Inflammatory Biomarkers and Intracranial Venous Thrombosis
Author Information
Author(s): Muacevic Alexander, Adler John R, Zhou Longmin, Jiang Wenting, Hou Pengwei, Cai Mingfa, Li Ziqi, Wang Shousen
Primary Institution: Fuzong Clinical Medical College of Fujian Medical University (The 900th Hospital), Fuzhou, CHN
Hypothesis
This study aimed to explore the role of inflammatory biomarkers in the diagnosis of cerebral venous thrombosis (CVT) and assess machine learning models for improving diagnostic accuracy.
Conclusion
Inflammatory biomarkers, particularly the systemic immune-inflammatory index (SII), have significant predictive value in diagnosing CVT, and machine learning models, especially artificial neural networks, show promise in improving diagnostic accuracy.
Supporting Evidence
- The SII and NLR levels were significantly higher in CVT patients.
- The ANN model based on SII and PLR achieved the best diagnostic performance, with an AUC of 0.94.
- Headache was the most common symptom in CVT patients, occurring in 78% of cases.
- PLR, NLR, and SII had high predictive value in the diagnosis of CVT.
Takeaway
Doctors can use certain blood tests to help find a rare condition called cerebral venous thrombosis, especially when combined with smart computer programs that learn from data.
Methodology
The study included 100 CVT patients and 50 controls, collecting demographic and clinical data, and used traditional statistical methods alongside various machine learning models to evaluate diagnostic performance.
Potential Biases
The study may introduce selection bias due to being conducted in a single center.
Limitations
The sample size was relatively small, conducted in a single center, and the cross-sectional design limits causal relationship establishment.
Participant Demographics
The study included 100 CVT patients and 50 age- and gender-matched controls.
Statistical Information
P-Value
< 0.001
Confidence Interval
95% confidence interval: 0.614-0.782
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website