Overview

Original Research

DIAGNOSIS OF STUDY URODYNAMICS RESULTS OF BLADDER DISEASE BASED ON DEEP LEARNING

ALIREZA IZADI 1, ROOHOLLAH BARZAMINI 2, FARSHID HAJATI 3, and NEGAR JANPORS 4.

Vol 18, No 10 ( 2023 )   |  DOI: 10.5281/zenodo.10069116   |   Author Affiliation: Electrical Engineering, Islamic Azad University Tehran Central Branch, Tehran, Iran 1,2; College of Engineering and Science, Victoria University Sydney, Sydney, Australia 3; Computer Engineering, Islamic Azad University South Tehran Branch, Tehran, Iran 4.   |   Licensing: CC 4.0   |   Pg no: 1358-1371   |   Published on: 24-10-2023

Abstract

Problems in the urinary tract can challenge the quality of life of humans, so that with proper and timely diagnosis of those problems, can be frequent urination, kidney stones and bladder stones or even prostate cancer or eliminate Prevented the kidneys from going. With the correct and quick diagnosis of the disease, the doctor can start the appropriate treatment and the patient will have a good quality of life. Today, to help doctors, artificial intelligence and computer-aided diagnosis (CAD) can play an important role, now deep learning has shown promising performance in computer vision systems [12]. However, no similar action has been taken so far, in this paper, we propose an enhanced approach for classifying diseases related to the urinary tract types using 1 dimensional convolutional neural network. A urologist clinic in Tehran, Iran, has donated the results of its research tests. We evaluate the proposed model on a benchmark dataset containing 1168 test of 3 Diseases related to the urinary tract types (Spastic Neurogenic Bladder, Neurogenic Bladder, and Flaccid Neurogenic Bladder). We have achieved 85 percent accuracy and so far no similar work has been done on this collection.

Keywords

One-dimensional convolutional; Bladder Urodynamic Study; Deep learning; Neural network; Urodynamic test; Bladder strip.