Artificial intelligence (AI) has shown great promise in the management pathway of cerebrovascular diseases such as brain arteriovenous malformations (AVMs). A brain AVM is a tangle of blood vessels that connects arteries and veins in the brain, disrupting the vital process of oxygen-rich blood being taken from the heart to the brain. AI techniques such as deep learning and machine learning can be used to improve patient outcomes through automated detection, prediction of rupture risk, and outcome prediction after treatment.
Deep learning, a subfield of AI, has been explored for its role and limitations in the aneurysm patient journey. It has also been applied to automated detection and prediction in cerebral arteriovenous malformations and Moyamoya disease. Machine learning techniques have also been used to predict final outcomes with greater accuracy and may help individualize treatment based on key predicting factors.
One study found that machine learning techniques were able to predict final outcome with an accuracy of 97.5%, identifying the presence or absence of nidal fistulae as the most important factor. This is significantly higher than the accuracy of conventional statistical analyses, which had an accuracy of only 43% in predicting final outcome.
Brain AVMs can be treated successfully by endovascular techniques or combined with surgery and radiosurgery with an acceptable risk profile. One study found that machine learning techniques were able to predict final outcome with greater accuracy than conventional statistical analyses. The study recorded clinical presentation, imaging, procedural details, complications, and outcome, and analyzed the data with AI techniques to identify predictors of outcome and assess accuracy in predicting clinical outcome at final follow-up.
The most common clinical presentation was intracranial hemorrhage (56%). Complications included ischemic stroke in 10%, symptomatic hemorrhage in 9.8%, and a mortality rate of 4.7%. The machine learning model showed superior accuracy of 97.5% in predicting outcome and identified the presence or absence of nidal fistulae as the most important factor. In conclusion, AI techniques such as deep learning and machine learning have shown great potential in improving patient outcomes for cerebrovascular diseases such as brain AVMs. These techniques can be used for automated detection, prediction of rupture risk, and outcome prediction after treatment, allowing for more individualized treatment based on key predicting factors.
Be the first to comment