Visiting Researcher
Department of Diagnostic, Interventional and Pediatric Radiology, Inselspital, Bern University Hospital, Switzerland.
Projects
AssessNet-19 is a 2-stage pipeline for assessing COVID-19 severity from CT scans. First, 2D U-Net models segment lungs and lesions from CT slices. Then, 3D volumes are created and radiomics features are extracted, selected, and input into an XGBoost classifier for severity assessment.
Read more [...]
DiagnoseNET is an open source framework for tailoring deep neural networks into different computational architectures from CPU-GPU implementation to multi-GPU and multi-nodes with an efficient ratio between accuracy and energy consumption.
Read more [...]
enerGyPU is an open source monitor-tool, designed to automate the capture, storage and modelling the causal factors that determine an energy efficient execution while the target application is processing on heterogeneous platforms.
Read more [...]
Sepsis Risk Predictor Project: An Open Source Platform for Clinical Data Integration and AI-driven Sepsis Detection ...
Read more [...]Professional Experience
2024-
2023-
Visiting Researcher
Computational Intelligence to Predict Health and Environmental Risks Center (CIPHER), University of North Carolina at Charlotte, USA.
2020- 2023
Postdoctoral Researcher
Medical Image Analysis Group (MIA), ARTORG Center for Biomedical Engineering Research, University of Bern, Switzerland.
2016- 2020
Research Assistant
Models and Algorithms for Artificial Intelligence Group (MAASAI), Inria Centre of Sophia-Antipolis, University of Côte d’Azur, France.
2014- 2016
Research Assistant
High Performance and Scientific Computing Unit, Industrial University of Santander, Colombia.
2015- 2015
Master Internship
Parallel and Distributed Processing Group (GPPD), Institute of Informatics, Federal University of Rio Grande do Sul, Brazil.
2011- 2013
Computer Systems Analyst
Investment Fund Castilla Riopaila Colombina, Cali, Colombia.
Education
University of Bern
Master of Advanced Studies in Translational Medicine and Biomedical Engineering
2022 - 2024
M.A.S. Thesis: AI-Powered Clinical Decision Support Platform for Interstitial Lung Disease
Advisors: Prof. Dr. med. Alexander Pöllinger and mba. Mark Illi
Sitem-Insel School for Translation and Entrepreneurship in Medicine, Switzerland
University of Côte d'Azur
Doctor in Computer Science
2017 - 2021
Ph.D. Thesis: Green Artificial Intelligence to Automate Medical Diagnosis with Low Energy Consumption
Advisors: Prof. Dr. Michel Riveill and Prof. Dr. med. Pascal Staccini
Doctoral School of Computer Science and Technology, France
Industrial University of Santander
Master of Science in Systems and Computer Engineering
2014 - 2016
M.Sc. Thesis: Energy-Aware to Scale Large Scientific Applications on Heterogeneous Architectures
Advisors: Prof. Dr. Carlos J. Barrios H. and Prof. Dr. Philippe O. A. Navaux
School of Systems and Computer Engineering, Colombia
Central University of Valle del Cauca
Bachelor of Science in Systems Engineering (with distinction)
2007 - 2012
B.Sc. Thesis: Implementation of a Grid Computing to Support Research Projects
Advisor: Prof. M.Sc. Vivian M. Orejuela R.
School of Systems and Computer Engineering, Colombia
Publications and Presentations
2023
John A. García H., Arno Depotter, Danielle V. Bower, Herkus Bajercius, et al. A Multi-class Radiomics Method-based WHO Severity Scale for Improving COVID-19 Patient Assessment and Disease Characterization from CT Scans. Investigative Radiology Journal, 2023.
2022
Cristian Toro, Erick Villarreal, Vivian Orejuela, John A. García H. A Machine Learning-based Missing Data Imputation with FHIR Interoperability Approach in Sepsis Prediction. Latin American High-Performance Computing Conference, CARLA2022. Porto Alegre, Brazil.
2021
Herkus Bajercius, John A. García H., Arno Depotter, Maria Barroso, et al. Clinical Evaluation And Multi-Class Delineation Of A Multi-Centric COVID-19 AI-Based Segmentation Study. Conference on Clinical Translation of Medical Image Computing & Computer Assisted Intervention (CLINICCAI) 2021. Strasbourg, France.
2020
John A. García H., Good Practices on Parallel and Distributed Programming for Training Neural Networks. 11th International SuperComputing Camp (SC-CAMP) 2020. Virtual.
John A. García H., Frédéric Precioso, Pascal Staccini and Michel Riveill. DiagnoseNET: Automatic Framework to Scale Neural Networks on Heterogeneous Systems Applied to Medical Diagnosis. International Conference on IT Convergence and Security (ICITCS) 2020. Nha Trang, Vietnam.
2018
John A. García H., Frédéric Precioso, Pascal Staccini and Michel Riveill. Scalability Analysis of Mini-Cluster Jetson TX2 for Training DNN Applied to Healthcare. Nvidia GPU Technology Conference (GTC) Europe 2018. Munich, Germany.
John A. García H., Frédéric Precioso, Pascal Staccini and Michel Riveill. Parallel and Distributed Processing for Unsupervised Patient Phenotype Representation. Latin American High-Performance Computing Conference, CARLA2018. Bucaramanga, Colombia.
2016
John A. García H., Esteban Hernández, Philippe O. A. Navaux and Carlos J. Barrios H. Energy-awareness to Accelerate Large-scale Scientific Applications in Heterogeneous Architectures. The International Conference for High Performance Computing, Networking, Storage, and Analysis (Supercomputing Conference) SC16. Salt Lake City, Utah, USA.
John A. García H., Esteban Hernández, Philippe O. A. Navaux and Carlos J. Barrios H. enerGyPU and enerGyPhi Monitor for Power Consumption and Performance Evaluation on Nvidia Tesla GPU and Intel Xeon Phi. IEEE/ACM 16th International Symposium on Cluster, Cloud and Grid Computing, CCGrid 2016, Cartagena, Colombia.
John A. García H., Philippe O. A. Navaux and Carlos J. Barrios H. enerGyPU for Monitoring Performance and Power Consumption on Multi-GPUs. NIvidia GPU Technology Conference (GTC) 2016. San Jose, USA.