Authors:
Dr. R. Rajasree, Assistant Professor/ECE, PSN Institute of Technology and Science, Tirunelveli, Tamil Nadu, India
K. Haritha, Assistant Professor/AI&DS, Vel Tech Multi Tech Dr.Rangarajan Dr.Sakunthala Engineering College, Avadi, Chennai, , Tamil Nadu, India
Mr. S. Santhosh Kumar, Research Scholar, Mechanical Engineering, St.Xaviers Catholic College of Engineering, Nagercoil, Tamil Nadu, India
Editors: Dr. Thirumurugan Shanmugam and Dr. Shweta A. Bansal
ISBN: 978-81-963849-7-5
DOI: https://doi.org/10.59646/csebookc19/004
Abstract:
Postoperative urinary retention is a medical condition characterized by the inability to urinate despite a full bladder. Bladder ultrasound imaging plays a crucial role in early diagnosis and management of urinary retention by estimating urine volume. Additionally, it can help reduce the need for urinary catheters, lowering the risk of catheter-associated urinary tract infections. The combination of wearable ultrasound devices and machine learning-based bladder volume estimation algorithms eases the workload of healthcare professionals in hospital settings and enhances outpatient care. However, existing algorithms are resource-intensive, requiring expensive GPUs for computation. In this study, we have developed and validated a memory-efficient deep learning model that demands minimal computational resources for precise bladder segmentation and urine volume calculation using B-mode ultrasound images from 360 patients for training and validation, and 74 patients for testing.