Klasifikasi Dehidrasi Tubuh Manusia Berdasarkan Citra RGB Pada Warna Urine Menggunakan Metode K-Nearest Neighbor

Abstract

Dehydration is a condition in the body where the amount of fluid that comes out is more than the amount of fluid that enters so that the body experiences a lack of fluids. The simplest way to find out if you are dehydrated is to check the color and amount of urine. If it is very dark and there is little water, then the body needs a lot of air. If the urine is clear, it means the body is in normal air balance. The level of dehydration between each person is different, so we need a system that can classify the level of dehydration objectively through urine, so that it can facilitate the process of early detection and diagnosis before being enforced. Based on the research that has been done, it is found that the Matrix Laboratory (Matlab) can classify dehydration in the human body, by utilizing urine images through several processes, namely preprocessing, segmentation, feature extraction of Red Green Blue and then the K-Nearest Neighbor method to classify a person's level of dehydration. From 30 urine image training data and 15 urine image test data with 3 classes of urine color levels, namely not dehydrated, mild dehydration and severe dehydration, the results obtained from the accuracy of the dehydration level classification using the K-Nearest Neighbor method of 93.3% obtained from 14 test data with accurate classification, and 1 test data with inaccurate classification.

Klasifikasi Tingkat Dehidrasi Berdasarkan Warna Urin Menggunakan Metode KNN

“…Namun dengan ukuran piksel yang cukup besar akan memberikan dampak waktu pemrosesan yang waktu cukup lama [18]. Setelah proses resize selesai dilakukan maka citra gambar dari sampel dapat langsung di klasifikasikan ke dalam sistem, kemudian diekstraksi menjadi citra RGB lalu di klasifikasikan menjadi dehidrasi atau tidak dehidrasi menggunakan metode KNN [19]. Penjelasan tingkat dehidrasi secara konvensional disajikan pada Tabel 1.…”

Detecting Dehydration Based on Urine Color Using Fuzzy Logic Image Processing and Regulating Water Intake with an Automatic Water Pump According to Dehydration Level Using an IoT-Based

“…Darker urine color and higher urine viscosity are associated with decreased plasma volume and increased serum osmolality. [4]- [6] Urine color can be used as an indicator to determine body fluid needs. Darker urine color may indicate a higher water need.…”