Ibrahim Karabayir

Dr. Ibrahim Karabayir has a strong background in artificial intelligence and mathematics, with early research focusing on integral inequalities and quantum calculus. During his PhD at Istanbul University, he shifted his focus to real-world AI applications, notably developing EVGO, an optimizer algorithm that enhances deep neural network training. Dr. Karabayir conducted cutting-edge research at Wake Forest, UTHSC and Loyola, focusing on critical decision-making problems in cardiovascular diseases, neurodegenerative disorders, surgical outcomes, and microbiome analysis.

Dr. Karabayir excels in developing and implementing novel deep learning architectures and applying advanced AI techniques across diverse domains, including but not limited to CNNs, RNNs, LSTM, ResNet, Diffusion Models, LLMs, Transfer Learning, Boosting algorithms, and Multi-Task Learning. His expertise extends to deploying models on edge devices through techniques like model quantization, pruning, and knowledge distillation. Dr. Karabayir’s current research focuses on cardiovascular diseases, particularly in predicting, detecting, and estimating outcomes such as heart failure, cardiomyopathy, fatal coronary heart disease, and biomarkers like EF, BNP, troponin, and coronary calcium.

His work has been published in prestigious journals/conferences, including IEEE Transactions on Neural Networks and Learning Systems (TNNLS), European Heart Journal DH, Annals of Surgery, American Journal of Obstetrics and Gynecology (AJOG), FASEB, ACC, AHA and AMIA. He is currently serving as investigator on two NIH R01 grants exploring ECG-AI for predicting heart failure and cardiomyopathy in childhood cancer survivors.