Biology subjects

Beattie, M. S.

Publications and source records attributed to Beattie, M. S..

3 recordsLinked to original sources

bioRxiv2021

Expert-integrated automated machine learning uncovers hemodynamic predictors in spinal cord injury

Automated machine learning (AutoML) is positioned to democratize artificial intelligence (AI) by reducing the amount of human input and ML expertise needed to create prediction models. However, successful translation of ML in biomedicine requires moving beyond optimizing only for prediction accuracy and towards discovering reproducible clinical and biological inferences. Here, we present a model-agnostic framework to reinforce AutoML using strategies and tools of explainable and reproducible AI, including novel metrics for performance precision and feature instability. The framework enables clinicians to interpret AutoML-generated models for clinical and biological verifiability and consequently integrate domain expertise during model development. We applied the framework towards spinal cord injury prognostication and identified a detrimental relationship between intraoperative hypertension and patient outcome. Furthermore, our analysis captured evolving clinical practices such as faster time-to-surgery and blood pressure management that affected clinical model validation. Altogether, we illustrate how augmenting AutoML for inferential reproducibility empowers biomedical discovery and builds trust in AI processes towards effective clinical integration.

bioinformatics↗

bioRxiv2021

Open Data Commons for Preclinical Traumatic Brain Injury Research: Empowering Data Sharing and Big Data Analytics

Traumatic brain injury (TBI) is a major unsolved public health problem worldwide with considerable preclinical research dedicated to recapitulating clinical TBI, deciphering the underlying pathophysiology, and developing therapeutics. However, the heterogeneity of clinical TBI and correspondingly in preclinical studies have made translation from bench to bedside difficult. Here, we present the potential of data sharing, data aggregation, and multivariate analytics to integrate heterogeneity and empower researchers. We introduce the Open Data Commons for Traumatic Brain Injury (ODC-TBI.org) as a user-centered web platform and cloudbased repository focused on preclinical TBI research that enables data citation with persistent identifiers, promotes data element harmonization, and follows FAIR data sharing principles. Importantly, the ODC-TBI implements data sharing at the level of individual subjects, thus enabling data reuse for granular big data analytics and data-hungry machine learning approaches. We provide use cases applying descriptive analytics and unsupervised machine learning on pooled ODC-TBI data. Descriptive statistics included subject-level data for 11 published papers (N = 1250 subjects) representing six distinct TBI models across mice and rats (implementing controlled cortical impact, closed head injury, fluid percussion injury, and CHIMERA TBI modalities). We performed principal component analysis (PCA) on cohorts of animals combined through the ODC-TBI to identify persistent inflammatory patterns across different experimental designs. Our workflow ultimately improved the sensitivity of our analyses in uncovering patterns of pro- vs anti-inflammation and oxidative stress without the multiple testing problems of univariate analyses. As the practice of open data becomes increasingly required by the scientific community, ODC-TBI provides a foundation that creates new scientific opportunities for researchers and their work, facilitates multi-dataset and multidimensional analytics, and drives collaboration across molecular and computational biologists to bridge preclinical research to the clinic.

neuroscience↗

bioRxiv2020

Blood RNA Profiles are Diagnostic for Severity in Human Acute Spinal Cord Injury

Biomarkers of spinal cord injury (SCI) could help determine the severity of the injury and facilitate early critical care decision making. We analyzed global gene expression in peripheral white blood cells during the acute injury phase and identified 197 genes whose expression changed after SCI compared to healthy and trauma controls and in direct relation to SCI severity. Unsupervised co-expression network analysis identified several gene modules that predicted injury severity (AIS grades) with an overall accuracy of 72.7% and included signatures of immune cell subtypes. Our findings indicate that global transcriptomic changes in peripheral blood cells have diagnostic and potentially prognostic value for SCI severity.

neuroscience↗