AVANÇOS TECNOLÓGICOS NA RADIOLOGIA: UMA REVISÃO INTEGRATIVA SOBRE O IMPACTO DA INTELIGÊNCIA ARTIFICIAL NO DIAGNÓSTICO POR IMAGEM

INTRODUCTION: Technological advancements, particularly in Artificial Intelligence (AI), have led to profound transformations in diagnostic imaging, one of the most impacted fields of modern medicine. Techniques such as machine learning and deep neural networks hold the potential to enhance accuracy, efficiency, and speed in the analysis of radiological exams, providing a powerful tool for radiologists. METHODOLOGY: This study was conducted as an integrative literature review. The guiding question was formulated using the PICO strategy, aiming to understand the impact of AI in diagnostic imaging. Inclusion criteria covered articles published between 2014 and 2024, in Portuguese or English, addressing AI's impact on radiology. The search was conducted across PubMed, Scopus, Web of Science, and IEEE Xplore databases, using specific descriptors related to AI and radiology. Data analysis followed a content analysis approach, categorizing results into three main themes: benefits, limitations, and the impact of AI on radiologists' roles. RESULTS: After an initial screening of 734 articles, eight studies were selected for final analysis, based on the inclusion criteria. Key findings indicate that AI is improving the accuracy and efficiency of diagnostic imaging, particularly in early disease detection. However, the analysis also highlighted challenges related to the implementation of the technology, such as ensuring the reliability of algorithms and proper professional training. Additionally, AI was viewed as a complementary tool for radiologists, rather than a substitute. DISCUSSION: The application of AI in diagnostic imaging has shown to be highly beneficial in various aspects. Studies demonstrate that AI enhances accuracy, enables early pathology detection, and improves radiologists' efficiency. However, the implementation of this technology is not without challenges. Trust in algorithms and ethical concerns surrounding patient data use remain critical issues. Although highly promising, AI requires careful integration into radiologists' workflows, ensuring it is used as a complement to human expertise rather than a replacement. Continuous professional training and updates are essential for the effective and ethical deployment of AI. CONCLUSION: Artificial Intelligence has a transformative impact on diagnostic imaging, offering significant benefits such as increased accuracy, efficiency, and early detection capabilities. However, its implementation in radiology should be done carefully, focusing on professional training, data security, and human oversight. AI should be viewed as a supportive tool, enhancing radiologists' work rather than replacing their skills. Future research should continue to explore the ethical, operational, and technical challenges of AI, as well as evaluate its long-term impact on clinical practice.