Page 106 - tech fest 2025 אשדוד חוברת תקצירים

106
“ECGenius”: An Adaptive Learning Platform for ECG Interpretation
SE-B-03
Angelina May Petrovsky; linapetrovsky@gmail.com Mohamed Alfker; mohamedalfker4@gmail.com
Advisors: Dr. Hadassa Daltrophe1, Dr. Tammar Shrot1 1SCE - Shamoon College of Engineering, Ashdod
“ECGenius” is a modern E-learning application designed to enhance the learning experience of medical students and professionals regarding ECG interpretation. This project addresses the critical need for effective ECG diagnostic training by providing an interactive and personalized learning environment. Our platform utilizes a sophisticated quiz generation system that adapts to each student's performance level. Each quiz incorporates real ECG samples and includes multiple-choice questions with only one correct diagnosis. The system tracks user performance over time to continuously adjust the difficulty level and focus of subsequent quizzes. Future development plans include implementing AI-based functionality using heatmaps to identify areas of ECG interpretation where students often err—though its effectiveness is still to be determined. The project’s importance lies in its potential to improve ECG interpretation skills, which are crucial for accurate cardiac diagnosis.
Keywords: adaptive learning, AI, ECG diagnosis, medical education
Gas Detection and Evaluation Using Machine Learning
SE-B-04
Gal Rabinovich; thegal9989@gmail.com Daniel Nekludov; danielnekludov@gmail.com
Advisors: Dr. Hadassa Daltrophe1, Dr. Tammar Shrot1 1SCE - Shamoon College of Engineering, Ashdod
Gas analysis using spectroscopy relies on the unique wavelength absorption patterns of different gases. This project, conducted in collaboration with “Spectronix Ltd.,” integrates spectroscopy with machine learning (ML) to classify gases and quantify their concentrations from spectral data. “Spectronix” specializes in advanced gas detection technologies, contributing expertise in sensor development and industrial applications. Multiple ML models are ‘trained’ and compared to determine the optimal approach for accurate gas identification and measurement. By leveraging pattern recognition, ML enhances accuracy and scalability in gas analysis. Our findings may advance environmental monitoring, healthcare, and industrial safety. This research bridges theoretical advancements using real-world applications, demonstrating the promise of AI-driven spectroscopy for automated and precise gas analysis.
Keywords: absorption spectra, gas analysis, ML, spectroscopy