Multimodal attention-driven network with blockchain integrity for reliable kidney stone diagnosis.
Authors
Affiliations (2)
Affiliations (2)
- Department of Computer Science, College of Computer Science, Applied College Tanumah, King Khalid University, Abha, Saudi Arabia. [email protected].
- Department of Public Health, College of Applied Medical Sciences, Khamis Mashayt, King Khalid University, Almahalah, 62561, Abha, Saudi Arabia.
Abstract
Renal calculi are the hard mineral deposits that obstructed in the renal system of the human. Dehydration, hormonal imbalance and urinary obstruction leads to formation of renal calculi leads to the chronic Kidney disease end with the renal failure. The early detection and treatment is the key to prevent the irreversible complication like scarring and renal failure. With the Ultrasound imaging, detection of renal calculi remains to be challenging till data due to its speckle noise, low tissue contrast and varying anatomical variability. To address this limitations, our study proposed the multimodal attention driven network with blockchain integrity that integrates the complementary global and local representations of the medical data. The developed cross attention fusion module with the interpretability layer strengthens the decision to extract outputs. Additionally, blockchain with smart contracts is utilized to ensure privacy, integrity, and auditability of data in clinical applications. Experimental outcomes prove that Mutimodal attention-Driven network acquires better diagnostic performance with 98.5% accuracy and 99.1% AUC, making it a secure and trustworthy solution for healthcare environments.