Towards the Selection of the Best Machine Learning Techniques and Methods for Urinalysis

Urinalysis is a significant technique used for determining and inspecting the urinary system. Urine has numerous chemical materials secreted; these materials can be used to diagnose diseases and conditions such as urinary tract infection, diabetes, kidney diseases, and pregnancy, at the earliest. Accessing medical care and health screenings are quite essential, but it has become extremely difficult since screening techniques are either very expensive or convoluted for people of low-income communities. Despite the fact, that numerous urinalysis methods and techniques have been brought forth by researchers over the years, no research has been conducted to scrutinize and review state-of-the-art developments in the stated area. Hence, this research aims to conduct a Systematic Literature Review of urinalysis methods proposed and assessed from 2007 to 2019 and recognizes 33 studies. This leads to the identification of 10 methods, 8 technologies, 12 challenges, and 10 diseases. An insight analysis of the identified methods and models reveals that Genetic Based Fuzzy Classifying Method and Automatic Urinary Particle Recognition Method are the most optimum options due to the fact that their computational time is minimum as well as they do not require any supportive hardware. Moreover, the analysis of technologies reveals that Mobile App (Augmented Reality) is the best option among the identified technologies. It has also been discovered that machine/deep learning classification techniques have been used to classify the sample and provide reliable and accurate results within time. Nevertheless, a complete analysis of methods and technologies has been presented. The findings of this article are extremely useful for practitioners as well as academics of the area.