Paul Le

Date of Award

12-8-2023

Document Type

Thesis

Publisher

Santa Clara : Santa Clara University, 2023.

Department

Computer Science and Engineering

First Advisor

David C. Anastasiu

Abstract

Chronic kidney disease is increasingly recognized as a leading public health problem over the world that affects more than 10 percent of the population worldwide, where electrolytes and wastes can build up in your system. Kidney failure might not be noticeable until more advanced stages where it may then become fatal if not for artificial filtering or a transplant. As a result, it is important to detect kidney disease early on to prevent it from progressing to kidney failure. The current main test of the disease is a blood test that measures the levels of a waste product called creatine and needs information such as age, size, gender, and ethnicity. They may be uncomfortable, can lead to infections, and are inconvenient and expensive.

I will re-engineer an Android application for Chronic Kidney Disease detection by working on test strip detection zone localization, detection zone focus, capture quality, and dynamic model loading. This uses a smartphone’s camera and allows users to manually focus on an area of the view to analyze. The camera detects where the test strip and its detection zone is and checks if it is in focus. The pixels are sent to the machine learning algorithm. The application can quickly determine the health of a users kidney and can display it. By only requiring a few drops of blood and an Android smartphone, it is very important for those who cannot afford insurance or live in developing countries. This can make a huge difference in early detection of CDK in these areas where people would otherwise disregard the tests in fear of not having enough money.


Download

Share

COinS