Cerebral aneurysm evolution modeling from microstructural computational models to machine learning: A review. (June 2022)

• Record Type: Journal Article
• Authors: Nabaei, Malikeh
• Is Part Of: Computational biology and chemistry. Volume 98(2022)
• Publication Date: 2022-06
• Subjects: Cerebral aneurysm -- Growth and remodeling -- Microstructure -- Machine learning -- Hemodynamics -- Risk factors; Chemistry -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; Biochemistry -- Data processing; Biology -- Data processing; Molecular biology -- Data processing; Periodicals; Electronic journals; 542.85
• DOI: 10.1016/j.compbiolchem.2022.107676
• Languages: English
• ISSNs: 1476-9271

Abstract

Abstract: Predicting the future behavior of cerebral aneurysms was the target of several studies in recent years. When an unruptured cerebral aneurysm is diagnosed, the physician has to decide about the treatment method. Often more giant aneurysms are diagnosed at higher risk of rupture and are candidates for intervention. However, several clinical and morphological parameters are introduced as risk factors. Therefore, some small size aneurysms with a higher growth rate and rupture risk may be missed. Nowadays, computational models and artificial intelligence can help physicians make more precise decisions, not only according to the aneurysm size. Therefore, the target can be developing a tool that receives the patient history and medical images as input and gives the aneurysm growth rate and rupture risk as output. Achieving this target can be possible by developing a proper computational growth model and using artificial intelligence. This requires knowledge of the vascular microstructure and the procedure of disease development, including degradation and remodeling mechanisms. Moreover, geometrical and clinical risk factors should also be recognized and considered. The present article is a step-by-step indication of this concept. In this paper, first, a review of different computational growth models is presented. Then, the morphological and clinical risk factors are described, and at last, the methods of combining the computational growth models with machine learning are discussed. This review can help the researchers learn the fundamentals and take the proper future steps. Graphical Abstract: ga1 Highlights: Micro-structural computational models predict the variation of vascular constituents, and aneurysm enlargement. Clinical statistical studies determine the risk factors for cerebral aneurysm genesis, progression, and rupture. Machine learning provides a prediction of the future behavior of a diagnosed aneurysm. Combining the above, a rapid and precise tool can be achieved to predict the future behavior of cerebral aneurysms.