Design of Cost Efficient VBIR Technique Using ICA and IVCA

• Published in: SN computer science 2024-06, Vol.5 (5), p.560, Article 560
• Main Authors: Pradeep Kumar, B. P., Srinidhi, N. N., Shiva Darshan, S. L., Naresh, E.
• Format: Article
• Language: English
• Subjects: AI Based Internet of Healthcare: Analysis and Future Perspectives; Blood vessels; Cancer therapies; Cluster analysis; Clustering; Computer Imaging; Computer Science; Computer Systems Organization and Communication Networks; Contrast agents; Data Structures and Information Theory; Disease; Image analysis; Image contrast; Image databases; Image enhancement; Image retrieval; Independent component analysis; Information retrieval; Information seeking behavior; Information Systems and Communication Service; Magnetic resonance imaging; Measurement techniques; Medical diagnosis; Medical imaging; Original Research; Patients; Pattern Recognition and Graphics; Permeability; Pharmacokinetics; Prostate cancer; R&D; Radiation; Recall; Research & development; Retrieval; Software Engineering/Programming and Operating Systems; Time lag; Tomography; Tumors; Uniqueness; Vector quantization; Vision
• ISSN: 2661-8907; 2662-995X; 2661-8907
• DOI: 10.1007/s42979-024-02936-9

With an emphasis on early-stage contrast agent transit through tumour vasculature, this study presents Adaptive Complex Independent Components Analysis (ACICA) as a unique method for evaluating intravascular responsiveness in prostatic tissue. Furthermore, a new SVM clustering method is introduced that outperforms the conventional k-means clustering for image retrieval based on vision. The study emphasises how dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) may be improved in conjunction with quantitative analysis. Within the field of DCE-MRI, ACICA stands out as a unique intravascular attention measurer. Utilising the spatial independence of extravascular and intravascular magnetic resonance imaging (MR) data, ACICA offers a strong foundation for DCE-MRI image analysis. It incorporates pharmacokinetic modelling to optimise the time lag, especially useful for arterial curves, and a reference region (RR)-based technique to adjust the intravascular concentration curve. The model's evaluation yields outstanding results, with recall and accuracy ranging from 83 to 99% and 82.8% to 99.6%, respectively. The average recall and precision across datasets are 92.86% and 92.82%).All things considered, this study demonstrates the effectiveness of ACICA in evaluating intravascular responsiveness and presents viable paths for enhancing clinical results and diagnostic precision in the treatment of prostate cancer.