AI-Driven Detection of Glomerular Diseases and Automated Clinical Reporting: A System Prototype

Abstract

Artificial intelligence (AI) is present in almost every aspect of our everyday lives. One of the key areas where its influence is promising is medicine. Medical decision-making is a complex process due to the potentially irreversible consequences of misdiagnosis and the lengthy time required to establish an accurate diagnosis. AI may help doctors improve diagnostic procedures and accelerate treatment by performing rapid, accurate analyses. The aim of the paper is to present an AI-driven prototype system for detecting glomerular diseases based on histopathological images and for supporting clinical reporting. The system was developed as part of a project aimed at supporting diagnostic processes using AI models and modern digital solutions. Three main solutions have been developed: (1) the intuitive, user-friendly, and minimalistic system's user interface, which allows for starting a new diagnosis or reviewing archived ones, viewing patients' data, and recording a description; (2) supporting clinical reporting by a speech-to-text and text-to-text system enabling the description of diagnostic results in words in natural language in real time; (3) automated medical image analysis and description, which allows for disease recognition, including the identification of pathological changes and the precise measurement of affected areas, which reduces diagnostic time and improves precision. The goal of the project is to create a system that becomes a reliable help for physicians, an intelligent tool that aids in diagnosis, enabling fast, precise, and error-free medical decision-making.