Development and validation of a deep learning model for prediction of intracranial aneurysm rupture risk based on multi-omics factor.

OBJECTIVE: The clinical ability of radiomics to predict intracranial aneurysm rupture risk remains unexplored. This study aims to investigate the potential uses of radiomics and explore whether deep learning (DL) algorithms outperform traditional statistical methods in predicting aneurysm rupture risk. METHODS: This retrospective study included 1740 patients with 1809 intracranial aneurysms confirmed by digital subtraction angiography at two hospitals in China from January 2014 to December 2018. We randomly divided the dataset (hospital 1) into training (80%) and internal validation (20%). External validation was performed using independent data collected from hospital 2. The prediction models were developed based on clinical, aneurysm morphological, and radiomics parameters by logistic regression (LR). Additionally, the DL model for predicting aneurysm rupture risk using integration parameters was developed and compared with other models. RESULTS: The AUCs of LR models A (clinical), B (morphological), and C (radiomics) were 0.678, 0.708, and 0.738, respectively (all p < 0.05). The AUCs of the combined feature models D (clinical and morphological), E (clinical and radiomics), and F (clinical, morphological, and radiomics) were 0.771, 0.839, and 0.849, respectively. The DL model (AUC = 0.929) outperformed the machine learning (ML) (AUC = 0.878) and the LR models (AUC = 0.849). Also, the DL model has shown good performance in the external validation datasets (AUC: 0.876 vs 0.842 vs 0.823, respectively). CONCLUSION: Radiomics signatures play an important role in predicting aneurysm rupture risk. DL methods outperformed conventional statistical methods in prediction models for the rupture risk of unruptured intracranial aneurysms, integrating clinical, aneurysm morphological, and radiomics parameters. KEY POINTS: • Radiomics parameters are associated with the rupture risk of intracranial aneurysms. • The prediction model based on integrating parameters in the deep learning model was significantly better than a conventional model. • The radiomics signature proposed in this study could guide clinicians in selecting appropriate patients for preventive treatment.

An immunogenic cell death-related regulators classification patterns and immune microenvironment infiltration characterization in intracranial aneurysm based on machine learning.

Background: Immunogenic Cell Death (ICD) is a novel way to regulate cell death and can sufficiently activate adaptive immune responses. Its role in immunity is still emerging. However, the involvement of ICD in Intracranial Aneurysms (IA) remains unclear. This study aimed to identify biomarkers associated with ICDs and determine the relationship between them and the immune microenvironment during the onset and progression of IA. Methods: The IA gene expression profiles were obtained from the Gene Expression Omnibus (GEO) database. The differentially expressed genes (DEGs) in IA were identified and the effects of the ICD on immune microenvironment signatures were studied. Techniques like Lasso, Bayes, DT, FDA, GBM, NNET, RG, SVM, LR, and multivariate analysis were used to identify the ICD gene signatures in IA. A consensus clustering algorithm was used for conducting the unsupervised cluster analysis of the ICD patterns in IA. Furthermore, enrichment analysis was carried out for investigating the various immune responses and other functional pathways. Along with functional annotation, the weighted gene co-expression network analysis (WGCNA), protein-protein interaction (PPI) network and module construction, identification of the hub gene, and co-expression analysis were also carried out. Results: The above techniques were used for establishing the ICD gene signatures of HMGB1, HMGN1, IL33, BCL2, HSPA4, PANX1, TLR9, CLEC7A, and NLRP3 that could easily distinguish IA from normal samples. The unsupervised cluster analysis helped in identifying three ICD gene patterns in different datasets. Gene enrichment analysis revealed that the IA samples showed many differences in pathways such as the cytokine-cytokine receptor interaction, regulation of actin cytoskeleton, chemokine signaling pathway, NOD-like receptor signaling pathway, viral protein interaction with the cytokines and cytokine receptors, and a few other signaling pathways compared to normal samples. In addition, the three ICD modification modes showed obvious differences in their immune microenvironment and the biological function pathways. Eight ICD-regulators were identified and showed meaningful associations with IA, suggesting they could severe as potential prognostic biomarkers. Conclusions: A new gene signature for IA based on ICD features was created. This signature shows that the ICD pattern and the immune microenvironment are closely related to IA and provide a basis for optimizing risk monitoring, clinical decision-making, and developing novel treatment strategies for patients with IA.