Detection and Identification of Triple Phosphate Crystals and Calcium Oxalate Crystals in Human Urine Sediment Using Harr Feature, Adaptive Boosting and Support Vector Machine via Open CV
Jessie R. Balbin, Glenn V. Magwili, Leonardo D. Valiente, Den Leonard B. Gawaran, Neil Ezekiel R. Lumapas, Anthony M. Umali · 2020
The importance of pattern recognition technology to help doctors during clinical diagnosis for urine crystals will be the topic in this paper. The paper focused on creating a program that would detect and identify both triple phosphate and calcium oxalate crystals from a urine sediment sample. By doing so, the professional who is supposed to conduct the diagnosis through visual inspection of microscope slides containing urine sample, aiming to identify crystals, bacteria and other relevant elements, resulting in a laborious task, will take less time in doing so and furthermore, lessen the probability for human error. Tests showed that the system was capable of detecting and classifying with accuracy of higher than 90%, individual as high as 89.70% and specificity of over 93% for both crystals.