Mining multi-site clinical data to develop machine learning MRI biomarkers: application to neonatal hypoxic ischemic encephalopathy

• Published in: Journal of translational medicine 2019-11, Vol.17 (1), p.385-385, Article 385
• Main Authors: Weiss, Rebecca J, Bates, Sara V, Song, Ya'nan, Zhang, Yue, Herzberg, Emily M, Chen, Yih-Chieh, Gong, Maryann, Chien, Isabel, Zhang, Lily, Murphy, Shawn N, Gollub, Randy L, Grant, P Ellen, Ou, Yangming
• Format: Article
• Language: English
• Subjects: Age; Algorithms; Archives & records; Artificial intelligence; Big data; Biological markers; Biomarkers; Biomarkers - metabolism; Brain cancer; Brain damage; Clinical trials; Clinical Trials as Topic; Computational biology; Data collection; Diagnostic imaging; Encephalopathy; Experts; Hospitals; Humans; Hypoxia; Hypoxia-Ischemia, Brain - diagnostic imaging; Hypoxic ischemic encephalopathy; Infant, Newborn; Informatics; International Classification of Diseases; Ischemia; Learning algorithms; Machine Learning; Magnetic Resonance Imaging; Medical records; Medical research; Mining industry; Mortality; MRI; Neonatal encephalopathy; Neonates; Newborn infants; Outcome prediction; Patients; Probability; Product development; Protocol; Traumatic brain injury; Treatment Outcome; Tumors
• ISSN: 1479-5876; 1479-5876
• DOI: 10.1186/s12967-019-2119-5

Secondary and retrospective use of hospital-hosted clinical data provides a time- and cost-efficient alternative to prospective clinical trials for biomarker development. This study aims to create a retrospective clinical dataset of Magnetic Resonance Images (MRI) and clinical records of neonatal hypoxic ischemic encephalopathy (HIE), from which clinically-relevant analytic algorithms can be developed for MRI-based HIE lesion detection and outcome prediction. This retrospective study will use clinical registries and big data informatics tools to build a multi-site dataset that contains structural and diffusion MRI, clinical information including hospital course, short-term outcomes (during infancy), and long-term outcomes (~ 2 years of age) for at least 300 patients from multiple hospitals. Within machine learning frameworks, we will test whether the quantified deviation from our recently-developed normative brain atlases can detect abnormal regions and predict outcomes for individual patients as accurately as, or even more accurately, than human experts. Trial Registration Not applicable. This study protocol mines existing clinical data thus does not meet the ICMJE definition of a clinical trial that requires registration.