Democritus University of Thrace (DUTH)
Ανάπτυξη έξυπνου συστήματος πρόληψης τραυματισμών σε δρομείς αναψυχής με μεθόδους μηχανικής μάθησης
Abstract
dc:descriptionThis thesis examines the development and evaluation of digital decision support systems for injury prevention and rehabilitation in recreational runners. In this context, an information system was designed and implemented which was initially based on an expert rule-based model, with rules defined by trainers, coaches and physiotherapists with many years of experience in the field of prevention and rehabilitation. These rules were integrated into a web-based platform in the form of if-then decisions and led to the categorization of discomforts, injuries and the display of corresponding exercises via video. A second version of the system was then developed using machine learning algorithms (Naive Bayes). The system was trained and evaluated from the data collected by the above implementation (rule-based system). Decision making in this system is based on data already labeled by experts (labeling), without the need to write new code. This approach allowed to extend the system with additional features while maintaining usability for the end user.Data collection was performed using a structured questionnaire and 170 records were collected, of which 137 included active musculoskeletal complaints. The pilot implementation of the platform showed high user acceptance, significant time spent on the instruction page and frequent repetition of the suggested exercises. The ability to save the link to the results allowed continuous access to the rehabilitation without repeating the process. The results of the study demonstrate that digital technology, combined with expertise and modern data analysis methods, can provide functional and personalized tools to recreational runners, improving compliance, self-care and ultimately quality of life. The system does not replace the specialist, but works supportively, reducing dependency and increasing early intervention. In the future, extending the system to other sporting activities and incorporating metadata such as exercise time, geographic location and climate factors could lead to even more advanced prediction and adaptation models, with the help of secondary machine learning algorithms.
Degree
thesis:*- Grantor dc:publisher
- Democritus University of Thrace (DUTH)
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Tzelepis, Theodoros
- Τζελέπης, Θεόδωρος
Subjects
dc:subject × 16- Μηχανική μάθηση
- Πρόληψη τραυματισμών
- Συστήματα Υποστήριξης Απόφασης
- Δρομείς αναψυχής
- Μαchine Learning
- Injury prevention
- Decision support systems
- Recreational runners
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Φυσικές Επιστήμες
- Τεχνητή νοημοσύνη
- Διεπιστημονικές εφαρμογές επιστήμης ηλεκτρονικών υπολογιστών
- Computer and Information Sciences
- Natural Sciences
- Artificial Intelligence
- Computer Science Interdisciplinary Applications
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/60871