Cerebral hemorrhage extraction with modified shuffled frog leaping algorithm based on the blood clot clustering

• Published in: Multimedia tools and applications Vol. 83; no. 2; pp. 5653 - 5681
• Main Authors: Fang, Lingling, Jiang, Yumeng
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.01.2024; Springer Nature B.V
• Subjects: Adhesion; Algorithms; Blood; Blood clots; Brain; Clustering; Computed tomography; Computer Communication Networks; Computer Science; Data Structures and Information Theory; Diagnosis; Hemorrhage; Lesions; Multimedia Information Systems; Physicians; Special Purpose and Application-Based Systems; Track 2: Medical Applications of Multimedia; Blood clot extraction; CT images; Shuffled frog leaping algorithm; Clustering algorithm; Cerebral hemorrhage
• ISSN: 1380-7501; 1573-7721
• DOI: 10.1007/s11042-023-15781-4

Rapid extraction of brain lesions can help doctors speed up clinical diagnosis and provide help for follow-up treatment. The spatial location, shape, and distribution of Cerebral hemorrhage are very changeable. Besides, there is a particular case where a blood clot adhesion to the skull (called Skull adhesion). Based on the features of a blood clot, combined with the modified shuffled frog leaping algorithm (MSFLA) and clustering ideas, this paper proposes a C-MSFLA based on the cerebral hemorrhage clot clustering algorithm and establishes the intracranial blood clot extraction framework. Here, the brain uncorrelated tissue is removed by the two-dimensional prefix summation eliminate algorithm, and then the complete blood clot is extracted by regional morphological operation. The proposed method can automatically and accurately extract blood clots. The experiments are tested using clinical cerebral hemorrhage CT images of patients in the Second Affiliated Hospital of Dalian Medical University and verified by the evaluation indicators of JAC , Dice , and Acc . Experiments verify that the proposed method has good performance. It can assist doctors in detecting lesions in time, which can make an efficient, automatic, and accurate clinical diagnosis.