Development and validation of a machine learning-based prognostic risk stratification model for acute ischemic stroke

Wang, Kai; Hong, Tao; Liu, Wencai; Xu, Chan; Yin, Chengliang; Liu, Haiyan; Wei, Xiu’e; Wu, Shi-Nan; Li, Wenle; Rong, Liangqun

doi:10.1038/s41598-023-40411-2

Cited by 3 publications

(1 citation statement)

References 44 publications

(43 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SHAP has numerous medical applications in interpreting a predictive machine learning model. For example, its interpretability has been demonstrated in a prognostic risk stratification model for acute ischemic stroke [45], a feature selection task for the diagnosis of Parkinson's disease [46], and a classification task for the diagnosis of glaucoma [47].…”

Section: Discussionmentioning

confidence: 99%

Assessing Machine Learning Models for Predicting Age with Intracranial Vessel Tortuosity and Thickness Information

Yoon,

Oh,

Kim

2023

Brain Sciences

View full text Add to dashboard Cite

This study aimed to develop and validate machine learning (ML) models that predict age using intracranial vessels’ tortuosity and diameter features derived from magnetic resonance angiography (MRA) data. A total of 171 subjects’ three-dimensional (3D) time-of-flight MRA image data were considered for analysis. After annotations of two endpoints in each arterial segment, tortuosity features such as the sum of the angle metrics, triangular index, relative length, and product of the angle distance, as well as the vessels’ diameter features, were extracted and used to train and validate the ML models for age prediction. Features extracted from the right and left internal carotid arteries (ICA) and basilar arteries were considered as the inputs to train and validate six ML regression models with a four-fold cross validation. The random forest regression model resulted in the lowest root mean square error of 14.9 years and the highest average coefficient of determination of 0.186. The linear regression model showed the lowest average mean absolute percentage error (MAPE) and the highest average Pearson correlation coefficient (0.532). The mean diameter of the right ICA vessel segment was the most important feature contributing to prediction of age in two out of the four regression models considered. An ML of tortuosity descriptors and diameter features extracted from MRA data showed a modest correlation between real age and ML-predicted age. Further studies are warranted for the assessment of the model’s age predictions in patients with intracranial vessel diseases.

show abstract

Section: Discussionmentioning

confidence: 99%

Assessing Machine Learning Models for Predicting Age with Intracranial Vessel Tortuosity and Thickness Information

Yoon,

Oh,

Kim

2023

Brain Sciences

View full text Add to dashboard Cite

show abstract

The impact of dysbiosis in oropharyngeal and gut microbiota on systemic inflammatory response and short-term prognosis in acute ischemic stroke with preceding infection

He,

Li,

Zhao

et al. 2024

Front. Microbiol.

View full text Add to dashboard Cite

BackgroundStroke is a devastating disease and ranks as the second leading cause of death and disability globally. Several studies have shown that preceding infection (PI) of upper respiratory tract are strongly associated with acute ischemic stroke (AIS). However, the clinical implications and underlying pathological mechanisms remain unclear.MethodsIn this study, 16S rRNA gene sequencing was employed to compare the structural characteristics of oropharyngeal and gut microbiota in AIS patients with or without PI and normal controls (NCs; 30 cases each), and systemic inflammatory markers were detected to explore the relationship between upper respiratory tract infections (URTIs) and subsequent stroke severity and functional outcome and the potential mechanism.ResultsWe found that patients with AIS-PI exhibited elevated serum WBC, NE, CRP, and Hcy levels, as well as a higher 90-day mRS score. Oropharyngeal and gut microbiota analysis showed that AIS and AIS-PI patients exhibited increased microbial richness in sequence. Principal coordinate analysis of the microbiota demonstrated significant differences in microbiota composition among the three groups. In AIS-PI patients, Megamonas, Megasphaera, Ruminococcaceae UCG 004, Rothia, and Streptococcus were significantly enriched in the gut. Opportunistic pathogens, including Thermus, uncultured Veillonella sp., and Oribacterium sinu, were found to be significantly enriched in the oropharynx. The dysregulated microbiota were positively correlated with systemic inflammatory markers, stroke severity, and poor prognosis. In contrast, short-chain fatty acid-producing bacteria Eisenbergiella, bacterium NLAE, Fusicatenibacter, Ruminococcaceae, and Faecalibacterium were enriched in NCs. Their abundances were negatively correlated with systemic inflammatory markers, stroke severity and poor prognosis.ConclusionOur findings suggest that PIs of the upper respiratory tract may contribute to poor short-term functional outcome in AIS patients by causing disturbance of the oropharyngeal and gut microbiota and promoting elevated systemic inflammation levels.

show abstract

Lung Cancer Biomarkers Associated with Increased Peripheral Arterial Stiffness in Middle-aged Chinese Adults

Li,

Gu,

et al. 2024

JAT

View full text Add to dashboard Cite

Aims: Previous evidence suggests that serum lung cancer biomarkers are associated with inflammatory conditions; however, their relationship with peripheral arterial stiffness remains unclear. Therefore, the present study investigated the relationship between serum lung cancer biomarkers and peripheral arterial stiffness in middle-aged Chinese adults. Methods:In total, 3878 middle-aged Chinese adults were enrolled in this study. Increased peripheral arterial stiffness was assessed using the brachial-ankle pulse wave velocity and ankle-brachial index. Univariate and multivariate logistic regression analyses were used to determine the independent effects of serum lung cancer biomarkers on the risk of increased peripheral arterial stiffness. A receiver operating characteristic curve analysis was used to assess the diagnostic ability of serum lung cancer biomarkers in distinguishing increased peripheral arterial stiffness.Results: Serum levels of carcinoembryonic antigen (CEA), neuron-specific enolase (NSE), cytokeratin-19 fragment 21-1, and pro-gastrin-releasing peptide were higher in subjects with increased peripheral arterial stiffness than in those without (P＜0.05). After adjusting for other risk factors, serum CEA and NSE levels were found to be independently associated with increased peripheral arterial stiffness. The corresponding adjusted odds ratios (ORs) for increased peripheral arterial stiffness in CEA level quartiles were 1.00, 1.57, 2.15, and 6.13. The ORs for increased peripheral arterial stiffness in the quartiles of NSE levels were 1.00, 4.92, 6.65, and 8.01. Conclusions: Increased serum CEA and NSE levels are closely linked to increased peripheral arterial stiffness, and high serum CEA and NSE levels are potential risk markers for peripheral arterial stiffness in middle-aged Chinese adults.

show abstract

Development and validation of a machine learning-based prognostic risk stratification model for acute ischemic stroke

Cited by 3 publications

References 44 publications

Assessing Machine Learning Models for Predicting Age with Intracranial Vessel Tortuosity and Thickness Information

Assessing Machine Learning Models for Predicting Age with Intracranial Vessel Tortuosity and Thickness Information

The impact of dysbiosis in oropharyngeal and gut microbiota on systemic inflammatory response and short-term prognosis in acute ischemic stroke with preceding infection

Lung Cancer Biomarkers Associated with Increased Peripheral Arterial Stiffness in Middle-aged Chinese Adults

Contact Info

Product

Resources

About