College of Computer and Information Science, Jouf University, Sakakah, Saudi Arabia.
Comput Intell Neurosci. 2022 Aug 31;2022:1133819. doi: 10.1155/2022/1133819. eCollection 2022.
Subarachnoid hemorrhage (SAH) is one of the serious strokes of cerebrovascular accidents. There is an approx. 15% probability of spontaneous subarachnoid hemorrhage in all acute cerebrovascular accidents (CVAs). Most spontaneous subarachnoid hemorrhages are caused by ruptures of intracranial aneurysms, accounting for about 85% of all occurrences. About 15% of acute cerebrovascular disorders are caused by spontaneous subarachnoid hemorrhage. This illness is mostly caused by brain/spinal arteriovenous malformations, extracranial aneurysms, and hypertension. Computed tomography (CT) scan is the common diagnostic modality to evaluate SAH, but it is very difficult to identify the abnormality. Thus, automatic detection of SAH is required to recognize the early signs and symptoms of SAH and to provide appropriate therapeutic intervention and treatment. In this article, the gray-level cooccurrence matrix (GLCM) is used to extract useful features from CT images. Then, the New Association Classification Frequent Pattern (NCFP-growth) algorithm is applied, which is based on association rules. Then, it is compared with FP-growth methods with association rules and FP-growth methods without association rules. The experimental results indicate that the suggested approach outperforms in terms of classification accuracy. The proposed approach equates to a 95.2% accuracy rate compared to the conventional data mining algorithm.
蛛网膜下腔出血 (SAH) 是脑血管意外的严重类型之一。所有急性脑血管意外 (CVA) 中约有 15%的自发性蛛网膜下腔出血概率。大多数自发性蛛网膜下腔出血是由颅内动脉瘤破裂引起的，约占所有病例的 85%。约 15%的急性脑血管病由自发性蛛网膜下腔出血引起。这种疾病主要由脑/脊髓动静脉畸形、颅内动脉瘤和高血压引起。计算机断层扫描 (CT) 扫描是评估蛛网膜下腔出血的常用诊断方式，但很难识别异常。因此，需要自动检测蛛网膜下腔出血以识别蛛网膜下腔出血的早期迹象和症状，并提供适当的治疗干预和治疗。在本文中，灰度共生矩阵 (GLCM) 用于从 CT 图像中提取有用的特征。然后，应用基于关联规则的新关联分类频繁模式 (NCFP-growth) 算法，然后将其与具有关联规则和不具有关联规则的 FP-growth 方法进行比较。实验结果表明，该方法在分类准确性方面表现出色。与传统的数据挖掘算法相比，该方法的准确率达到了 95.2%。
