Design and Development of an AI-Driven Mobile Application for Smart Menstrual Cycle Tracking Using Agile Methodology

Abstract

In developing countries such as Nigeria, the unavailability of secure and user-friendly menstrual tracking tool presents an all-pervasive problem for young women. Stable menstrual health condition significantly contributes to maintaining public composure and coordinated mental interactions among the women folk. Mobile app-based smart menstrual tracker represents a cutting-edge innovation for enhancing menstrual health. This study developed a model of mobile app that leverages user-friendly interfaces, advanced artificial intelligence algorithms to monitor, analyze and predict menstrual cycles. The app offers personalized insights into cycle patterns, ovulation and fertility windows, as well as track symptoms such as cramps, mood changes and energy level. The study adopted agile methodology software development process to facilitate integration of predictive algorithm for identifying cycle patterns and forecasting likely future occurrence using previous data. This approach harnessed critical hardware and software tools such as visual studio code editor, React.js with Vite, Node.js with Express and MongoDB to develop and implement both the frontend and backend functionalities of the app. It explored AI models to accurately identify menstrual cycle phases using last short-time memory (LSTM) algorithm to advance privacy-preserving reproductive health monitoring. The app was tested using locally accessed real user data assigned pseudo-names, summary results were presented as test-case I, test-case II and test-case III. The results showed optimal performance where the date-based computation model provided accurate outputs that matched the expected results. This study contributes to women’s health by enhancing flexible user-centric and personalized menstrual health information management system for improved, secure and precision oriented cycle tracking.