Intelligent clinical decision support system for intracranial aneurysm lesion-based detection using biomedical imaging data

An intracranial aneurysm (IA) is a type of cerebrovascular condition that occurs due to focal bulging or expansion in the wall of a cerebral artery, most frequently affecting the internal carotid artery. An IA is seemingly connected with conditions such as fibrous dysplasia, adult-dominant polycystic kidney disease (ADPKD), and coarctation of aorta. A ruptured brain aneurysm can result in a severe and life-threatening subarachnoid hemorrhage if not treated early, often leading to long-lasting neurological damage and a higher chance of death.

However, the traditional diagnostic technique is time-consuming, and it necessitates a trained and skilled medical practitioner. Hence, building an automated model to help physicians precisely detect IA can reduce the clinicians’ burden. Recently, deep learning has gained significant interest in the domain of medical imaging as it automatically learns patterns and accurately detects several types of diseases, such as IAs.

Therefore, this article introduces a Clinical Decision Support System for Intracranial Aneurysm Lesion-Based Detection (CDSS-IALD). The main purpose of this work is to develop a robust system that can automatically detect and identify intracranial aneurysms using medical imaging data to assist clinicians with treatment decisions. Initially, the proposed CDSS-IALD model pre-processes the input biomedical images using image normalization, noise reduction, contrast improvement, and skull stripping to refine image quality and consistency for subsequent analysis.

Following that, a ConvNeXt hierarchical model is employed for feature extraction, which effectively extracts local details and global context from pre-processed images. Finally, these extracted representations are then forwarded to a spiking neural network, which performs accurate aneurysm detection and classification. To ensure the enhanced performance of the CDSS-IALD system, extensive experiments were conducted on the benchmark Intracranial Aneurysm Dataset.

The experimental outcomes clearly demonstrate the betterment of the proposed CDSS-IALD system over the other approaches in detecting and localizing intracranial aneurysms from biomedical imaging. Open access funding provided by Vellore Institute of Technology- AP University. Department of Electronic and communication Engineering, Mahatma Gandhi Institute of Technology, Gandipet, Hyderabad, Telangana, India Department of Computer Science and Engineering, Aditya University, Surampalem, India Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, Guntur, AP, India Department of Computer Science and Engineering, Malla Reddy University, Hyderabad, 500100, India Principal & Professor, Capital Engineering College, Bajpur Khordha, Mahatpalla, Bhuabneswar, Odisha, India School of Computer Engineering, KIIT Deemed to be University, Bhubaneswar, India School of Electronics Engineering, VIT-AP University, Amaravati, India The authors declare no competing interests.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This article is published under an open access license. Please check the 'Copyright Information' section either on this page or in the PDF for details of this license and what re-use is permitted.