3 papers indexed here
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
Stroke-associated pneumonia (SAP) is a frequent complication after acute ischemic stroke (AIS) and is associated with poor outcomes. This study aimed to develop an interpretable multimodal deep learning model integrating MRI, lesion-related brain regions, and clinical variables for early SAP prediction. A total of 426 AIS patients were retrospectively enrolled, including 71 patients with SAP. Multimodal MRI data (DWI, T1WI, and T2-FLAIR) were processed using standardized registration and lesion segmentation. A 3D convolutional neural network was used to extract imaging representations, which were fused with clinical variables and AAL3-based brain-region features. Model performance was assessed using stratified five-fold cross-validation and nested cross-validation when applicable, with further evaluation based on receiver operating characteristic (ROC) analysis, calibration analysis, and decision curve analysis. Grad-CAM was applied for model interpretation. The multimodal fusion model achieved the best performance for SAP prediction, with an AUC of 0.782 (95% CI: 0.712–0.839), compared with the clinical model based on conventional clinical variables (AUC = 0.756), the imaging model based on 3D CNN representations (AUC = 0.693), and the brain-region model based on AAL3-derived lesion location features (AUC = 0.501). The fusion model showed superior clinical utility and favorable calibration. Grad-CAM visualization demonstrated that model predictions were mainly driven by lesion-related cortical and subcortical regions. A multimodal deep learning framework integrating MRI, brain-region information, and clinical characteristics improved SAP prediction after AIS and provided an interpretable approach for individualized risk stratification.
Background Social media is a key channel for public health information, but its open nature leads to mixed quality of information regarding Ischemic Stroke (IS) treatment, which may mislead patient decisions. This study aimed to systematically compare IS treatment information published by professional and lay sources on two major Chinese social media platforms, Weibo and REDnote, in terms of content, engagement, and quality. Methods This study employed computational social science and Natural Language Processing (NLP) techniques to analyze posts from four sources (Weibo-Pro, Weibo-Lay, REDnote-Pro, REDnote-Lay). We used topic modeling and co-occurrence networks to analyze content features and developed an automated scoring system based on a Large Language Model (LLM) to quantitatively evaluate information quality on two dimensions: “linguistic features” and “evidence-based medical content”. Results Lay-sourced content exceeded professional-sourced content in both volume and user engagement. Content and quality patterns differed across the two platforms: on Weibo, the evidence-based quality of professional content was significantly higher than that of lay content (p < 0.001), whereas no significant professional–lay difference was detected on REDnote. We refer to this observed convergence as a “quality paradox,” while recognizing that the cross-sectional design does not establish a platform effect. Content themes also differed: Weibo-Lay content centered on Traditional Chinese Medicine, whereas REDnote-Lay content focused on rehabilitation experiences and family support. Conclusion Distinct platform- and source-related patterns were observed across four “discursive communities.” On REDnote, professional- and lay-sourced posts showed similar evidence-based quality in this sample. These findings support further investigation of platform-specific communication environments and may inform cautious, context-sensitive approaches for patients, clinicians, and platform managers.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
