Digital twins are virtual replicas of physical objects or systems. This new technology is increasingly being adopted in industry to improve the monitoring and efficiency of products and organizations. In healthcare, digital human twins (DHTs) represent virtual copies of patients, including tissues, organs, and physiological processes. Their application has the potential to transform patient care in the direction of increasingly personalized data-driven medicine. The use of DHTs can be integrated with digital twins of healthcare institutions to improve organizational management processes and resource allocation. By modeling the complex multi-omics interactions between genetic and environmental factors, DHTs help monitor disease progression and optimize treatment plans. Through digital simulation, DHT models enable the selection of the most appropriate molecular therapy and accurate 3D representation for precision surgical planning, together with augmented reality tools. Furthermore, they allow for the development of tailored early diagnosis protocols and new targeted drugs. Furthermore, digital twins can facilitate medical training and education. By creating virtual anatomy and physiology models, medical students can practice procedures, enhance their skills, and improve their understanding of the human body. Overall, digital twins have immense potential to revolutionize healthcare, improving patient care and outcomes, reducing costs, and enhancing medical research and education. However, challenges such as data security, data quality, and data interoperability must be addressed before the widespread adoption of digital twins in healthcare. We aim to propose a narrative review on this hot topic to provide an overview of the potential applications of digital twins to improve treatment and diagnostics, but also of the challenges related to their development and widespread diffusion.

Background and purpose: Accurate instance segmentation of teeth in panoramic dental X-rays is a challenging task due to variations in tooth morphology and overlapping regions. In this study, we propose a new algorithm, for instance, segmentation of the different teeth in panoramic dental X-rays. Methods: An instance segmentation model was trained using the architecture of a Mask Region-based Convolutional Neural Network (Mask-RCNN). The data for the training, validation, and testing were taken from the Tuft dental database (1000 panoramic dental radiographs). The number of the predicted label was 52 (20 deciduous and 32 permanent). The size of the training, validation, and test sets were 760, 190, and 70 images, respectively, and the split was performed randomly. The model was trained for 300 epochs, using a batch size of 10, a base learning rate of 0.001, and a warm-up multistep learning rate scheduler (gamma = 0.1). Data augmentation was performed by changing the brightness, contrast, crop, and image size. The percentage of correctly detected teeth and Dice in the test set were used as the quality metrics for the model. Results: In the test set, the percentage of correctly classified teeth was 98.4%, while the Dice score was 0.87. For both the left mandibular central and lateral incisor permanent teeth, the Dice index result was 0.91 and the accuracy was 100%. For the permanent teeth right mandibular first molar, mandibular second molar, and third molar, the Dice indexes were 0.92, 0.93, and 0.78, respectively, with an accuracy of 100% for all three different teeth. For deciduous teeth, the Dice indexes for the right mandibular lateral incisor, right mandibular canine, and right mandibular first molar were 0.89, 0.91, and 0.85, respectively, with an accuracy of 100%. Conclusions: A successful instance segmentation model for teeth identification in panoramic dental X-ray was developed and validated. This model may help speed up and automate tasks like teeth counting and identifying specific missing teeth, improving the current clinical practice.

**Abstract text** Introduction The integration of Artificial Intelligence (AI) in radiology has gained significant attention in recent years. To | POSTER

Purpose: This study examines patients' perspectives on the integration of artificial intelligence (AI) in radiology through focus groups, aiming to identify the main issues and areas for improvement. It is part of a larger research project that employs various methodologies to explore the views of both patients and radiologists regarding AI tools. Methods: We conducted two focus groups using a narrative story and vignettes: one with patients who self-assessed as AI experts and the other with non-AI experts. Results: The focus groups revealed diverse opinions on AI use in diagnostics, focusing on six main topics: acceptance, concerns, communication between radiologists and patients, explainability of AI, medical records, and emotional aspects. Conclusions: The findings underscore the importance of developing patient-centric AI solutions to build trust in AI-assisted diagnostic tools, considering emotional and communicational aspects and addressing both physician and patient concerns to facilitate smoother integration of AI in radiology.