Design and Preliminary Evaluation of a Multi-Indicator Intelligent System for Physical Fitness Monitoring Based on Performance Data

Despite the growing use of digital tools in sports sciences, transforming heterogeneous data obtained from physical fitness tests into integrated and interpretable information remains a practical challenge. This study aimed to design and preliminarily evaluate an intelligent multi-index system for collecting, processing, normalizing, scoring, and displaying selected physical fitness data. This developmental study was based on the design and testing of a software prototype. The system was developed using the Python programming language and an object-oriented structure, with tools for data processing, structured data storage, and analytical visualization. Input variables included body mass index, push-ups, pull-ups, core muscle endurance, squats, 2.4-km running, 100-m sprint, and breath-hold time. The system successfully processed individual and population-level data and presented the results through textual reports and multidimensional visualizations. The mean scores obtained from the synthetic data were 68.5% for push-ups, 58.4% for pull-ups, 80.7% for core endurance, 81.2% for squats, 49.5% for aerobic endurance, 72.2% for the 100-m sprint, and 71.2% for breath-hold time. The simple mean of the seven indicators was 68.8%. The findings indicate that the prototype is technically capable of transforming heterogeneous performance data into an integrated and visually interpretable profile and may facilitate the processes of data recording, processing, and reporting. However, the present findings do not establish the clinical or normative validity of the system because the population data were synthetic and some scoring rules were simplified.