Track: Artificial Intelligence

Abstract

Supervised training of deep learning models has become a tool for resolving multidisciplinary issues for the different disciplines around the world. The medical field is one of these areas mentioned where machine learning participates for medical image analysis to improve and support the existing medical instrumentation. Globally, different diagnostic tests for urinary tract infections (UTIs) are used in clinical practices. However, UTI’s tend to be asymptomatic for people and cause complications through time because of late detection. In this project we propose the development of an automatic urinalysis system for pyuria detection. For this, a convolutional neural network is trained and tested to identify White Blood Cells (WBCs). The automatic urinalysis system is able to detect and count the number of WBCs in a urine sample.