Background:Large language models (LLMs) have shown considerable potential for extracting information from free-text radiology reports, enabling efficient data use, large-scale data mining, and a wide range of secondary analyses and clinical applications. This study aimed to evaluate the performance of LLMs in extracting diagnostically relevant information from multicenter free-text liver magnetic resonance imaging (MRI) reports, explore the clinical utility of LLM-generated structured reports, and investigate optimal prompting strategies for multicenter data. Methods:In this retrospective multicenter study, 800 free-text liver MRI reports from four medical centers (Beijing Friendship Hospital, Tianjin Medical University General Hospital, The Second Affiliated Hospital of Xi'an Jiaotong University, Sir Run Run Shaw Hospital) were collected to evaluate the information extraction performance of two LLMs-DeepSeek-V3 and ChatGPT-4o-using radiologist-annotated structured data as the reference standard. Three prompting strategies were applied: zero-shot prompting, global few-shot prompting (using shared examples across centers), and center-specific few-shot prompting (using examples specific to each center), with example counts set to 2-12. Model performance was evaluated using field-level F1 scores, and report-level extraction success was defined as the correct extraction of ≥80% report fields. Additionally, an exploratory clinical evaluation was conducted using 20 reports from one center, in which 10 radiologists and 10 clinicians evaluated the readability and clinical usability of free-text, manually structured, and LLM-generated reports on a 5-point Likert scale. Results:Few-shot prompting significantly outperformed zero-shot prompting for both LLMs, with the largest gains in macro F1 observed when k was increased from 0 to 2 (∆DeepSeek-V3: global 0.106, center-specific 0.127; ∆ ChatGPT-4o: global 0.086, center-specific 0.107). Performance plateaued at k=4 [DeepSeek-V3: global 0.848 (0.838-0.858), center-specific 0.865 (0.856-0.875); ChatGPT-4o: global 0.835 (0.824-0.845), center-specific 0.861 (0.851-0.870)], with adjacent-k gains <0.01. Center-specific prompting consistently outperformed global prompting (∆F1: 0.017-0.024 for DeepSeek-V3; 0.014-0.026 for ChatGPT-4o). In the exploratory clinical evaluation, structured reports received higher scores for clarity and communication than free-text reports (both P<0.001), while LLM-generated reports received scores comparable to those of manually structured reports (both P>0.05). Conclusions:LLMs demonstrated strong performance in extracting diagnostically relevant information from Chinese multicenter liver MRI reports. In the exploratory clinical evaluation, the LLM-generated structured reports showed the potential to improve report clarity and facilitate clinical communication. Global prompting showed good performance across centers, while center-specific prompting further improved accuracy by adapting to local reporting styles.

Objective:To analyze the correlations between multimodal MRI examination parameters - apparent diffusion coefficient (ADC), cerebral blood flow (CBF), and choline (Cho)/creatine (Cr)-and microvessel density (MVD) in patients with glioma, as well as the prognostic value of these parameters. Methods:The medical records of 200 patients diagnosed with glioma at our hospital from January 2019 to January 2024 were retrospectively analyzed. According to the WHO classification, patients were divided into a low-grade group (n = 122) and a high-grade group (n = 78). Patients were further divided into a good prognosis group (n = 114) and a poor prognosis group (n = 86). The Pearson correlation test was used to analyze the relationship between multimodal MRI parameters and MVD in glioma patients. The area under the curve (AUC) of receiver operating characteristic (ROC) curve was used to evaluate the diagnostic value of multimodal MRI parameters and MVD for high-grade glioma and their predictive value for poor prognosis. The Hosmer-Lemeshow test was used to assess the goodness of fit of the model. Results:The ADC value in the low-grade group was higher than in the high-grade group, while CBF, Cho/Cr, and MVD were lower in the low-grade group (P < 0.05). The ADC value in the good prognosis group was higher than in the poor prognosis group, while CBF and Cho/Cr values were lower in the poor prognosis group (P < 0.05). ADC was negatively correlated with MVD in glioma patients (r = －0.226, P = 0.001); CBF and Cho/Cr were positively correlated with MVD (r = 0.235 and 0.396, P = 0.001 and < 0.001, respectively). The AUCs of ADC, CBF, Cho/Cr, and MVD for diagnosing high-grade glioma were 0.870, 0.696, 0.926, and 0.950, respectively. The combined AUC of these four parameters was 0.997. For prognosis prediction based on multimodal MRI parameters, the combined AUC of ADC+CBF+Cho/Cr was 0.876, while the AUCs of ADC, CBF, and Cho/Cr alone were 0.718, 0.688, and 0.808, respectively. The Hosmer-Lemeshow test showed that the combined model (ADC + CBF + Cho/Cr) fit was good (χ 2 = 8.570, P = 0.380), indicating that the combined prediction model had high accuracy. Conclusion:ADC, CBF, and Cho/Cr have a weak correlation with MVD in glioma patients. The combination of these four parameters has high value in diagnosing high-grade glioma, and the combination of the first three parameters has higher predictive value for patient prognosis.

Background:Stress-induced hyperglycemia (SIH) has been associated with poor outcomes in stroke patients. However, the relationship between SIH and sepsis in this population remains understudied. We aimed to evaluate the association of SIH, measured using the stress hyperglycemia ratio (SHR), with the development of sepsis and mortality among critically ill stroke patients. Methods:We retrospectively analyzed stroke patients requiring ICU admission from the MIMIC-IV database. Primary outcome was sepsis, and secondary outcomes were 30-day and 90-day all-cause mortality. Multivariable Cox and logistic regression models were used to evaluate associations. Results:A total of 3018 patients were included (66.8% ischemic stroke). After full adjustment for confounders, SHR was independently associated with an increased risk of sepsis (Q4 vs Q1: OR 1.46, 95% CI: 1.12-1.89, P = 0.005; continuous SHR: OR 1.31, P = 0.010). SHR also demonstrated a strong dose-response relationship with mortality; patients in Q4 had significantly higher risks of 30-day (OR 2.95, 95% CI: 2.25-3.88, P < 0.001) and 90-day mortality (OR 2.25, 95% CI: 1.80-2.82, P < 0.001). Subgroup analyses revealed significant interactions between SHR and stroke type for sepsis (P for interaction = 0.014), with a more pronounced effect observed in ischemic stroke patients. The associations between SHR and both sepsis and mortality were consistently maintained regardless of the presence of diabetes (all P < 0.050). Conclusion:Elevated stress hyperglycemia ratio is independently associated with higher risks of sepsis and short-to long-term mortality among critically ill patients with stroke, with consistent associations observed irrespective of diabetes status. In contrast, no statistically significant association between SHR and sepsis was identified in the hemorrhagic stroke subgroup.

RATIONALE AND OBJECTIVES:This study aimed to develop and validate an interpretable prediction model integrating CT-derived imaging biomarkers and clinical laboratory parameters to predict prolonged length of stay (LOS) in ulcerative colitis (UC). MATERIALS AND METHODS:In this multi-center retrospective study, 273 UC patients were enrolled from three hospitals. Patients were divided into training (n = 184) and external validation (n = 89) cohorts. CT visual features and body composition parameters were evaluated. Univariate and multivariable logistic regression were used to select predictive features. The imaging, clinical, and combined model were developed and evaluated. The feature contribution was qualified using the Shapley additive explanation (SHAP) algorithm. RESULTS:The combined model with six key predictors including visceral adipose tissue (VAT) density, visceral obesity, bowel wall thickening, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR) and albumin demonstrated superior performance with area under the curve (AUCs) of 0.862 and 0.867 in the training and validation cohorts for predicting prolonged LOS(LOS≥14 days), respectively. It significantly outperformed both the imaging-only and clinical-only models (all p < 0.05). SHAP analysis confirmed VAT density as the most influential feature. With the addition of the clinical biomarkers to the CT-derived prediction model, the prediction performance significantly improved with integrated discrimination improvement (IDI=0.187, 0.189) and net reclassification improvement (NRI=0.193, 0.268) in the training and validation cohort, respectively (all p < 0.05). CONCLUSION:The combined prediction model is promising for predicting prolonged LOS in UC. Moreover, VAT density was established as an independent predictor of LOS and was identified by SHAP analysis as the most influential feature in the predictive model.

OBJECTIVES:To evaluate the ability of the maximum standardized uptake value (SUVmax) to predict the lymphovascular space invasion (LVSI) status in endometrial cancer (EC). METHOD:PubMed/MEDLINE, Web of Science, Embase, and the Cochrane Library were systematically searched for all original studies evaluating the diagnostic efficacy of LVSI using PET/CT or PET/MR. Methodological quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). A bivariate random effects model was used to acquire pooled sensitivity, specificity, heterogeneity, and the area under the summary receiver operating characteristic curve (AUROC). Meta-regression and sensitivity analysis were performed to identify sources of heterogeneity. RESULTS:A total of 6 studies (257 patients) were included. Most studies had a low risk of bias, and all studies had minimal applicability concerns. The summary AUROC values, pooled sensitivity and specificity of SUVmax in detecting LVSI in EC were 0.77, 62% and 83%, respectively. One study may have contributed to the unstable results of this study according to the sensitivity analysis. CONCLUSION:Our study showed that SUVmax has moderate accuracy in noninvasively predicting LVSI in EC. More original studies with large samples are needed in the future to evaluate the role of SUVmax in differentiating LVSI. Advances in knowledge: LVSI is closely related to the prognosis of EC, and it can only be obtained by surgical pathology. SUVmax has moderate diagnostic performance in preoperatively predicting LVSI in EC. Future studies with large samples are needed to confirm the clinical value of SUVmax in the preoperative prediction of LVSI.

OBJECTIVE:Accurate evaluation of inflammation severity in ulcerative colitis (UC) can guide treatment strategy selection. The potential value of the pericolic fat attenuation index (FAI) on CT as an indicator of disease severity remains unknown. This study aimed to assess the diagnostic accuracy of pericolic FAI in predicting UC severity. MATERIALS AND METHODS:This retrospective study enrolled 148 patients (mean age 48 years; 87 males). The fat attenuation on CT was measured in four different locations: the mesocolic vascular side (MS) and opposite side of MS (OMS) around the most severe bowel lesion, the retroperitoneal space (RS), and the subcutaneous area. The fat attenuation indices (FAIMS, FAIOMS, and FAIRS) were calculated as the fat attenuation measured in MS, OMS, and RS, respectively, minus that of the subcutaneous area, and were obtained in the non-enhanced, arterial, and delayed phases. Correlations between the FAI and UC Endoscopic Index of Severity (UCEIS) were assessed using Spearman's correlation. Predictors of severe UC (UCEIS ≥7) were selected by univariable analysis. The performance of FAI in predicting severe UC was evaluated using the area under the receiver operating characteristic curve (AUC). RESULTS:The FAIMS and FAIOMS scores were significantly higher than FAIRS in three phases (all P < 0.001). The FAIMS and FAIOMS scores moderately correlated with the UCEIS score (r = 0.474-0.649 among the three phases). Additionally, FAIMS and FAIOMS identified severe UC, with AUC varying from 0.77 to 0.85. CONCLUSION:Increased CT attenuation of pericolic adipose tissue could serve as a noninvasive marker for evaluating UC severity. FAIMS and FAIOMS of three phases showed similar prediction accuracies for severe UC identification.

BACKGROUND:Systolic blood pressure (BP) is a key factor in the outcomes of patients with acute ischemic stroke (AIS) receiving endovascular thrombectomy (EVT). However, the factors that mediate the association between BP and clinical outcome are unclear. METHODS:Consecutive patients with AIS in the anterior circulation underwent continuous BP monitoring for 24 hours. The 3-month modified Rankin scale (mRS) score was defined as the clinical functional outcome. The systolic BPI indices (BPIs) were successive variation, standard deviation, variability independent of mean BP (VIM), and 24-hour mean BP. Regression analysis was used to assess the correlation between different BPIs and functional outcomes, whereas mediation analysis was employed to assess the potential mediating effects of baseline risk factors through BP on functional outcomes. RESULTS:A total of 140 of 292 patients (47.9%) achieved functional independence, and 87 (29.8%) experienced hemorrhagic transformation (HT). A history of stroke or hypertension and NIHSS score at onset were associated with SD and VIM (P < 0.05). BP variation (BPV) was still strongly associated with functional outcomes after adjustment for different risk factors. Mediation analysis revealed that stroke affected functional outcomes by affecting BPV, while the hypertension history affected functional prognosis by impacting the 24-hour mean BP and BPV. In addition, higher National Institute of Health stroke scale (NIHSS) scores were associated with increased BPV, whereas increased BPV was correlated with a greater proportion of unfavorable outcomes. CONCLUSIONS:To our knowledge, this study is the first to explore the mediating effects of different BPIs on the relationships between risk factors and functional outcomes and may provide new insights and potential mechanisms for improving AIS prognosis.

OBJECTIVES:Chronic internal carotid artery occlusion (CICAO) poses a considerable risk for stroke. While endovascular revascularization holds promise as a potential therapy, its real-world efficacy, safety, and long-term outcomes remain underexplored. This study aims to assess the effectiveness, safety, and long-term outcomes of endovascular revascularization in symptomatic CICAO patients refractory to medical therapy. MATERIALS AND METHODS:A retrospective cohort study was conducted to collect clinical and surgical data from CICAO patients meeting the inclusion criteria for endovascular revascularization therapy. Patients were categorized into groups based on the success or failure of revascularization procedures. Follow-up assessments were undertaken to ascertain patients' prognoses and survival outcomes. Logistic multivariate analysis was employed to identify risk factors associated with primary and secondary outcome events. COX proportional hazard regression was used to compare the risk ratios of these events between the two groups. RESULTS:The study included 59 patients undergoing 62 procedures with a 75.81 % success rate for revascularization. Perioperative complications were 6.45 %, and the average follow-up duration was 36.53 ± 3.92 months. In the successful revascularization group, the primary endpoint event rate was 6.52 %, contrasting with 23.08 % in the non-revascularization group. Carotid artery occlusion and diabetes emerged as independent risk factors for primary endpoint events. A significant difference was observed between the two groups in both primary endpoint (RR 0.16, [95 %CI, 0.03-0.84]) and total endpoint event rates (RR 0.27, [95 %CI, 0.08-0.96]) CONCLUSIONS: Failure of revascularization may be associated with an increased risk of recurrent cerebrovascular events in patients with CICAO, while successful endovascular revascularization appears to be linked to a lower incidence of such events. However, these results should be interpreted with caution due the relatively small sample size.

OBJECTIVE:Preoperative differentiation of benign and malignant endometrial lesions, along with the identification of aggressive histological types of endometrial cancer (EC), is crucial for guiding treatment strategies. Time-dependent diffusion magnetic resonance imaging (TDD-MRI), which allows the characterization of tissue microstructure at the cellular level, is not currently applied for endometrial lesions. This study aimed to evaluate TDD-MRI-derived microstructural parameters for noninvasively distinguishing benign and malignant endometrial lesions and predicting aggressive histological types of EC. MATERIALS AND METHODS:This prospective study enrolled 177 patients with clinically suspected EC who underwent TDD-MRI between January 2024 and March 2025. The Imaging Microstructural Parameters Using Limited Spectrally Edited Diffusion method was used to extract microstructural parameters, including the cell diameter (d), intracellular volume fraction (vin), cellularity (number of cells per unit area), cellularity index (vin/d), and extracellular diffusivity (Dex), along with three apparent diffusion coefficient measurements. The area under the receiver operating characteristic curve (AUC) was used to assess diagnostic performance. The Pearson correlation coefficient between the microstructural parameters and histopathological measurements was calculated. RESULTS:A total of 130 women (mean ± standard deviation age: 56 ± 14 years) administered uterine curettage or surgery were included in the final analysis. All microstructural parameters showed significant differences between benign endometrial lesions and EC (P < 0.05), as well as between nonaggressive and aggressive EC (P < 0.05). Cellularity exhibited the highest AUC of 0.86 for distinguishing benign endometrial lesions from EC, whereas the cellularity index showed the highest AUC of 0.88 for distinguishing aggressive histological types. D0Hz was positively correlated with Dex (P < 0.05) and negatively correlated with diameter (P < 0.05), cellularity index (P < 0.01) and vin (P < 0.001) in patients with benign endometrial lesions. D0Hz was positively correlated with Dex (P < 0.001) and negatively correlated with vin (P < 0.001) in patients with EC. Microstructural parameters strongly correlated with corresponding pathological features (r = 0.77-0.83; P < 0.001). CONCLUSION:TDD-MRI-derived microstructural parameters demonstrated high performance in differentiating benign from malignant endometrial diseases and identifying aggressive types of EC.

Peri-bowel fat inflammation is a prominent feature of inflammatory bowel disease (IBD). The peri-bowel fat attenuation index (FAI) can capture fat inflammation on abdominal CT. This study aimed to investigate the prognostic value of the peri-bowel FAI in IBD patients. Totally, 207 IBD patients were retrospectively collected. Regions of interest were placed at 5 different locations, namely, mesenteric side (MS) and opposite side of MS (OMS) around the most severe bowel lesion, spaces around the normal bowel wall (Nor), retroperitoneal space (RS), and subcutaneous area. The Kaplan–Meier curves were plotted. The prognostic value of the peri-bowel FAI was evaluated by multivariable Cox regression models. High peri-bowel FAI values of MS and OMS were predictors of disease progression and correlated strongly with each other (r = 0.840, p < 0.001), while the FAI of Nor and RS were not. Therefore, peri-bowel FAI of MS was used as a representative biomarker for the prediction of IBD disease progression (HR = 1.161 [1.110–1.215], p < 0.001) with an optimum cutoff of 25.1 HU, which was confirmed in the subgroup analysis with different disease subtypes. With the addition of the peri-bowel FAI to the current noninvasive risk prediction model, the AUC increased from 0.706 (0.638–0.767) to 0.864 (0.810–0.90) with integrated discrimination improvement (IDI = 0.293 [0.229–0.356], p < 0.001) and net reclassification improvement (NRI = 1.053 [0.821–1.284], p < 0.001). The peri-bowel FAI is promising for IBD disease progression prediction and risk stratification by quantifying peri-bowel fat inflammation. High peri-bowel FAI values are an independent indicator of increased IBD disease progression and could guide early targeted prevention and intensive therapy. Questions The peri-bowel fat attenuation index (FAI) helps detect peri-bowel fat inflammation noninvasively, but its importance for risk stratification and prediction of clinical outcomes remains unknown. Findings The peri-bowel FAI was an independent predictor of inflammatory bowel disease (IBD) disease progression with an optimum cutoff of 25.1 HU. Clinical relevance The peri-bowel FAI is a promising biomarker for contributing to the identification of so-called high-risk patients with uncontrolled inflammation, who might be candidates for more intensive treatment for addressing underlying inflammation at early stages and ultimately improve long-term prognosis.

OBJECTIVE:To investigate the value of conventional MRI sequences in predicting the occurrence of postoperative pancreatic fistula (POPF) in patients undergoing pancreaticoduodenectomy (PD). METHODS:A total of 122 patients from August 2019 to April 2023 were enrolled. All patients underwent pancreatic histological evaluation, including fibrosis, fat deposition, and acinar cell atrophy. The preoperative image features of pancreas were obtained, including morphological features, pancreas-muscle signal intensity ratio, pancreatic fat fraction and multi-phase enhancement features. The patients were divided into two groups according to whether pancreatic fistula occurred after operation. The related risk factors of pancreatic fistula, the correlation between imaging and pathological changes were analyzed, and the value of preoperative imaging in predicting pancreatic fistula was evaluated. RESULTS:Of the 122 patients, 23(18.9 %) developed POPF. Pathological score showed that there was a significant difference in pancreatic fat deposition between the two groups (P = 0.006), the fat deposition score was higher in the POPF group. Pancreatic fat deposition was the only independent risk factor for POPF(OR,1.933; P = 0.018). MRI showed that proton density fat fraction(PDFF) (P = 0.001), pancreas-to-aorta signal intensity ratio(P-A SI ratio) of equilibrium phase(P = 0.023) and delay phase(P = 0.020) had significant differences. PDFF was positively correlated with fat deposition(r = 0.404, P < 0.001), P-A SI ratio of equilibrium phase and delay phase were positively correlated with fibrosis(r = 0.313, P = 0.002; r = 0.315, P = 0.002, respectively). ROC analysis showed that PDFF had the best efficacy in predicting postoperative pancreatic fistula (AUROC = 0.810), better than P-A SI ratio of equilibrium phase(AUROC = 0.752) and delayed phase(AUROC = 0.766). CONCLUSIONS:Pancreatic fat deposition is a high risk factor for POPF, PDFF can reflect fat deposition and predict POPF.

Background: Increasing evidence has documented cortical involvement at all stages of PD. The local vulnerabilities within certain brain regions in PD have been previously demonstrated, whereas its underlying genetic and neurochemical factors remain unclear. This study aims to investigate the spatial spectrum of cortical atrophy in Parkinson’s disease (PD) and link these variances in gray matter properties and curvature respectively to putative molecular pathways and neurotransmitter factors. Methods: We recruited 141 clinically diagnosed PD patients and 70 healthy controls. Cortical morphological abnormalities of PD were obtained by intergroup comparisons in gray matter properties metrics and curvature measurements. Then we performed gene-category enrichment and spatial correlation analyses to evaluate the specific correspondence between cortical alteration in PD and genetic expression from the Allen Human Brain Atlas and normative neurotransmitter atlases from Neuromaps. Results: We found decreased gray matter properties in temporal, somatomotor, cingulate and occipital cortices, decreased curvature measures in occipital, temporal and orbitofrontal cortices, and increased curvature measures in somatomotor, prefrontal and posterior parietal cortices for PD patients. The related genes were enriched for the glucose metabolism, mitochondrial function, and post-translational histone modifications processes. In addition, the serotonin and norepinephrine transporter devoted more to gray matter properties alterations while the dopamine, gamma-aminobutyric acid receptors, and norepinephrine transporter were strong contributors of curvature abnormalities in PD. Conclusions: Collectively, the present study offered interpretation of cortical morphological alterations and the cortical pathogenic theory in PD from genetic and neurochemical perspectives, which inspire further research on new pharmacotherapeutic approaches.

Background: Substantial evidence emphasizes the dysregulation of iron homeostasis, demyelination and oxidative stress in the neurodegenerative process of multiple system atrophy (MSA) and Parkinson’s disease (PD), although its clinical implications remain unclear. Recent MRI post-processing techniques leveraging magnetic susceptibility properties provide a noninvasive means to characterize iron, myelin content and oxygen metabolism alterations. This study aims to investigate subcortical alterations of susceptibility-derived metrics in these two synucleinopathies. Methods: A cohort comprising 180 patients (122 with PD and 58 with MSA) and 77 healthy controls (HCs) underwent clinical evaluation and multi-echo gradient echo MRI scans. Susceptibility source separation, susceptibility-based oxygen extraction fraction (OEF) mapping and semiautomatic subcortical nuclei segmentation were utilized to derive parametric values of deep gray matter in all subjects. Results: MSA patients showed markedly elevated paramagnetic susceptibility values in the putamen, globus pallidus (GP) and thalamus; increased diamagnetic susceptibility values in the putamen and dentate nucleus; and reduced OEF values across all nuclei compared with PD patients and HCs. Whereas PD exhibited increased positive susceptibility values in the substantia nigra and enhancing negative values in the GP, similar to MSA. Notably, age-related reductions in OEF were evident in HCs, which was altered by the MSA pathology. Paramagnetic susceptibility was correlated with disease severity. Moreover, the susceptibility-derived metrics of striatum and midbrain nuclei proved to be effective predictors to distinguish PD from MSA (AUC = 0.833). Conclusion: Susceptibility-derived metrics could detect pathological involvement distinct to each disease, offering significant potential for differentiating between MSA and PD in clinical settings.

OBJECTIVES:Iron deposition and mitochondrial dysfunction are closely associated with the genesis and progression of Parkinson's disease (PD). This study aims to extract susceptibility and oxygen extraction fraction (OEF) values of deep grey matter (DGM) to explore spatiotemporal progression patterns of brain iron-oxygen metabolism in PD.METHODS:Ninety-five PD patients and forty healthy controls (HCs) were included. Quantitative susceptibility mapping (QSM) and OEF maps were computed from MRI multi-echo gradient echo data. Analysis of covariance (ANCOVA) was used to compare mean susceptibility and OEF values in DGM between early-stage PD (ESP), advanced-stage PD (ASP) patients and HCs. Then Granger causality analysis on the pseudo-time-series of MRI data was applied to assess the causal effect of early altered nuclei on iron content and oxygen extraction in other DGM nuclei.RESULTS:The susceptibility values in substantia nigra (SN), red nucleus, and globus pallidus (GP) significantly increased in PD patients compared with HCs, while the iron content in GP did not elevate obviously until the late stage. The mean OEF values for the caudate nucleus, putamen, and dentate nucleus were higher in ESP patients than in ASP patients or/and HCs. We also found that iron accumulation progressively expands from the midbrain to the striatum. These alterations were correlated with clinical features and improved AUC for early PD diagnosis to 0.824.CONCLUSIONS:Abnormal cerebral iron deposition and tissue oxygen utilization in PD measured by QSM and OEF maps could reflect pathological alterations in neurodegenerative processes and provide valuable indicators for disease identification and management.CLINICAL RELEVANCE STATEMENT:Noninvasive assessment of cerebral iron-oxygen metabolism may serve as clinical evidence of pathological changes in PD and improve the validity of diagnosis and disease monitoring.KEY POINTS:• Quantitative susceptibility mapping and oxygen extraction fraction maps indicated the cerebral pathology of abnormal iron accumulation and oxygen metabolism in Parkinson's disease. • Iron deposition is mainly in the midbrain, while altered oxygen metabolism is concentrated in the striatum and cerebellum. • The susceptibility and oxygen extraction fraction values in subcortical nuclei were associated with clinical severity.

Explore the correlation between basal ganglia nuclei and cortical gray matter volume changes in tremor-dominant and postural instability-gait difficulty (PIGD) Parkinson’s disease subtypes for Parkinson’s disease diagnosis and individualized treatment. High-resolution 3D-T1WI MRI data from 35 tremor-dominant and 30 PIGD patientsand 35 healthy controls were analyzed. Voxel-based morphometry identified gray matter volume differences. Automated basal ganglia segmentation quantified subcortical volumes, followed by multivariate analysis of covariance and Spearman correlation analyses. Compared with healthy control, patients with PIGD exhibited severe gray matter loss ( P < 0.0001), while tremor-dominant showed nonsignificant reductions. Subcortically, different basal ganglia volumes were atrophied in the tremor-dominant and PIGD groups compared with the healthy control ( P < 0.05). PIGD demonstrated greater left putamen atrophy than tremor-dominant ( P < 0.05). Spearman correlation analysis revealed that the volume of the right globus pallidus was positively correlated with that of the left medial and lateral cingulate gyrus in patients with tremor-dominant ( r = 0.35, P = 0.04); and between the left globus pallidus volume and the right superior temporal gyrus volume in patients with PIGD ( r = 0.47, P = 0.01). Compared with the tremor-dominant subtype, the PIGD subtype exhibits more severe GM atrophy, with different basal ganglia volume changes across subtypes. These altered anatomical features and the correlation between degeneration of the basal ganglia region and cortical gray matter changes may provide insights into the differential functional changes in patients with different motor subtypes and help to elucidate the underlying pathologic mechanisms.

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. 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. The random forest model had the best performance in the test set, with the highest area under the curve (AUC, 0.840; 95

BackgroundBlood pressure (BP) is a key factor for the clinical outcomes of acute ischemic stroke (AIS) receiving endovascular thrombectomy (EVT). However, the effect of the circadian pattern of BP on functional outcome is unclear.MethodsThis multicenter, retrospective, observational study was conducted from 2016 to 2023 at three hospitals in China (ChiCTR2300077202). A total of 407 patients who underwent endovascular thrombectomy (EVT) and continuous 24-h BP monitoring were included. Two hundred forty-one cases from Beijing Hospital were allocated to the development group, while 166 cases from Peking University Shenzhen Hospital and Hainan General Hospital were used for external validation. Postoperative systolic BP (SBP) included daytime SBP, nighttime SBP, and 24-h average SBP. Least absolute shrinkage and selection operator (LASSO), support vector machine-recursive feature elimination (SVM-RFE), Boruta were used to screen for potential features associated with functional dependence defined as 3-month modified Rankin scale (mRS) score ≥ 3. Nine algorithms were applied for model construction and evaluated using area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy.ResultsThree hundred twenty-eight of 407 (80.6%) patients achieved successful recanalization and 182 patients (44.7%) were functional independent. NIHSS at onset, modified cerebral infarction thrombolysis grade, atrial fibrillation, coronary atherosclerotic heart disease, hypertension were identified as prognostic factors by the intersection of three algorithms to construct the baseline model. Compared to daytime SBP and 24-h SBP models, the AUC of baseline + nighttime SBP showed the highest AUC in all algorithms. The XGboost model performed the best among all the algorithms. ROC results showed an AUC of 0.841 in the development set and an AUC of 0.752 in the validation set for the baseline plus nighttime SBP model, with a brier score of 0.198.ConclusionThis study firstly explored the association between circadian BP patterns with functional outcome for AIS. Nighttime SBP may provide more clinical information regarding the prognosis of patients with AIS after EVT.

Background:Paramagnetic rim lesions (PRLs) on susceptibility magnetic resonance sequences have been suggested as an imaging marker of disease progression in multiple sclerosis. This retrospective cross-sectional study aimed to investigate the impact of PRLs on cortical thickness and gray matter (GM) to white matter (WM) contrast in relapsing-remitting multiple sclerosis (RRMS).Methods:A total of 82 RRMS patients (40 patients with at least 1 PRL and 42 patients without PRL) and 43 healthy controls (HC) were included in this study. The T1-weighted images (T1WI) were processed with the FreeSurfer pipeline. GM to WM signal intensity ratio (GWR) was obtained from T1WI by dividing the GM signal intensity by the WM signal intensity for each vertex. Group differences in cortical thickness and GWR were tested on reconstructed cortical surface.Results:Compared to HC, patients with PRL had thinner mean cortical thickness (P<0.001), higher mean GWR (P=0.001), and lower brain structure volumes (cortex volume, P=0.001; WM volume, P<0.001; deep GM volume, P<0.001). Vertex-based analysis found significant cortical thinning in several regions and increased GWR in a wider range of regions in patients with PRL. The two types of clusters had both overlapping regions and independent regions. However, in patients without PRL, only a few regions showed significant cortical thickness changes. Correlation analysis found that in patients with PRL, only PRL volume showed a significant negative correlation with mean cortical thickness (P=0.048), and PRL volume and count, non-PRL count, and total lesion volume were significantly and positively correlated with mean GWR (P<0.05).Conclusions:There were significant changes in cortical thickness, GWR, and brain structure volume in RRMS patients with PRL that may contribute to further understanding of the pathological mechanisms underlying neurological tissue damage.

Background: Multiple system atrophy (MSA) and Parkinson's disease (PD) are neurodegenerative disorders characterized by α-synuclein pathology, disrupted iron homeostasis and impaired neurochemical transmission. Considering the critical role of iron in neurotransmitter synthesis and transport, our study aims to identify distinct patterns of whole-brain iron accumulation in MSA and PD, and to elucidate the corresponding neurochemical substrates. Methods: A total of 122 PD patients, 58 MSA patients and 78 age-, sex-matched health controls underwent multi-echo gradient echo sequences and neurological evaluations. We conducted voxel-wise and regional analyses using quantitative susceptibility mapping to explore MSA or PD-specific alterations in cortical and subcortical iron concentrations. Spatial correlation approaches were employed to examine the topographical alignment of cortical iron accumulation patterns with normative atlases of neurotransmitter receptor and transporter densities. Furthermore, we assessed the associations between the colocalization strength of neurochemical systems and disease severity. Results: MSA patients exhibited increased susceptibility in the striatal, midbrain, cerebellar nuclei, as well as the frontal, temporal, occipital lobes, and anterior cingulate gyrus. In contrast, PD patients displayed elevated iron levels in the left inferior occipital gyrus, precentral gyrus, and substantia nigra. The excessive iron accumulation in MSA or PD correlated with the spatial distribution of cholinergic, noradrenaline, glutamate, serotonin, cannabinoids, and opioid neurotransmitters, and the degree of this alignment was related to motor deficits. Conclusions: Our findings provide evidence of the interaction between iron accumulation and non-dopamine neurotransmitters in the pathogenesis of MSA and PD, which inspires research on potential targets for pharmacotherapy.