Hailong Li, PhD, is a Research Associate in the lab. He received his BS degree in Automation, and MS degree in Control Theory and Engineering from Northeastern University (Shenyang, China). He earned his Ph.D. degree in Computer Science in 2013 from The University of Cincinnati (Cincinnati, OH, US). He finished a post-doctoral training in Biomedical Informatics of Cincinnati Children’s Hospital Medical Center (Cincinnati, Oh, US) in 2016. His research focuses on applying machine learning algorithms to medical image analysis. Specifically, he applied novel deep learning models (e.g., deep neural networks (DNN), deep convolutional neural networks (CNN), U-Net, and graph convolutional networks (GCN), etc.) to diagnosis and prognosis of neurodevelopment, Autism, ADHD, as well as liver fibrosis. His current studies include risk prediction in very preterm neonates using structural and functional MRI data, Autism of adolescents and adults using functional connecotme, and liver stiffness and fibrosis of adolescents using anatomical MRI abdominal images. He is familiar with advanced machine learning tools, such as TensorFlow, Keras, and Python.
Machine learning; deep learning; medical image processing and analysis
Post-doc Research Fellow
Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, 2016.
Doctor of Philosophy (Ph.D.)
Computer Science and Engineering, University of Cincinnati, Cincinnati, OH, 2013.
Master of Science (M.S.)
Control Theory and Engineering, Northeastern University, China, 2007.
Bachelor of Science (B.S)
Automation, Northeastern University, China, 2004
ConCeptCNN: A novel multi-filter convolutional neural network for the prediction of neurodevelopmental disorders using brain connectome.
Introduction: Deep convolutional neural network (CNN) and its derivatives have recently shown great promise in the prediction of brain disorders using brain connectome data. Existing deep CNN methods using single global row and column convolutional filters have limited ability to extract discriminative information from brain connectome for prediction tasks. Our study presents a novel deep Connectome–Inception CNN (ConCeptCNN) model, which is developed based on multiple convolutional filters. The proposed model is used to extract topological features from brain connectome data for neurological disorders classification and analysis. The ConCeptCNN uses multiple vector-shaped filters to extract topological information from the brain connectome at different levels for complementary feature embeddings of brain connectome. The proposed model is validated using two datasets: the Neuro Bureau ADHD-200 dataset and the Cincinnati Early Prediction Study (CINEPS) dataset. In a cross-validation experiment, the ConCeptCNN achieved a prediction accuracy of 78.7% for the detection of attention deficit hyperactivity disorder (ADHD) in adolescents and an accuracy of 81.6% for the prediction of cognitive deficits at 2 years corrected age in very preterm infants. In addition to the classification tasks, the ConCeptCNN identified several brain regions that are discriminative to neurodevelopmental disorders.