The use of artificial intelligence in diagnostic imaging

Abstract

Health metrology plays a crucial role in modern medicine. This discipline faces significant challenges that demand innovative solutions to ensure precise and reliable medical measurements. Health metrology thus positions itself as a fundamental pillar to address future challenges in healthcare, guaranteeing reliable measurements traceable to the International System of Units.A cutting-edge technology on the rise today is Artificial Intelligence (AI), which plays a key role in the healthcare field, enabling more advanced analysis and automation of measurements. Machine Learning (ML) and AI algorithms help us identify subtle patterns, improving early disease detection, with the consequent savings for the healthcare system that this implies.The Spanish National Metrology Institute (CEM) has recently established a Health Laboratory with the purpose of strengthening the metrological infrastructure in the healthcare sector. This laboratory seeks to promote research and development of innovative measurement technologies to address emerging challenges in health metrology. Among its current objetives are:• Exploring the application of Artificial Intelligence and Machine Learning in healthcare, specifically in diagnostic imaging.• Actively participating in European projects and networks related to health metrology.• Contributing and supporting projects in the EMP Health Call 2025, in the specific line dedicated to the health sector.The challenge that CEM aims to address is the evaluation of the quality of AI systems based on the analysis of medical image data from ultrasound and CT scans. Similarly, it intends to study the implementation of AI for analyzing the minimum radiation dose limit that would allow achieving results similar to those obtained with higher dose values and provide quality diagnostic images.This challenge would ensure that doses remain below established legal limits with available AI tools. It will allow identifying response patterns to different radiation doses, predicting the efficacy of lower doses based on individual patient characteristics, and optimizing dosing regimens to maximize efficacy while minimizing side effects.This objective would lead to the reduction of adverse effects and toxicity, optimization of resources and cost reduction in the healthcare system, as well as personalization of treatments based on individual patient characteristics.This new approach to Health metrology that is presented in the coming years will provide the integration of artificial intelligence in terms of specific regulation in metrology and, above all, data protection.This challenge arises from the need for synergy between Machine Learning (ML), Artificial Intelligence (AI), and Personalized and Precision Medicine in the healthcare field. This technological convergence will allow a more precise, personalized, and effective approach to disease diagnosis and treatment.