{"id":4503,"date":"2026-07-03T11:38:15","date_gmt":"2026-07-03T09:38:15","guid":{"rendered":"https:\/\/crta-robotics.com\/?p=4503"},"modified":"2026-07-03T12:31:30","modified_gmt":"2026-07-03T10:31:30","slug":"inspiration-projekt-na-medunarodnoj-ljetnoj-skoli-o-umjetnoj-inteligenciji-u-medicini","status":"publish","type":"post","link":"https:\/\/crta-robotics.com\/en\/inspiration-projekt-na-medunarodnoj-ljetnoj-skoli-o-umjetnoj-inteligenciji-u-medicini\/","title":{"rendered":"The INSPIRATION project at the international summer school on artificial intelligence in medicine"},"content":{"rendered":"

As part of the INSPIRATION project, led by Assoc. Ph.D. Filip \u0160uligoj, PhD student and young researcher Ines Frajtag participated in the 7th Summer School on Deep Learning for Medical Imaging, held at \u00c9cole de technologie sup\u00e9rieure (\u00c9TS) in Montr\u00e9al, Canada.<\/p>\n\n\n\n

The summer school focused on advanced deep learning methods and their application in medical image analysis, with particular emphasis on the development, training, evaluation, and clinical applicability of artificial intelligence models. The program included lectures, technical sessions, and hands-on workshops covering state-of-the-art approaches in computer vision and medical image management.<\/p>\n\n\n\n

The program covered various neural network architectures used for tasks such as medical image segmentation, classification, and interpretation. Particular attention was paid to models that enable the automated extraction of anatomical structures, pathological changes, or relevant image features from medical images. Such approaches play a significant role in the development of clinical decision support systems and medical procedure planning, as well as in enhancing the precision of image-guided interventions.<\/p>\n\n\n\n

In addition to the models themselves, the program covered practical aspects of developing deep learning-based systems. Discussions addressed data preparation and preprocessing, the organization of training, validation, and testing sets, model training strategies, and the selection of appropriate performance metrics. Particular emphasis was placed on evaluating model robustness, generalization to new data, and the importance of reliable validation prior to potential deployment in a clinical setting.<\/p>\n\n\n\n

One important part of the program addressed challenges particularly prevalent in medical imaging. These include the limited availability of high-quality annotated datasets, the variability of medical images, differences between imaging devices and protocols, class imbalance, and the need for interpretable and clinically reliable models. Discussions also covered the transfer of developed algorithms from the research environment to real-world medical and robotic applications.<\/p>\n\n\n\n

The knowledge acquired is directly linked to research activities within the INSPIRATION project, particularly in the fields of computer vision, medical and biometric data analysis, and non-invasive patient registration. Deep learning methods for image segmentation and interpretation provide a crucial foundation for developing advanced systems capable of contributing to more precise planning and execution of robotic medical procedures.<\/p>\n\n\n\n

Participation in this summer school enabled further professional development in the field of artificial intelligence in medical imaging, the exchange of knowledge with international experts and researchers, and insight into current research trends and challenges regarding the application of deep learning in medicine. This experience further contributes to the development of research competencies within the INSPIRATION project and strengthens international collaboration in the fields of computer vision, medical robotics, and robot-assisted medical procedures.<\/p>\n\n\n\n

The project is funded by the European Union - NextGenerationEU.<\/p>\n\n\n\n