ARTIFICIAL INTELLIGENCE IN HEALTHCARE: OPTIMIZING MEDICAL IMAGE ANALYSIS FOR MORE ACCURATE AND HUMANIZED DIAGNOSES

Abstract

This study investigates the use of artificial intelligence (AI), with a focus on convolutional neural networks, to improve diagnostic accuracy in diseases such as cancer, through the analysis of clinical images. The research used convolutional neural networks trained on medical imaging data, evaluating metrics such as accuracy, sensitivity, and specificity. Machine learning models specialized in medical image analysis have been developed, aiming at the accurate diagnosis of diseases such as cancer. The platform chosen for prototyping was Orange, which allows you to build machine learning applications without manual coding. This process includes tasks such as data collection, cleanup, and downsizing. The models achieved an accuracy of xxx% in detecting patterns associated with cancer in X-ray images. In addition to technological advances, the study discusses the responsibility of professionals in the management of AI-assisted decisions, as well as the need for ethical validation in the collection of sensitive data. Collaboration between artificial intelligence and healthcare professionals is seen as key to improving disease control and keeping clinical reasoning an important part of the medical field. In short, the results show the potential of AI in transforming medical diagnosis, making them faster and more accurate, but the implementation must be careful to be safe and effective in clinical practice.