Integrated lasso loops in aneurysm risk data

A set of indicators, natural chemicals found in the blood, known as biomarkers, can help predict when an aneurysm in the brain might rupture. The work, published in the International Journal of Data Mining and Bioinformatics, looks at the risk of rupture associated with the ballooning out of a weakened blood vessel in the brain, that can lead to catastrophic bleeding. By analysing genetic data from three groups of patients, the team has identified characteristics associated with increased instability of an aneurysm.

The researchers used genetic profiling to look at activity associated with stable and ruptured aneurysms, as well as interactions between proteins that were linked to the latter and not the former. Across all the data, they found two interactions that were active in aneurysms prone to rupture. Then, by using machine learning techniques, specifically Least Absolute Shrinkage and Selection Operator (LASSO) regression, they were able to develop a prediction curve that gives a patient rupture risk based on the presence of the biomarkers.

The findings highlight an underlying mechanism that links chronically raised elevated blood pressure, hypertension, and inflammation in the vasculature of the brain. Hypertension puts mechanical strain on the walls of the blood vessels, while, at the same, activating a hormone-driven regulator of blood pressure, known as the local renin-angiotensin system. This system triggers inflammation and can weaken blood vessels. The research suggests that those genes associated with these biological systems come together to increase a person’s risk of a ruptured aneurysm. As such, they also now become targets for the development of novel therapies that are aimed at reducing mechanical stress in the brain’s blood vessels as well as lowering local inflammation.

This new understanding might improve the medical outcome for at-risk patients as well as precluding unnecessary medical intervention for those at lower risk who happen to have other risk factors. Given the prevalence of intracranial aneurysms and the high morbidity associated with rupture, such strategies could shift management from reactive emergency treatment to proactive, targeted prevention.

Liu, J-Y., Yuan, J., Luo, L. and Yin, X. (2026) ‘Hypertension-driven mechano-immune crosstalk related novel genes may be potential targets for IA rupture progression’, Int. J. Data Mining and Bioinformatics, Vol. 30, No. 5, pp.1–14.