junction proteins and blood vessels in the center of the ablation region immediately after histotripsy, partial recovery in the periphery of the ablation zone 1 wk following histotripsy and near-complete recovery of tight junction complex after 4 wk.

CONCLUSION

These results provide the first evidence of transcranial histotripsy-induced BBBO and repair at the periphery of the ablation zone.

Aqueductal developmental venous anomaly causing obstructive hydrocephalus: A case report and review of the literature

A developmental venous anomaly (DVA) is a venous drainage of the associated parenchyma that is normally asymptomatic. However, a DVA located adjacent to the aqueduct can cause obstructive hydrocephalus by blocking the flow of cerebrospinal fluid. We describe a rare case of obstructive hydrocephalus due to aqueductal stenosis secondary to a DVA. A 43-year-old man presented with sudden bilateral temporal pain during weight training. Using a 3-Tesla scanner, cranial magnetic resonance imaging (MRI) was performed, and hydrocephalus was found with mild enlargement of the lateral and third ventricles. Susceptibility-weighted imaging and postcontrast MRI revealed that the DVA from the bilateral thalami narrowed the orifice of the aqueduct on its drainage route towards the vein of Galen. We assumed that force exerted during weight training may have caused dilation of the anomalous veins, leading to his symptom. A review of the relevant English-language literature yielded only 19 cases of aqueductal stenosis due to DVA. In comparison to these cases, the duration of symptom in our case was extremely short because the patient had a history of ventriculomegaly detected on plain computed tomography and was diagnosed quickly based on the characteristic finding of DVA: the caput medusae appearance.

Computational limits to the legibility of the imaged human brain

Our knowledge of the organisation of the human brain at the population-level is yet to translate into power to predict functional differences at the individual-level, limiting clinical applications and casting doubt on the generalisability of inferred mechanisms. It remains unknown whether the difficulty arises from the absence of individuating biological patterns within the brain, or from limited power to access them with the models and compute at our disposal. Here we comprehensively investigate the resolvability of such patterns with data and compute at unprecedented scale. Across 23 810 unique participants from UK Biobank, we systematically evaluate the predictability of 25 individual biological characteristics, from all available combinations of structural and functional neuroimaging data. Over 4526 GPU*hours of computation, we train, optimize, and evaluate out-of-sample 700 individual predictive models, including fully-connected feed-forward neural networks of demographic, psychological, serological, chronic disease, and functional connectivity characteristics, and both uni- and multi-modal 3D convolutional neural network models of macro- and micro-structural brain imaging. We find a marked discrepancy between the high predictability of sex (balanced accuracy 99.7%), age (mean absolute error 2.048 years, R2 0.859), and weight (mean absolute error 2.609Kg, R2 0.625), for which we set new state-of-the-art performance, and the surprisingly low predictability of other characteristics. Neither structural nor functional imaging predicted an individual's psychology better than the coincidence of common chronic disease (p < 0.05). Serology predicted chronic disease (p < 0.05) and was best predicted by it (p < 0.001), followed by structural neuroimaging (p < 0.05). Our findings suggest either more informative imaging or more powerful models will be needed to decipher individual level characteristics from the human brain. We make our models and code openly available.

Non-invasive imaging for predicting labial salivary gland biopsy outcomes in patients with suspected primary Sjögren syndrome

To identify the value of salivary gland ultrasound (SGUS) combined with magnetic resonance imaging (MRI) and magnetic resonance sialography (MRS) in predicting the results of labial salivary gland biopsy (LSGB) in patients with suspected primary Sjögren syndrome (pSS), and construct a nomogram model to predict LSGB results. A total of 181 patients who were admitted with suspected pSS from December 2018 to April 2023 were examined and divided into a training set (n = 120) and a validation set (n = 61). Baseline data of the two groups were examined, and the value of SGUS, MRI, and MRS in predicting LSGB was analyzed. Multivariate logistic analysis was used to screen for risk factors, and nomogram prediction models were constructed using these results. In the training set, the SGUS, MRI, and MRS scores of patients in the LSGB + group were higher than those in the LSGB - group (all P < 0.001). The positive prediction value (PPV) was 91% for an SGUS score of 3, and 82% for MRI and MRS scores of 2 or more. We developed a nomogram prediction model based on SGUS, MRI, and MRS data, and it had a concordance index (C-index) of 0.94. The Hosmer-Lemeshow test (χ2 = 3.17, P = 0.92) also indicated the nomogram prediction model had good accuracy and calibration for prediction of LSGB results. A nomogram model based on SGUS, MRI, and MRS results can help rheumatologists decide whether LSGB should be performed in patients with suspected pSS.

The Impact of Omitting Contralateral Systematic Biopsy on the Surgical Planning of Patients with a Unilateral Suspicious Lesion on Magnetic Resonance Imaging Undergoing Robot-assisted Radical Prostatectomy for Prostate Cancer

Background and objective

A combined approach of magnetic resonance imaging (MRI)-targeted biopsy (TBx) and bilateral systematic biopsy (SBx) is advised in patients who have an increased risk of prostate cancer (PCa). The diagnostic gain of SBx in detecting PCa for treatment planning of patients undergoing robot-assisted radical prostatectomy (RARP) is unknown. This study aims to determine the impact of omitting contralateral SBx on the surgical planning of patients undergoing RARP in terms of nerve-sparing surgery (NSS) and extended pelvic lymph node dissection (ePLND).

Methods

Case files from 80 men with biopsy-proven PCa were studied. All men had a unilateral suspicious lesion on MRI, and underwent TBx and bilateral SBx. Case files were presented to five urologists for the surgical planning of RARP. Each case file was presented randomly using two different sets of information: (1) results of TBx + bilateral SBx, and (2) results of TBx + ipsilateral SBx. The urologists assessed whether they would perform NSS and/or ePLND.

Key findings and limitations

A change in the surgical plan concerning NSS on the contralateral side was observed in 9.0% (95% confidence interval [CI] 6.4-12.2) of cases. Additionally, the indication for ePLND changed in 5.3% (95% CI 3.3-7.9) of cases. Interobserver agreement based on Fleiss' kappa changed from 0.44 to 0.15 for the indication of NSS and from 0.84 to 0.83 for the indication of ePLND.

Conclusions and clinical implications

In our series, the diagnostic information obtained from contralateral SBx has limited impact on the surgical planning of patients with a unilateral suspicious lesion on MRI scheduled to undergo RARP.

Patient summary

In patients with one-sided prostate cancer on magnetic resonance imaging, omitting biopsies on the other side rarely changed the surgical plan with respect to nerve-sparing surgery and the indication to perform extended lymph node dissection.

Ultrasound assessment of synovitis in total knee replacements: concordance with MR imaging and joint aspiration

OBJECTIVE

To determine if diagnostic ultrasound can reliably distinguish between synovial patterns of prosthetic joint infection and polymeric wear in total knee replacements.

METHODS AND METHODS

This retrospective study was approved by our hospital IRB. Using the radiology report database, MR examinations performed within a week of diagnostic ultrasound and/or ultrasound-guided aspiration were identified. This yielded (1) 24 cases with MR and ultrasound comparisons; (2) 44 cases with MR, ultrasound, and aspiration comparisons; and (3) 92 cases with ultrasound and aspiration comparisons. The MR studies were reviewed by a musculoskeletal radiologist. The ultrasound studies were each reviewed by 2 other musculoskeletal radiologists. Each study was graded for synovial pattern indicating infection, polymeric wear, or normal/nonspecific. Agreement between the MR grader and the ultrasound graders as well as ultrasound inter-rater agreement were assessed using k statistics. Sensitivity, specificity, positive predictive value, and negative predictive value of ultrasound were calculated.

RESULTS

Agreement between ultrasound and MR imaging was fair to moderate: k of 0.27 (95% CI: 0.04, 0.50) and 0.44 (95% CI: 0.23, 0.65) for ultrasound raters 1 and 2, respectively. Inter-rater agreement between ultrasound graders 1 and 2 was moderate, with k of 0.56 (95% CI: 0.42, 0.70). Ultrasound sensitivities for infection were 0.13 and 0.09, while specificities were 0.94 and 0.98. Ultrasound sensitivities for polymeric wear were 0.38 and 0.62 with specificities of 0.90 and 0.76.

CONCLUSION

Ultrasound does not perform well in distinguishing patterns of synovitis either compared to MRI or a reference standard of aspiration/clinical follow-up.

Development of a whole spinal MRI-based tumor burden scoring method in participants with multiple myeloma: a pilot study of prognostic significance

The aim of the study was to develop a new whole spinal MRI-based tumor burden scoring method in participants with newly diagnosed multiple myeloma (MM) and to explore its prognostic significance. We prospectively recruited participants with newly diagnosed MM; performed whole spinal MRI (sagittal FSE T1WI, sagittal IDEAL T2WI, and axial FLAIR T2WI) on them; and collected their clinical data, early treatment response, progression-free survival (PFS), and overall survival (OS). We developed a new tumor burden scoring method according to the extent of bone marrow infiltration in five MRI patterns. All participants were divided into good response and poor response groups after four treatment cycles. Univariate, multivariate analyses, and ROC were used to determine the performance of independent predictors. Thresholds for PFS and OS were calculated using X-tile, and their prognostic significance were assessed by Kaplan-Meier. The Kruskal-Wallis H test was used to compare the differences of tumor burden score between the revised International Staging System (R-ISS) stages. The new tumor burden scoring method was used in 62 participants (median score, 12; range, 0-18). The tumor burden score (OR 1.266, p = 0.002) was an independent predictor of poor response and the AUC was 0.838. Higher tumor burden scores were associated with shorter PFS (p = 0.002) and OS (p = 0.011). The tumor burden score was higher in R-ISS-III than in R-ISS-I and R-ISS-II (p = 0.016 and p = 0.006, respectively). The tumor burden score was an excellent predictor of prognosis and may serve as a supplemental marker for R-ISS.

The Transatlantic Recommendations for Prostate Gland Evaluation with Magnetic Resonance Imaging After Focal Therapy (TARGET): A Systematic Review and International Consensus Recommendations

BACKGROUND AND OBJECTIVE

Magnetic resonance imaging (MRI) can detect recurrences after focal therapy for prostate cancer but there is no robust guidance regarding its use. Our objective was to produce consensus recommendations on MRI acquisition, interpretation, and reporting after focal therapy.

METHODS

A systematic review was performed in July 2022 to develop consensus statements. A two-round consensus exercise was then performed, with a consensus meeting in January 2023, during which 329 statements were scored by 23 panellists from Europe and North America spanning urology, radiology, and pathology with experience across eight focal therapy modalities. Using RAND Corporation/University of California-Los Angeles methodology, the Transatlantic Recommendations for Prostate Gland Evaluation with MRI after Focal Therapy (TARGET) were based on consensus for statements scored with agreement or disagreement.

KEY FINDINGS AND LIMITATIONS

In total, 73 studies were included in the review. All 20 studies (100%) reporting suspicious imaging features cited focal contrast enhancement as suspicious for cancer recurrence. Of 31 studies reporting MRI assessment criteria, the Prostate Imaging-Reporting and Data System (PI-RADS) score was the scheme used most often (20 studies; 65%), followed by a 5-point Likert score (six studies; 19%). For the consensus exercise, consensus for statements scored with agreement or disagreement increased from 227 of 295 statements (76.9%) in round one to 270 of 329 statements (82.1%) in round two. Key recommendations include performing routine MRI at 12 mo using a multiparametric protocol compliant with PI-RADS version 2.1 standards. PI-RADS category scores for assessing recurrence within the ablation zone should be avoided. An alternative 5-point scoring system is presented that includes a major dynamic contrast enhancement (DCE) sequence and joint minor diffusion-weighted imaging and T2-weighted sequences. For the DCE sequence, focal nodular strong early enhancement was the most suspicious imaging finding. A structured minimum reporting data set and minimum reporting standards for studies detailing MRI data after focal therapy are presented.

CONCLUSIONS AND CLINICAL IMPLICATIONS

The TARGET consensus recommendations may improve MRI acquisition, interpretation, and reporting after focal therapy for prostate cancer and provide minimum standards for study reporting.

PATIENT SUMMARY

Magnetic resonance imaging (MRI) scans can detect recurrent of prostate cancer after focal treatments, but there is a lack of guidance on MRI use for this purpose. We report new expert recommendations that may improve practice.

Mesial temporal atrophy in preoperative MRI rather than steep Trendelenburg position is associated with postoperative delirium in patients undergoing a major urologic surgery

PURPOSE

To investigate whether steep Trendelenburg in a major urologic surgery is associated with postoperative delirium, and to examine other potential clinical and radiologic factors predictive of postoperative delirium.

METHODS

182 patients who received a major urologic surgery and underwent a 3.0-T brain MRI scan within 1 year prior to the date of surgery were retrospectively enrolled. Preoperative brain MRIs were used to analyze features related to small vessel disease burden and mesial temporal atrophy. Presence of a significant mesial temporal atrophy was defined as Scheltens' scale ≥ 2. Patients' clinico-demographic data and MRI features were used to identify significant predictors of postoperative delirium using the logistic regression analysis. Independent predictors found significant in the univariate analysis were further evaluated in the multivariate analysis.

RESULTS

Incidence of postoperative delirium was 6.0%. Patients with postoperative delirium had lower body mass index (21.3 vs. 25.0 kg/m2, P = 0.003), prolonged duration of anesthesia (362.7 vs. 224.7 min, P < 0.001) and surgery (302.2 vs. 174.5 min, P < 0.001), and had more significant mesial temporal atrophy (64% vs. 30%, P = 0.046). In the univariate analysis, female sex, type of surgery (radical prostatectomy over cystectomy), prolonged duration of anesthesia (≥ 6 h), and presence of a significant mesial temporal atrophy were significant predictors (all P-values < 0.050), but only the presence of significant mesial temporal atrophy was significant in the multivariate analysis [odds ratio (OR), 3.69; 95% CI 0.99-13.75; P = 0.046].

CONCLUSION

Steep Trendelenburg was not associated with postoperative delirium. Significant mesial temporal atrophy (Scheltens' scale ≥ 2) in preoperative brain MRI was predictive of postoperative delirium.

TRIAL REGISTRATION

Not applicable.

A model-based MR parameter mapping network robust to substantial variations in acquisition settings

Deep learning methods show great potential for the efficient and precise estimation of quantitative parameter maps from multiple magnetic resonance (MR) images. Current deep learning-based MR parameter mapping (MPM) methods are mostly trained and tested using data with specific acquisition settings. However, scan protocols usually vary with centers, scanners, and studies in practice. Thus, deep learning methods applicable to MPM with varying acquisition settings are highly required but still rarely investigated. In this work, we develop a model-based deep network termed MMPM-Net for robust MPM with varying acquisition settings. A deep learning-based denoiser is introduced to construct the regularization term in the nonlinear inversion problem of MPM. The alternating direction method of multipliers is used to solve the optimization problem and then unrolled to construct MMPM-Net. The variation in acquisition parameters can be addressed by the data fidelity component in MMPM-Net. Extensive experiments are performed on R2 mapping and R1 mapping datasets with substantial variations in acquisition settings, and the results demonstrate that the proposed MMPM-Net method outperforms other state-of-the-art MR parameter mapping methods both qualitatively and quantitatively.

Intrauterine exposure to chorioamnionitis and neuroanatomical alterations at term-equivalent age in preterm infants

PURPOSE

Infants born to mothers with chorioamnionitis (CAM) are at increased risk of developing adverse neurodevelopmental disorders in later life. However, clinical magnetic resonance imaging (MRI) studies examining brain injuries and neuroanatomical alterations attributed to CAM have yielded inconsistent results. We aimed to determine whether exposure to histological CAM in utero leads to brain injuries and alterations in the neuroanatomy of preterm infants using 3.0- Tesla MRI at term-equivalent age.

METHODS

A total of 58 preterm infants born before 34 weeks of gestation at Nagoya University Hospital between 2010 and 2018 were eligible for this study (CAM group, n = 21; non-CAM group, n = 37). Brain injuries and abnormalities were assessed using the Kidokoro Global Brain Abnormality Scoring system. Gray matter, white matter, and subcortical gray matter (thalamus, caudate nucleus, putamen, pallidum, hippocampus, amygdala, and nucleus accumbens) volumes were evaluated using segmentation tools (SPM12 and Infant FreeSurfer).

RESULTS

The Kidokoro scores for each category and severity in the CAM group were comparable to those observed in the non-CAM group. White matter volume was significantly smaller in the CAM group after adjusting for covariates (postmenstrual age at MRI, infant sex, and gestational age) (p = 0.007), whereas gray matter volume was not significantly different. Multiple linear regression analyses revealed significantly smaller volumes in the bilateral pallidums (right, p = 0.045; left, p = 0.038) and nucleus accumbens (right, p = 0.030; left, p = 0.004) after adjusting for covariates.

CONCLUSIONS

Preterm infants born to mothers with histological CAM showed smaller volumes in white matter, pallidum, and nucleus accumbens at term-equivalent age.

Whole-lesion assessment of volume and signal changes after sclerotherapy of extremity venous malformations

PURPOSE

To investigate quantitative changes in MRI signal intensity (SI) and lesion volume that indicate treatment response and correlate these changes with clinical outcomes after percutaneous sclerotherapy (PS) of extremity venous malformations (VMs).

METHODS

VMs were segmented manually on pre- and post-treatment T2-weighted MRI using 3D Slicer to assess changes in lesion volume and SI. Clinical outcomes were scored on a 7-point Likert scale according to patient perception of symptom improvement; treatment response (success or failure) was determined accordingly.

RESULTS

Eighty-one patients with VMs underwent 125 PS sessions. Treatment success occurred in 77 patients (95 %). Mean (±SD) changes were -7.9 ± 24 cm3 in lesion volume and -123 ± 162 in SI (both, P <.001). Mean reduction in lesion volume was greater in the success group (-9.4 ± 24 cm3) than in the failure group (21 ± 20 cm3) (P =.006). Overall, lesion volume correlated with treatment response (ρ = -0.3, P =.004). On subgroup analysis, volume change correlated with clinical outcomes in children (ρ = -0.3, P =.03), in sodium tetradecyl sulfate-treated lesions (ρ = -0.5, P =.02), and in foot lesions (ρ = -0.6, P =.04). SI change correlated with clinical outcomes in VMs treated in 1 PS session (ρ = -0.3, P =.01) and in bleomycin-treated lesions (ρ = -0.4, P =.04).

CONCLUSIONS

Change in lesion volume is a reliable indicator of treatment response. Lesion volume and SI correlate with clinical outcomes in specific subgroups.

Spatial attention-based implicit neural representation for arbitrary reduction of MRI slice spacing

Magnetic resonance (MR) images collected in 2D clinical protocols typically have large inter-slice spacing, resulting in high in-plane resolution and reduced through-plane resolution. Super-resolution technique can enhance the through-plane resolution of MR images to facilitate downstream visualization and computer-aided diagnosis. However, most existing works train the super-resolution network at a fixed scaling factor, which is not friendly to clinical scenes of varying inter-slice spacing in MR scanning. Inspired by the recent progress in implicit neural representation, we propose a Spatial Attention-based Implicit Neural Representation (SA-INR) network for arbitrary reduction of MR inter-slice spacing. The SA-INR aims to represent an MR image as a continuous implicit function of 3D coordinates. In this way, the SA-INR can reconstruct the MR image with arbitrary inter-slice spacing by continuously sampling the coordinates in 3D space. In particular, a local-aware spatial attention operation is introduced to model nearby voxels and their affinity more accurately in a larger receptive field. Meanwhile, to improve the computational efficiency, a gradient-guided gating mask is proposed for applying the local-aware spatial attention to selected areas only. We evaluate our method on the public HCP-1200 dataset and the clinical knee MR dataset to demonstrate its superiority over other existing methods.

Adding quantitative T1rho-weighted imaging to conventional MRI improves specificity and sensitivity for differentiating malignant from benign breast lesions

OBJECTIVES

To investigate the feasibility of T1rho-weighted imaging in differentiating malignant from benign breast lesions and to explore the additional value of T1rho to conventional MRI.

MATERIALS AND METHODS

We prospectively enrolled consecutive women with breast lesions who underwent preoperative T1rho-weighted imaging, diffusion-weighted imaging, and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) between November 2021 and July 2023. The T1rho, apparent diffusion coefficient (ADC), and semi-quantitative parameters from DCE-MRI were obtained and compared between benign and malignant groups. The diagnostic performance was analyzed and compared using receiver operating characteristic (ROC) curves and the Delong Test.

RESULTS

This study included 113 patients (74 malignant and 39 benign lesions). The mean T1rho value in the benign group (92.61 ± 22.10 ms) was significantly higher than that in the malignant group (72.18 ± 16.37 ms) (P < 0.001). The ADC value and time to peak (TTP) value in the malignant group (1.13 ± 0.45 and 269.06 ± 106.01, respectively) were lower than those in the benign group (1.57 ± 0.45 and 388.30 ± 81.13, respectively) (all P < 0.001). T1rho combined with ADC and TTP showed good diagnostic performance with an area under the curve (AUC) of 0.896, a sensitivity of 81.0%, and a specificity of 87.1%. The specificity and sensitivity of the combination of T1rho, ADC, and TTP were significantly higher than those of the combination of ADC and TTP (87.1% vs. 84.6%, P < 0.005; 81.0% vs. 77.0%, P < 0.001).

CONCLUSION

T1rho-weighted imaging was a feasible MRI sequence for differentiating malignant from benign breast lesions. The combination of T1rho, ADC and TTP could achieve a favorable diagnostic performance with improved specificity and sensitivity, T1rho could serve as a supplementary approach to conventional MRI.

Cerebrospinal fluid flow artifact reduction with deep learning to optimize the evaluation of spinal canal stenosis on spine MRI

PURPOSE

The aim of study was to employ the Cycle Generative Adversarial Network (CycleGAN) deep learning model to diminish the cerebrospinal fluid (CSF) flow artifacts in cervical spine MRI. We also evaluate the agreement in quantifying spinal canal stenosis.

METHODS

For training model, we collected 9633 axial MR image pairs from 399 subjects. Then, additional 104 image pairs from 19 subjects were gathered for the test set. The deep learning model was developed using CycleGAN to reduce CSF flow artifacts, where T2 TSE images served as input, and T2 FFE images, known for fewer CSF flow artifacts. Post training, CycleGAN-generated images were subjected to both quantitative and qualitative evaluations for CSF artifacts. For assessing the agreement of spinal canal stenosis, four raters utilized an additional 104 pairs of original and CycleGAN-generated images, with inter-rater agreement evaluated using a weighted kappa value.

RESULTS

CSF flow artifacts were reduced in the CycleGAN-generated images compared to the T2 TSE and FFE images in both quantitative and qualitative analysis. All raters concordantly displayed satisfactory estimation results when assessing spinal canal stenosis using the CycleGAN-generated images with T2 TSE images (kappa = 0.61-0.75) compared to the original FFE with T2 TSE images (kappa = 0.48-0.71).

CONCLUSIONS

CycleGAN demonstrated the capability to produce images with diminished CSF flow artifacts. When paired with T2 TSE images, the CycleGAN-generated images allowed for more consistent assessment of spinal canal stenosis and exhibited agreement levels that were comparable to the combination of T2 TSE and FFE images.