Urine Dataset having eight particles classes

Taner Tuncer; Merve Erkuş; Ahmet Çınar; Hakan Ayyıldız; Seda Arslan Tuncer

Urine Dataset having eight particles classes

Quantitative Methods 2023-02-23 v2 Image and Video Processing

Authors: Taner Tuncer , Merve Erkuş , Ahmet Çınar , Hakan Ayyıldız , Seda Arslan Tuncer

Abstract

Urine sediment examination (USE) is one of the main tests used in the evaluation of diseases such as kidney, urinary, metabolic, and diabetes and to determine the density and number of various cells in the urine. USE's manual microscopy is a labor-intensive and time-consuming, imprecise, subjective process. Recently, automatic analysis of urine sediment has become inevitable in the medical field. In this study, we propose a dataset that can be used by artificial intelligence techniques to automatically identify particles in urine sediment images. The data set consists of 8509 particle images obtained by examining the particles in the urine sediment obtained from 409 patients from the Biochemistry Clinics of Elazig Fethi Sekin Central Hospital. Particle images are collected in 8 classes in total and these are Erythrocyte, Leukocyte, Epithelial, Bacteria, Yeast, Cylinders, Crystals, and others (sperm, etc.).

Cite

@article{arxiv.2302.09312,
  title  = {Urine Dataset having eight particles classes},
  author = {Taner Tuncer and Merve Erkuş and Ahmet Çınar and Hakan Ayyıldız and Seda Arslan Tuncer},
  journal= {arXiv preprint arXiv:2302.09312},
  year   = {2023}
}

Comments

3 pages, 2 figures This study includes the introduction of a new dataset

Urine Dataset having eight particles classes

Abstract

Cite

Comments

Related papers