We're excited to share the fantastic accomplishments of our TO_AITION consortium partners at the 45th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), held at the International Convention Centre (ICC) in Sydney, Australia, from July 24th to 27th, 2023. Work Presented:
A Comparison Study on Creating Simulated Patient Data for Individuals Suffering from Chronic Coronary Disorders: Angela Koloi, Vasileios S. Loukas, Antonis Sakellarios, Jos A. Bosch, Rick Quax, Karina Nowakowska, Nikolaos Tachos, Jakub Kaźmierski, Costas Papaloukas, Dimitrios Fotiadis In brief— The emergence of virtual population (VP) and synthetic data generation presents promising avenues to augment clinical research databases. In this study, we conducted a comparative analysis of five distinct approaches for generating virtual data populations from real patient data related to chronic coronary disorders. Our findings demonstrate the effectiveness of the Gaussian Copula Model in creating high-quality virtual data with consistent distributions and correlation patterns. Clinical Relevance— Utilizing a virtual population can significantly enhance the predictive performance of machine learning models, surpassing smaller non-augmented populations.
Graph-guided Gaussian Process-based Diagnosis of CVD Severity with Uncertainty Measures: Stavroula Tassi, Vassiliki Kigka, Panagiotis Siogkas, Silvia Rocchiccioli, Gualtriero Pelosi, Dimitrios I. Fotiadis, IEEE Fellow, Antonis I. Sakellarios In brief— This work introduces a Gaussian process (GP) framework over graphs for predicting the Fractional Flow Reserve (FFR) index, a crucial diagnostic tool for coronary artery disease, using only non-invasive imaging and clinical features. The novel medical EGP (MedEGP) method shows impressive merits in predicting the FFR index with well-calibrated uncertainty, establishing an accurate non-invasive approach for coronary artery disease diagnosis. Clinical Relevance— Our paper presents a significant advancement in non-invasive prediction of FFR, enhancing the diagnosis of coronary artery disease.
Congratulations to the teams from UVA and UOI and LODZ for their outstanding contributions to the IEEE EMBC conference. Their innovative research underscores the commitment of the TO_AITION consortium to advancing biomedical engineering and healthcare innovation. We eagerly anticipate further advancements in this field driven by collaborative efforts and groundbreaking research.