Search

Abstract Background Femoral head collapse is a critical pathological change and is regarded as turning point in disease progression in osteonecrosis of the femoral head (ONFH). In this study, we aim to build an automatic femoral head collapse prediction pipeline for ONFH based on magnetic resonance imaging (MRI) radiomics. Methods In the segmentation model development dataset, T1-weighted MRI of 222 hips from two hospitals were retrospectively collected and randomly split into training (n = 190) and test (n = 32) sets. In the prognosis prediction model development dataset, 206 hips were also retrospectively collected from two hospitals and divided into training set (n = 155) and external test set (n = 51) according to data source. A deep learning model for automatic lesion segmentation was trained with nnU-Net, from which three-dimensional regions of interest were segmented and a total of 107 radiomics features were extracted. After intra-class correlation coefficients screening, feature correlation coefficient screening and Least Absolute Shrinkage and Selection Operator regression feature selection, a machine learning model for ONFH prognosis prediction was trained with Logistic Regression (LR) and Light Gradient Boosting Machine (LightGBM) algorithm. Results The segmentation model achieved an average dice similarity coefficient of 0.848 and an average 95% Hausdorff distance of 3.794 in the test set, compared to the manual segmentation results. After feature selection, nine radiomics features were included in the prognosis prediction model. External test showed that the LightGBM model exhibited acceptable predictive performance. The area under the curve (AUC) of the prediction model was 0.851 (95% CI: 0.7268–0.9752), with an accuracy of 0.765, sensitivity of 0.833, and specificity of 0.727. Decision curve analysis showed that the LightGBM model exhibited favorable clinical utility. Conclusion This study presents an automated pipeline for predicting femoral head collapse in ONFH with acceptable performance. Further research is necessary to determine the clinical applicability of this radiomics-based approach and to assess its potential to assist in treatment decision-making for ONFH.

Background: Enlarged perivascular spaces (EPVSs) are commonly detected via magnetic resonance imaging. It is unclear whether EPVSs are associated with cognitive impairment within one month after an acute ischemic stroke (AIS) (i.e., early AIS with cognitive impairment (EAIS-CI)). This study explored the severity and location of EPVSs and their association with EAIS-CI severity and provides clinicians with early warning indicators before the onset of typical clinical symptoms in the Chinese population. Methods: The clinical data of 208 patients (176 AIS patients and 32 controls) were prospectively analyzed using the Montreal Cognitive Assessment Beijing version (MoCA-BJ) score as the primary group criterion and the Mini-Mental State Examination (MMSE) score as a supplementary criterion. When EPVS I as the main EPVS type detected by imaging, the basal ganglia (BG) is the area most severely affected. Statistical analysis was conducted on the relevant clinical data. Results: AIS patients were grouped based on MoCA-BJ scores. Age (p < 0.01), education level (p = 0.02), EPVS I as the main EPVS type (p < 0.01), the number of right-sided BG-EPVSs (p = 0.04), white matter hyperintensities (WMHs) (Fazekas scores: p = 0.02), brain atrophy (global cortical atrophy scores: p < 0.01, Koedam posterior atrophy visual scale scores: p = 0.01, medial temporal lobe atrophy scores: p < 0.01) and AIS lesion volume (p = 0.01) were significantly greater in the EAIS-CI group than in the EAIS without cognitive impairment group. The cognitive domains of attention (p = 0.04) and orientation (p < 0.01) were more closely associated with EPVS I as the main EPVS type. However, multivariate regression analysis did not identify EPVS I as the main EPVS type as the main risk factor for EAIS-CI (p = 0.098). Grouping by MMSE scores revealed that EPVS I as the main EPVS type was linked to low education level (p < 0.01) and was significantly associated with EAIS in individuals with cognitive dementia (p < 0.01). Conclusions: As a result of multiple factors, EAIS-CI is significantly associated with a low education level, BG-EPVS, WMHs, and worsening brain atrophy severity. Imaging markers, such as the severity of BG-EPVS, can assist in the early diagnosis and assessment of EAIS-CI. Clinical Trial Registration: The study was registered with the China Clinical Trial Registry (https://www.chictr.org.cn/), registration number: ChiCTR2000038819.

Abstract Introduction The widespread use of brain magnetic resonance imaging (MRI) has revealed the correlation between enlarged perivascular spaces (EPVS) and cognitive impairment (CI). However, few studies have examined the correlation between MRI-visible EPVS and CI in patients with Parkinson’s disease (PD) and vascular parkinsonism (VaP). This study explored how the number and main location of EPVS in PD and VaP are correlated with the occurrence of CI in these diseases to provide radiology markers and other evidence for early clinical diagnosis in a Chinese cohort. Methods Clinical data were prospectively collected from 77 patients: 26 patients clinically diagnosed with PD or probable PD, 19 patients clinically diagnosed with VaP, and 32 control subjects with normal cognitive function and no stroke or parkinsonism. The patients with PD and VaP were divided into a CI group and a no CI (NCI) group according to the Montreal Cognitive Assessment Beijing version (MoCA-BJ). The relevant clinical data were statistically analysed. Results The centrum semiovale (CSO)-EPVS, lacunes, Fazekas scores, global cortical atrophy scale (GCA) scores, Koedam posterior atrophy visual scale (KS) scores, and medial temporal atrophy (MTA) scores were higher in the PD-CI and VaP-CI groups than in the control group (adjusted P 0.05). Conclusion VaP-CI results from multiple factors and is significantly associated with BG-EPVS, lacunes, white matter hyperintensities and brain atrophy. BG-EPVS can be used as an imaging marker to distinguish VaP-CI from PD-CI.

Background: The immunoglobulin G synthesis rate (IgG SR) and immunoglobulin G (IgG) index are indicators of abnormal intrathecal humoural immune responses, and the albumin quotient (QALB) is an indicator used to evaluate the completeness of the blood-cerebrospinal fluid barrier (BCB). No systematic reports regarding differences in Guillain-Barré syndrome (GBS) and chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) are available. We assessed differences in the IgG SR, IgG index and QALB between GBS and CIDP patients in a Chinese cohort.Methods: A total of 234 patients were retrospectively enrolled in this study, and 167 clinically confirmed GBS and CIDP patients were selected. Meanwhile, 67 non-GBS and non-CIDP patients requiring cerebrospinal fluid (CSF) examination were enrolled as the control group. The IgG SR, IgG index and QALB were calculated using formulas. The relevant clinical data were subjected to statistical analysis.Results: Among the GBS and CIDP study groups and the control group, the QALB had the highest positive rate (80.00%) in the CIDP group (P < 0.01). The QALB stratification analysis showed that the ranges of 10 < QALB ≤ 30 were dominant in the GBS and CIDP groups, and the positive rate of CIDP was higher than that of GBS. Furthermore, a QALB ≤ 7 was dominant in the control group, and a QALB > 30 was dominant in the CIDP group. In receiver operating characteristic (ROC) curve analysis with the CIDP group as the trial group and the GBS group as the control group, the differences in the QALB were statistically significant (P < 0.01). To achieve a high specificity of 98~99%, the diagnostic cut-off value for the QALB was above 57.37 (sensitivity: 9.33%) or below 0.60 (sensitivity: 4.35%). Multivariate logistic regression analysis showed that the CIDP patients had a QALB higher than 57.37, and compared with that in the GBS patients, the difference in the QALB was statistically significant (P < 0.01).Conclusion: QALB elevation was associated with CIDP, while QALB values above 57.37 or below 0.60 had high specificity in differentiating between GBS and CIDP. In CIDP, the BCB is generally moderately to severely damaged; in GBS, the BCB is generally moderately damaged.