Cerebral white matter injury in haemodialysis patients: a cross-sectional tract-based spatial statistics and fixel-based analysis.

Background: End-stage renal disease (ESRD) patients on maintenance haemodialysis (HD) often have damage to brain white matter (WM) and cognitive impairment. However, whether this damage is caused by maintenance HD or renal dysfunction is unclear. Herein we investigate the natural progression of WM damage in patients with ESRD and the effects of HD on WM using tract-based spatial statistics (TBSS) and fixel-based analysis (FBA). Methods: Eighty-one ESRD patients, including 41 with no dialysis (ND) and 40 on HD, and 46 healthy controls (HCs) were enrolled in this study. The differences in WM among the three groups [ESRD patients with HD (ESRD-HD), ESRD patients without HD (ESRD-ND) and HCs] were analysed using TBSS and FBA. Pairwise comparison was then used to compare the differences in WM between two groups. The relationships between WM and neurocognitive assessments/clinical data were analysed in ESRD patients with and without HD. Results: The damage to WM in ESRD-ND and ESRD-HD appeared around the lateral ventricles in TBSS, while FBA reflected that the changes had extended to adjacent WM in the anterior hemisphere, with a larger region in ESRD-HD compared with ESRD-ND and the brainstem was also widely affected in ESRD-HD. The Montreal Cognitive Assessment (MoCA) scores were lower in the ESRD-HD group. RD in the body of the corpus callosum were negatively correlated with MoCA scores in both groups. Fiber density and cross-section (FDC) in the left thalamo-prefrontal projection (T_PREFL) and left and right cingulum (CGL and CGR) were positively correlated with MoCA scores in both groups. Creatinine (Cr) was positively correlated with FDC in some frontal projection fibres in the striatum and thalamus, CG and fronto-pontine tract and was positively correlated with FD mainly in premotor projection fibres in the striatum and thalamus in the ESRD-HD group. Cr was negatively correlated with mean and radial diffusivity in regions of the corona radiata in the ESRD-ND group. Conclusions: FBA is more sensitive in detecting differences between ESRD patients and HCs. When ESRD patients receive maintenance HD, the degree of WM damage may not be aggravated, but the range of damaged WM may be expanded, especially in the anterior hemisphere and brainstem. Some of these changes in the anterior hemisphere may contribute to cognitive decline.

Work stress and competency among radiology residents: the mediating effect of resilience.

Background: Competency of health professionals stands as a fundamental element in ensuring the quality of care. Notably, work stress and resilience are found to be associated with competency of health professionals. However, the role of resilience between work stress and competency remains unexplored. This study aims to investigate competency, work stress and resilience of radiology residents, and to examine the mediating effect of resilience. Methods: A cross-sectional survey was conducted among 3,666 radiology residents from 31 provinces of China in 2021. The diagnostic radiology milestones were used to measure the competency. Results of work stress and resilience were derived from self-reports and assessment of the Connor Davidson Resilience Scale (CD-RISC) respectively. To examine the association between work stress, resilience, and competency, Spearman correlation analysis and hierarchical linear regression were employed. The mediating effect of resilience was tested by bootstrapping analysis. Results: Descriptive statistics show that the average score (mean ± SD) of work stress, competency and resilience among radiology residents were 1.55 ± 1.53 (range from 0 to 6), 28.14 ± 13 (range from 0 to 81) and 5.97 ± 1.92 (range from 0 to 8), respectively. Work stress was negatively associated with competency (ß = -0.51, p < 0.001) and resilience (-0.57, p < 0.001). In particular, resilience mediated the relationship between work stress and competence, and the total mediating effect was -0.49 (= - 0.57 * 0.86), accounting for 49.06% of the total effect. Conclusion: Work stress is a significant contributor to competence among radiology residents. Resilience significantly mediated the association. This study highlights incorporating resilience training courses into the standardized residency training program to reduce intensive work stress and promote competency of radiology residents.

MRI-based radiomics model for predicting endometrial cancer with high tumor mutation burden.

PURPOSE: To evaluate the performance of MRI-based radiomics in predicting endometrial cancer (EC) with a high tumor mutation burden (TMB-H). METHODS: A total of 122 patients with pathologically confirmed EC (40 TMB-H, 82 non-TMB-H) were included in this retrospective study. Patients were randomly divided into training and testing cohorts in a ratio of 7:3. Radiomics features were extracted from sagittal T2-weighted images and contrast-enhanced T1-weighted images. Then, the logistic regression (LR), random forest (RF), and support vector machine (SVM) algorithms were used to construct radiomics models. The area under the receiver operating characteristic curve (AUC) was calculated to evaluate the diagnostic performance of each model, and decision curve analysis was used to determine their clinical application value. RESULTS: Four radiomics features were selected to build the radiomics models. The three models had similar performance, achieving 0.771 (LR), 0.892 (RF), and 0.738 (SVM) in the training cohort, and 0.787 (LR), 0.798 (RF), and 0.777 (SVM) in the testing cohort. The decision curve demonstrated the good clinical application value of the LR model. CONCLUSIONS: The MRI-based radiomics models demonstrated moderate predictive ability for TMB-H EC and thus may be a tool for preoperative, noninvasive prediction of TMB-H EC.

Disruption of network hierarchy pattern in bulimia nervosa reveals brain information integration disorder.

The human brain works as a hierarchical organization that is a continuous axis spanning sensorimotor cortex to transmodal cortex (referring to cortex that integrates multimodal sensory information and participates in complex cognitive functions). Previous studies have demonstrated abnormalities in several specific networks that may account for their multiple behavioral deficits in patients with bulimia nervosa (BN), but whether and how the network hierarchical organization changes in BN remain unknown. This study aimed to investigate alterations of the hierarchy organization in BN network and their clinical relevance. Connectome gradient analyses were applied to depict the network hierarchy patterns of fifty-nine patients with BN and thirty-nine healthy controls (HCs). Then, we evaluated the network- and voxel-level gradient alterations of BN by comparing gradient values in each network and each voxel between patients with BN and HCs. Finally, the association between altered gradient values and clinical variables was explored. In the principal gradient, patients with BN exhibited reduced gradient values in dorsal attention network and increased gradient values in subcortical regions compared to HCs. In the secondary gradient, patients with BN showed decreased gradient values in ventral attention network and increased gradient values in limbic network. Regionally, the areas with altered principal or secondary gradient values in BN group were mainly located in transmodal networks, i.e., the default-mode and frontoparietal network. In BN group, the principal gradient values of right inferior frontal gyrus were negatively associated with external eating behavior. This study revealed the disordered network hierarchy patterns in patients with BN, which suggested a disturbance of brain information integration from attention network and subcortical regions to transmodal networks in these patients. These findings may provide insight into the neurobiological underpinnings of BN.

Spatial position changes in the semicircular canals may be the anatomical basis of Meniere's disease: a preliminary study based on ultra-high-resolution computed tomography (CT) and intelligent segmentation.

Background: Meniere's disease (MD) is an ear-related vestibular disorder accompanied by vertigo, hearing loss, and tinnitus. The anatomical structure and spatial position of the semicircular canals are important for understanding vestibular function and disease; however, research on MD and the effect of anatomical changes in the semicircular canals is limited. This study explored the relationship between the spatial location of the semicircular canals and MD using ultra-high-resolution computed tomography (U-HRCT) and intelligent segmentation. Methods: Isotropic U-HRCT images obtained from patients with MD and healthy controls (HCs) were retrospectively analyzed. We extracted the semicircular canal structures and extracted their skeleton. The plane of the skeleton of each semicircular canal was fitted separately. The mutual angles between the semicircular canals, and the angles between each semicircular canal and each plane of the coordinate system were measured. Results: Among 45 MD-affected ears (MDAEs), 33 MD-healthy ears (MDHEs), and 45 HC ears, the angle between the superior and lateral semicircular canals (LSCs) and the angle between the superior and posterior semicircular canals (PSCs) were larger in the MDAE and MDHE groups than the HC group (P<0.01), while the angle between the posterior and LSCs was smaller in the MDAE group than the HC group (P<0.001). The angles between the superior and PSCs and coronal plane (CP) of the coordinate system were significantly smaller in the MDAE and MDHE groups than the HC group (P<0.01); however, the angles between the LSC and axial plane and CP were significantly larger in the MDAE and MDHE groups than the HC group (P<0.001). Conclusions: Spatial position changes in the semicircular canals may be the anatomical basis of MD.

Knowledge-Augmented Deep Learning for Segmenting and Detecting Cerebral Aneurysms With CT Angiography: A Multicenter Study.

Background Deep learning (DL) could improve the labor-intensive, challenging processes of diagnosing cerebral aneurysms but requires large multicenter data sets. Purpose To construct a DL model using a multicenter data set for accurate cerebral aneurysm segmentation and detection on CT angiography (CTA) images and to compare its performance with radiology reports. Materials and Methods Consecutive head or head and neck CTA images of suspected unruptured cerebral aneurysms were gathered retrospectively from eight hospitals between February 2018 and October 2021 for model development. An external test set with reference standard digital subtraction angiography (DSA) scans was obtained retrospectively from one of the eight hospitals between February 2022 and February 2023. Radiologists (reference standard) assessed aneurysm segmentation, while model performance was evaluated using the Dice similarity coefficient (DSC). The model's aneurysm detection performance was assessed by sensitivity and comparing areas under the receiver operating characteristic curves (AUCs) between the model and radiology reports in the DSA data set with use of the DeLong test. Results Images from 6060 patients (mean age, 56 years ± 12 [SD]; 3375 [55.7%] female) were included for model development (training: 4342; validation: 1086; and internal test set: 632). Another 118 patients (mean age, 59 years ± 14; 79 [66.9%] female) were included in an external test set to evaluate performance based on DSA. The model achieved a DSC of 0.87 for aneurysm segmentation performance in the internal test set. Using DSA, the model achieved 85.7% (108 of 126 aneurysms [95% CI: 78.1, 90.1]) sensitivity in detecting aneurysms on per-vessel analysis, with no evidence of a difference versus radiology reports (AUC, 0.93 [95% CI: 0.90, 0.95] vs 0.91 [95% CI: 0.87, 0.94]; P = .67). Model processing time from reconstruction to detection was 1.76 minutes ± 0.32 per scan. Conclusion The proposed DL model could accurately segment and detect cerebral aneurysms at CTA with no evidence of a significant difference in diagnostic performance compared with radiology reports. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Payabvash in this issue.

Delicate and thin fibrous septa indicate a regression tendency in metabolic dysfunction-associated steatohepatitis patients with advanced fibrosis.

BACKGROUND AND AIMS: Metabolic dysfunction-associated steatohepatitis (MASH)-related fibrosis is reversible. However, the dynamic morphology change in fibrosis regression remains unclear. We aim to explore the morphological characteristics of fibrosis regression in advanced MASH patients. METHODS: Clinical and histological data of 79 biopsy-proved MASH patients with advanced fibrosis (F3-F4) were reviewed. The second harmonic generation/two-photon excitation fluorescence (SHG/TPEF) image technology was used to quantitatively identify the R (regressive) septa from P (progressive) septa and PS (perisinusoidal) fibrosis. Non-invasive tests were used to compare the fibrosis level with and without R septa groups. Transcriptomics was used to explore hub genes and the underlying mechanism of the formation of R septa. RESULTS: The R septa were different from the P septa and PS fibrosis in detail collagen quantitation identified by SHG/TPEF technology. The R septa were found in MASH fibrosis-regressed patients, which met the definition of the "Beijing classification". Therefore, patients were divided into two groups according to septa morphology: with R septa (n = 10, 12.7%), and without R septa (n = 69, 87.3%). Patients with R septa had lower values in most non-invasive tests, especially for liver stiffness assessed by TE (12.3 vs. 19.4 kPa, p = 0.010) and FAST (FibroScan®-AST) score (0.43 vs. 0.70, p = 0.003). Transcriptomics analysis showed that the expressions of five hub fibrogenic genes, including Col3A1, BGN, Col4A1, THBS2, and Col4A2 in the R septa group, were significantly lower. CONCLUSIONS: The R septa can be differentiated from the P septa and PS fibrosis by quantitative assessment of SHG/TPEF, and it represents a tendency of fibrosis regression in MASH patients. TRIAL REGISTRATION: NCT03386890, 29/12/2017.

A Feasibility Study of AI-Assisted Compressed Sensing in Prostate T2-Weighted Imaging.

RATIONALE AND OBJECTIVES: To evaluate the image quality and PI-RADS scoring performance of prostate T2-weighted imaging (T2WI) based on AI-assisted compressed sensing (ACS). MATERIALS AND METHODS: In this prospective study, adult male urological outpatients or inpatients underwent prostate MRI, including T2WI, diffusion-weighted imaging and apparent diffusion coefficient maps. Three accelerated scanning protocols using parallel imaging (PI) and ACS: T2WIPI, T2WIACS1 and T2WIACS2 were evaluated through comparative analysis. Quantitative analysis included signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), slope profile, and edge rise distance (ERD). Image quality was qualitatively assessed using a five-point Likert scale (ranging from 1 = non-diagnostic to 5 = excellent). PI-RADS scores were determined for the largest or most suspicious lesions in each patient. The Friedman test and one-way ANOVA with post hoc tests were utilized for group comparisons, with statistical significance set at P < 0.05. RESULTS: This study included 40 participants. Compared to PI, ACS reduced acquisition time by over 50%, significantly enhancing the CNR of sagittal and axial T2WI (P < 0.05), significantly improving the image quality of sagittal and axial T2WI (P < 0.05). No significant differences were observed in slope profile, ERD, and PI-RADS scores between groups (P > 0.05). CONCLUSION: ACS reduced prostate T2WI acquisition time by half while improving image quality without affecting PI-RADS scores.

Bulimia nervosa selectively reshapes the structure and intrinsic function of anterior insula subregions associated with cognition-emotion integration.

BACKGROUND: Existing evidence suggests that anterior insula plays a crucial role in cognitive control and emotional regulation and is implicated in the onset and maintenance of bulimia nervosa (BN). However, it remains unclear how structural and functional abnormalities in specific subregions of anterior insula contribute to BN. METHODS: In this study, we analyzed structural MRI and resting-state functional MRI data from 54 BN patients and 56 healthy controls (HCs). We conducted voxel-based morphometry, amplitude of low frequency fluctuation (conventional band: 0.01-0.08 Hz, slow-5: 0.01-0.027 Hz) and seed-based whole-brain functional connectivity (FC) analysis of the anterior insula subregions for both groups. Additionally, we investigated the correlation between neuroimaging findings and clinical characteristics in the BN group. RESULTS: Our findings revealed that BN patients exhibited reduced gray matter volume in the right dorsal anterior insula (dAI) and bilateral ventral anterior insula (vAI) and demonstrated decreased ALFF in slow-5 band of bilateral dAI. The BN group also showed increased FC between bilateral dAI and precuneus or right superior frontal gyri which significantly correlated with the severity of BN or its key symptom. In addition, the decreased FC between bilateral vAI and anterior cingulate and paracingulate gyri and/or median cingulate and paracingulate gyri were both significantly correlated with the severity and its restrained eating behavior. CONCLUSIONS: Our findings further indicate that the functional separation of anterior insula subregions may underlie the pathophysiology of BN. Notably, the vAI associated with emotional processing may serve as a promising neuroimaging biomarker which could inform therapeutic strategy.

Paracetamol-induced Liver Injury in An Experimental Rat Model: Noninvasive Assessment using DKI and the Effect of Gadoxetate on DKI Parameters.

PURPOSE: To explore the potential of diffusion kurtosis imaging (DKI) for assessing the degree of liver injury in a paracetamol-induced rat model and to simultaneously investigate the effect of intravenous gadoxetate on DKI parameters. METHODS: Paracetamol was used to induce hepatoxicity in 39 rats. The rats were pathologically classified into 3 groups: normal (n=11), mild necrosis (n=18), and moderate necrosis (n=10). DKI was performed before and, 15 min, 25 min, and 45 min after gadoxetate administration. Repeated-measures ANOVA with Tukey's multiple comparison test was used to investigate the effect of gadoxetate on mean diffusivity (MD) and mean diffusion kurtosis (MK) and to assess the differences in MD and MK among the three groups. A receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic accuracy of the MD values when discriminating between the necrotic groups. RESULTS: Gadoxetate had no significant effect on either the MD or the MK, and the effect size was small. The MD in the moderate necrosis group was significantly lower than that in the other two groups (F = 13.502, p < 0.001; η2 = 0.428 [95% CI: 0.082-0.637]), while the MK did not significantly differ among the three groups (F = 2.702, p = 0.081; η2 = 0.131 [95% CI: 0.001-0.4003]). The AUCs of MD for discriminating the moderate necrosis or normal group from the other groups were 0.921 (95% CI: 0.832-1.000) and 0.831 (95% CI: 0.701-0.961), respectively. CONCLUSION: It would be better to measure the MD and MK before gadoxetate injection. MD showed potential for assessing the degree of liver necrosis in a paracetamol-induced liver injury rat model.

Multiparametric magnetic resonance imaging-based assessment of the effect of adenomyosis on determining the depth of myometrial invasion in endometrial cancer.

Background: Accurate preoperative diagnosis of endometrial cancer (EC) with deep myometrial invasion (DMI) is critical to deciding whether to perform lymphadenectomy. However, the presence of adenomyosis makes distinguishing DMI from superficial myometrial invasion (SMI) on magnetic resonance imaging (MRI) challenging. We aimed to evaluate the accuracy of multiparametric MRI (mpMRI) in diagnosing DMI in EC coexisting with adenomyosis (EC-A) compared with EC without coexisting adenomyosis and to evaluate the effect of different adenomyosis subtypes on myometrial invasion (MI) depth in EC. Methods: Patients with histologically confirmed International Federation of Gynecology and Obstetrics (FIGO) stage I EC who underwent preoperative MRI were consecutively included in this 2-center retrospective study. Institution 1 was searched from January 2017 to November 2022 and institution 2 was searched from June 2017 to March 2021. Patients were divided into 2 groups: group A, patients with EC-A; group B, EC patients without coexisting adenomyosis, matched 1:2 according to age ±5 years and tumor grade. A senior radiologist assessed the MRI adenomyosis classification in group A. Then, 2 radiologists (R1/R2) independently interpreted T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), T1-weighted contrast-enhanced (T1CE), and a combination of all images (mpMRI) respectively, and then assessed MI depth. Accuracy, sensitivity, specificity, and the areas under the receiver operating curve (AUC) were calculated. The chi-square test was used to compare the accuracy of diagnosing DMI. Interobserver agreement was evaluated using the Kappa test. Results: A total of 70 cases in group A and 140 cases in group B were included. The accuracy, sensitivity, and specificity of consensus were 94.3% [95% confidence interval (CI): 88.9-99.7%] vs. 92.1% (95% CI: 87.7-96.6%), 60.0% (95% CI: 17-92.7%) vs. 86.7% (95% CI: 68.4-95.6%), and 96.9% (95% CI: 88.4-95.5%) vs. 93.6% (95% CI: 86.8-97.2%) (group A vs. group B, respectively). There was no significant difference in the diagnostic accuracy of DMI on each sequence between the groups (Reviewer 1/Reviewer 2): PT2WI=0.14/0.17, PDWI=0.50/0.33, PT1CE=0.90/0.18, PmpMRI=0.50/0.37. The AUC for T2WI, DWI, T1CE, and mpMRI (Reviewer 1/Reviewer 2), respectively, were 0.54 (95% CI: 0.42-0.66)/0.78 (95% CI: 0.67-0.87), 0.63 (95% CI: 0.50-0.74)/0.77 (95% CI: 0.65-0.86), 0.69 (95% CI: 0.57-0.80)/0.79 (95% CI: 0.68-0.88), and 0.91 (95% CI: 0.82-0.97)/0.89 (95% CI: 0.79-0.95) (group A) and 0.83 (95% CI: 0.76-0.89)/0.85 (95% CI: 0.78-0.90), 0.83 (95% CI: 0.76-0.89)/0.86 (95% CI: 0.79-0.91), 0.88 (95% CI: 0.82-0.93)/0.86 (95% CI: 0.80-0.92), and 0.91 (95% CI: 0.85-0.95)/0.87 (95% CI: 0.80-0.92) (group B). Interobserver agreement was highest with mpMRI [κ=0.387/0.695 (case/control)]. The consensus results of MRI categorization of adenomyosis revealed no significant difference in the accuracy of diagnosing DMI by adenomyosis subtype (Pspatial relationship>0.99, Paffected area=0.52, Paffected pattern=0.58, Paffected size>0.99). Conclusions: The presence of adenomyosis or adenomyosis subtype had no significant effect on the interpretation of the depth of MI. T1CE can increase the contrast between adenomyosis and cancer foci; therefore, the information provided by T1CE should be valued.

Amygdala structural and functional reorganization as an indicator of affective dysfunction in patients with tinnitus.

The aim of this study was to systematically investigate structural and functional alterations in amygdala subregions using multimodal magnetic resonance imaging (MRI) in patients with tinnitus with or without affective dysfunction. Sixty patients with persistent tinnitus and 40 healthy controls (HCs) were recruited. Based on a questionnaire assessment, 26 and 34 patients were categorized into the tinnitus patients with affective dysfunction (TPAD) and tinnitus patients without affective dysfunction (TPWAD) groups, respectively. MRI-based measurements of gray matter volume, fractional anisotropy (FA), fractional amplitude of low-frequency fluctuations (fALFF), regional homogeneity (ReHo), degree centrality (DC), and functional connectivity (FC) were conducted within 14 amygdala subregions for intergroup comparisons. Associations between the MRI properties and clinical characteristics were estimated via partial correlation analyses. Compared with that of the HCs, the TPAD and TPWAD groups exhibited significant structural and functional changes, including white matter integrity (WMI), fALFF, ReHo, DC, and FC alterations, with more pronounced WMI changes in the TPAD group, predominantly within the left auxiliary basal or basomedial nucleus (AB/BM), right central nucleus, right lateral nuclei (dorsal portion), and left lateral nuclei (ventral portion containing basolateral portions). Moreover, the TPAD group exhibited decreased FC between the left AB/BM and left middle occipital gyrus and right superior frontal gyrus (SFG), left basal nucleus and right SFG, and right lateral nuclei (intermediate portion) and right SFG. In combination, these amygdalar alterations exhibited a sensitivity of 65.4% and specificity of 96.9% in predicting affective dysfunction in patients with tinnitus. Although similar structural and functional amygdala remodeling were observed in the TPAD and TPWAD groups, the changes were more pronounced in the TPAD group. These changes mainly involved alterations in functionality and white matter microstructure in various amygdala subregions; in combination, these changes could serve as an imaging-based predictor of emotional disorders in patients with tinnitus.

Reduced functional connectivity induced by longitudinal alterations of structure and perfusion may be associated with cognitive impairment in patients on maintenance hemodialysis.

Hemodialysis (HD) leads to cognitive impairment; however, the pathophysiology of maintenance HD remains unclear. This study aimed to investigate the longitudinal alterations in gray matter volume (GMV) and cerebral blood flow (CBF) in patients on HD at follow-up compared with baseline, examine the alterations in functional connectivity (FC) by defining co-changed brain regions as seed points, and investigate the correlation between the co-changed brain regions and neuropsychological test scores. Twenty-seven patients with HD and 30 healthy controls were enrolled in this study. All participants underwent high-resolution T1-weighted imaging, arterial spin labeling, and functional MR imaging to measure GMV, CBF, and FC. The patients on HD were assessed at baseline and 3 years subsequently. The right and left medial superior frontal gyrus (SFGmed.L) exhibited significantly lower GMV and CBF in patients on HD at follow-up compared with baseline and lower FC between the SFGmed.L and left middle temporal gyrus (MTG.L). FC between the SFGmed.L and MTG.L was positively correlated with neuropsychological test scores in the HD group at follow-up. Reduced GMV and CBF may result in decreased FC between the SFGmed.L and MTG.L, which may be associated with cognitive impairment in patients on maintenance HD. Our findings provide unique insights into the pathological mechanisms of patients on maintenance HD with cognitive impairment.

Mendelian randomization analyses support causal relationships between tinnitus of different stages and severity and structural characteristics of specific brain regions.

This study aims to delineate the causal relationships between idiopathic tinnitus in different stages and severity and the morphological properties in specific brain regions. We utilized a two-sample bidirectional Mendelian randomization (MR) analysis to ascertain the causal effects of brain structural attributes on varying severities and stages of tinnitus. Our approach involved harnessing genetic variables derived from extensive genome-wide association studies as instrumental variables, centered mainly on pertinent single-nucleotide polymorphisms associated with tinnitus. Subsequently, we integrated this data with brain structural imaging inputs to facilitate the MR analysis. We also applied reverse MR analysis to pinpoint the critical brain regions implicated in the onset of tinnitus. Our analysis revealed a demonstrable causal relationship between tinnitus and brain structural alterations, including changes primarily within the auditory cortex and hub regions of the limbic system, as well as portions of the frontal-temporal-occipital circuit. We found that individuals exhibiting cortical thickness alterations in the bilateral peri-calcarine and right superior occipital gyrus might have previously experienced tinnitus. Changes in the cortical areas of the right rectus, left inferior frontal gyrus, and right pars-orbitalis appeared unrelated to tinnitus. Furthermore, moderate tinnitus patients showed more pronounced structural alterations. This study substantiates that tinnitus could instigate substantial structural alterations mainly within the auditory-limbic-frontal-visual system, while the reciprocal causality was not supported. Moreover, the data underscores that moderate, rather than severe, tinnitus precipitates the most significant structural changes. Morphological alterations in several specific brain areas either indicate a history of tinnitus or bear no relation to it.

Comparative analysis of vestibular endolymphatic hydrops grading methods and hearing loss in Ménière's disease: a retrospective MRI study using 3D-real inversion recovery sequence.

PURPOSE: To compare the correlation between different grading methods of vestibular endolymphatic hydrops (EH) and the severity of hearing loss in Ménière's disease (MD), and evaluate the diagnostic value of these methods in diagnosing MD. METHODS: This retrospective study included 30 patients diagnosed with MD from June 2021 to August 2023. All patients underwent inner ear MR gadolinium-enhanced imaging using three-dimensional (3D)-real inversion recovery sequences and pure-tone audiometry. The EH levels were independently evaluated according to the classification methods outlined by Nakashima et al. (Acta Otolaryngol Suppl 5-8, 2009. https://doi.org/10.1080/00016480902729827 ) (M1), Fang et al. (J Laryngol Otol 126:454-459, 2012. https://doi.org/10.1017/S0022215112000060 ) (M2), Barath et al. (Am J Neuroradiol 35:1387-1392, 2014. https://doi.org/10.3174/ajnr.A3856 ), (M3), Liu et al. (Front Surg 9:874971, 2022. https://doi.org/10.3389/fsurg.2022.874971 ), (M4), and Bernaerts et al. (Neuroradiology 61:421-429, 2019. https://doi.org/10.1007/s00234-019-02155-7 ) (M5), with a subsequent comparison of interobserver agreement. After achieving a consensus, an analysis was performed to explore the correlations between vestibular EH grading using different methods, the average hearing thresholds at low-mid, high-, and full frequencies and clinical stages. The diagnostic capabilities of these methods for MD were then compared. RESULTS: The interobserver consistency of M2-M5 was superior to that of M1. The EH grading based on M4 showed a significant correlation with the average hearing thresholds at low-mid, high-, and full frequencies and clinical stages. M1, M2, M3, and M5 correlated with some parameters. A receiver operating characteristic curve analysis indicated that M5 significantly outperformed M1, M2, M3, and M4 in terms of diagnostic efficiency for MD. CONCLUSION: M4 showed the strongest correlation with the degree of hearing loss in patients with MD, whereas M5 showed the highest diagnostic performance.

Posterior circulation ischemic stroke: radiomics-based machine learning approach to identify onset time from magnetic resonance imaging.

PURPOSE: Posterior circulation ischemic stroke (PCIS) possesses unique features. However, previous studies have primarily or exclusively relied on anterior circulation stroke cases to build machine learning (ML) models for predicting onset time. To date, there is no research reporting the effectiveness and stability of ML in identifying PCIS onset time. We aimed to build diffusion-weighted imaging-based ML models to identify the onset time of PCIS patients. METHODS: Consecutive PCIS patients within 24 h of definite symptom onset were included (112 in the training set and 49 in the independent test set). Images were processed as follows: volume of interest segmentation, image feature extraction, and feature selection. Five ML models, naïve Bayes, logistic regression, tree ensemble, k-nearest neighbor, and random forest, were built based on the training set to estimate the stroke onset time (binary classification: ≤ 4.5 h or > 4.5 h). Relative standard deviations (RSD), receiver operating characteristic (ROC) curves, and the calibration plot was performed to evaluate the stability and performance of the five models. RESULTS: The random forest model had the best performance in the test set, with the highest area under the curve (AUC, 0.840; 95% CI: 0.706, 0.974). This model also achieved the highest accuracy, sensitivity, specificity, positive predictive value, and negative predictive value (83.7%, 64.3%, 91.4%, 75.0%, and 86.5%, respectively). Furthermore, the model had high stability (RSD = 0.0094). CONCLUSION: The PCIS case-based ML model was effective for estimating the symptom onset time and achieved considerably high specificity and stability.

Deep learning-based arterial subtraction images improve the detection of LR-TR algorithm for viable HCC on extracellular agents-enhanced MRI.

PURPOSE: To determine the role of deep learning-based arterial subtraction images in viability assessment on extracellular agents-enhanced MRI using LR-TR algorithm. METHODS: Patients diagnosed with HCC who underwent locoregional therapy were retrospectively collected. We constructed a deep learning-based subtraction model and automatically generated arterial subtraction images. Two radiologists evaluated LR-TR category on ordinary images and then evaluated again on ordinary images plus arterial subtraction images after a 2-month washout period. The reference standard for viability was tumor stain on the digital subtraction hepatic angiography within 1 month after MRI. RESULTS: 286 observations of 105 patients were ultimately enrolled. 157 observations were viable and 129 observations were nonviable according to the reference standard. The sensitivity and accuracy of LR-TR algorithm for detecting viable HCC significantly increased with the application of arterial subtraction images (87.9% vs. 67.5%, p < 0.001; 86.4% vs. 75.9%, p < 0.001). And the specificity slightly decreased without significant difference when the arterial subtraction images were added (84.5% vs. 86.0%, p = 0.687). The AUC of LR-TR algorithm significantly increased with the addition of arterial subtraction images (0.862 vs. 0.768, p < 0.001). The arterial subtraction images also improved inter-reader agreement (0.857 vs. 0.727). CONCLUSION: Extended application of deep learning-based arterial subtraction images on extracellular agents-enhanced MRI can increase the sensitivity of LR-TR algorithm for detecting viable HCC without significant change in specificity.

Intravoxel incoherent motion assessment of liver fibrosis staging in MASLD.

PURPOSE: Partial correlation analysis was performed to account for the interference of steatosis changes and inflammatory factors, to determine the true correlation between fibrosis and IVIM parameters (Dfast, Dslow, and F), and to evaluate the diagnostic efficacy of IVIM for liver fibrosis. METHODS: A total of 106 patients with metabolic dysfunction-associated steatotic liver disease (MASLD) examined by IVIM from November 2016 to November 2023 at our hospital were retrospectively included. Preliminary analysis of each IVIM parameter and correlations with pathological findings were performed using Spearman correlation analysis, and partial correlation analysis was used to exclude the interference of other pathological factors, thus yielding the true correlations between IVIM parameters (Dfast, Dslow, and F) and pathology. The diagnostic efficacy of IVIM parameters for diagnosing MASLD was assessed via receiver operating characteristic (ROC) curve analysis. RESULTS: Spearman correlation analysis of all the IVIM parameters revealed correlations with steatosis, lobular inflammation, and ballooning. Partial correlation analysis indicated that Dfast was correlated with the pathological fibrosis stage (r = - 0.593, P < 0.001), Dslow was correlated with the pathological steatosis score (r = - 0.313, P < 0.05), and F was correlated with the pathological fibrosis stage and steatosis score (r = - 0.456 and 0.255, P < 0.001 and P < 0.05). In the diagnosis of hepatic fibrosis, significant hepatic fibrosis, advanced liver fibrosis and cirrhosis, Dfast achieved areas under the ROC curve of 0.763, 0.801, 0.853, and 0.897, respectively. The threshold values for diagnosing different fibrosis stages using Dfast (10-3 mm2/s) were 57.613, 54.587, 52.714, and 51.978, respectively. CONCLUSION: According to our partial correlation analysis, there was a moderate correlation between Dfast and F according to fibrosis stage, and Dfast was not influenced by inflammation or steatosis when diagnosing fibrosis in MASLD patients. A relatively close Dfast threshold is insufficient for accurately and noninvasively assessing various stages of MASLD fibrosis. In clinical practice, this approach can be considered an alternative method for the preliminary assessment of fibrosis in MASLD patients.

Algorithms for classification of sequences and segmentation of prostate gland: an external validation study.

OBJECTIVES: The aim of the study was to externally validate two AI models for the classification of prostate mpMRI sequences and segmentation of the prostate gland on T2WI. MATERIALS AND METHODS: MpMRI data from 719 patients were retrospectively collected from two hospitals, utilizing nine MR scanners from four different vendors, over the period from February 2018 to May 2022. Med3D deep learning pretrained architecture was used to perform image classification,UNet-3D was used to segment the prostate gland. The images were classified into one of nine image types by the mode. The segmentation model was validated using T2WI images. The accuracy of the segmentation was evaluated by measuring the DSC, VS,AHD.Finally,efficacy of the models was compared for different MR field strengths and sequences. RESULTS: 20,551 image groups were obtained from 719 MR studies. The classification model accuracy is 99%, with a kappa of 0.932. The precision, recall, and F1 values for the nine image types had statistically significant differences, respectively (all P < 0.001). The accuracy for scanners 1.436 T, 1.5 T, and 3.0 T was 87%, 86%, and 98%, respectively (P < 0.001). For segmentation model, the median DSC was 0.942 to 0.955, the median VS was 0.974 to 0.982, and the median AHD was 5.55 to 6.49 mm,respectively.These values also had statistically significant differences for the three different magnetic field strengths (all P < 0.001). CONCLUSION: The AI models for mpMRI image classification and prostate segmentation demonstrated good performance during external validation, which could enhance efficiency in prostate volume measurement and cancer detection with mpMRI. CLINICAL RELEVANCE STATEMENT: These models can greatly improve the work efficiency in cancer detection, measurement of prostate volume and guided biopsies.