Franko Hržić

Franko Hržić was born in 1993 in Rijeka, Croatia. At the Faculty of Engineering University of Rijeka, he obtained all three degrees: a bachelor’s and master’s degree in computer engineering and a PhD, with the latter acquired in 2022.

Currently, he works as a teaching assistant at the same university. Throughout his doctoral studies, he collaborated extensively with doctors from the Division of Pediatric Radiology, Department of Radiology, at the Medical University of Graz in Austria. This collaboration ultimately resulted in the Scandinavian format of his doctoral thesis and These en Cotuttelle (joint degree) from the Medical University of Graz. His research primarily focuses on addressing practical problems in radiology, intending to enhance the quality of radiologists’ work, support their decision making and reduce their workload.

In the field of medicine, Computed Tomography, X-ray screening, and Magnetic Resonance Imaging play a crucial role in diagnosis setting and medical operations planning. One of the possible reasons to get a medical screening is trauma caused by an external force on the body which may cause various internal damage. In the domain of pediatric radiology, fractures of the wrist, ulna, and radius bones are among the most commonly encountered fractures.

This presentation will offer challenges of fracture detection and machine-learning-based solutions for those challenges. The challenges encompass fracture detection and segmentation through YOLO-based and U-NET-based models,  a Cycle-GAN-based approach for cast removal and estimating the age of fractures while incorporating a simple measure of uncertainty.