Data-driven biomarkers outperform theory-based biomarkers in predicting stroke motor outcomes.

Autor:	Olafson ER; Department of Radiology, Weill Cornell Medicine, New York City, New York, USA., Sperber C; Department of Neurology, Inselspital, University Hospital Bern, University of Bern, Bern, Switzerland., Jamison KW; Department of Radiology, Weill Cornell Medicine, New York City, New York, USA., Bowren MD Jr; Department of Neurology, Carver College of Medicine, Iowa City, IA, USA., Boes AD; Departments of Neurology, Psychiatry, and Pediatrics, Carver College of Medicine, Iowa City, IA, USA., Andrushko JW; Department of Physical Therapy, Faculty of Medicine, The University of British Columbia, Vancouver, BC, Canada., Borich MR; Division of Physical Therapy, Department of Rehabilitation Medicine, Emory University School of Medicine, Atlanta, GA, USA., Boyd LA; Department of Physical Therapy, Faculty of Medicine, The University of British Columbia, Vancouver, BC, Canada.; Djavad Mowafaghian Centre for Brain Health, University of British Columbia, Vancouver, BC, Canada., Cassidy JM; Department of Health Sciences, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA., Conforto AB; Hospital das Clinicas HCFMUSP, Faculdade de Medicina, Universidade de Sao Paulo, Sao Paolo, Brazil.; Hospital Israelita Albert Einstein, São Paulo, Brazil., Cramer SC; Dept. Neurology, UCLA; California Rehabilitation Institute, Los Angeles, CA, USA., Dula AN; Department of Neurology, Dell Medical School at The University of Texas Austin, Austin, TX, USA., Geranmayeh F; Clinical Language and Cognition Group. Department of Brain Sciences, Imperial College London, London, United Kingdom., Hordacre B; Innovation, Implementation and Clinical Translation (IIMPACT) in Health, Allied Health and Human Performance, University of South Australia, Adelaide, Australia., Jahanshad N; Imaging Genetics Center, Mark and Mary Stevens Neuroimaging and Informatics Institute, Keck School of Medicine, University of Southern California, Charleston, SC, USA., Kautz SA; Department of Health Sciences & Research, Medical University of South Carolina, Charleston, SC, USA.; Ralph H Johnson VA Health Care System, Charleston, SC, USA., Lo B; Chan Division of Occupational Science and Occupational Therapy, University of Southern California, Los Angeles, CA, USA., MacIntosh BJ; Sandra Black Centre for Brain Resilience and Recovery, Hurvitz Brain Sciences Program, Sunnybrook Research Institute, Toronto, ON, Canada.; Computational Radiology and Artificial Intelligence (CRAI), Department of Physics and Computational Radiology, Clinic for Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway., Piras F; Laboratory of Neuropsychiatry, Santa Lucia Foundation IRCCS, Rome, Italy., Robertson AD; Sandra Black Centre for Brain Resilience and Recovery, Hurvitz Brain Sciences Program, Sunnybrook Research Institute, Toronto, ON, Canada.; Schlegel-UW Research Institute for Aging, Waterloo, ON, Canada., Seo NJ; Department of Health Sciences & Research, Medical University of South Carolina, Charleston, SC, USA.; Ralph H Johnson VA Health Care System, Charleston, SC, USA.; Department of Rehabilitation Sciences, Medical University of South Carolina, Charleston, SC, USA., Soekadar SR; Dept. of Psychiatry and Neurosciences, Charité Campus Mitte (CCM), Charité - Universitätsmedizin Berlin, Berlin, Germany., Thomopoulos SI; Imaging Genetics Center, Mark and Mary Stevens Neuroimaging and Informatics Institute, Keck School of Medicine, University of Southern California, Charleston, SC, USA., Vecchio D; Laboratory of Neuropsychiatry, Santa Lucia Foundation IRCCS, Rome, Italy., Weng TB; Department of Neurology, Dell Medical School at The University of Texas Austin, Austin, TX, USA.; Department of Diagnostic Medicine, Dell Medical School, The University of Texas at Austin, Austin, TX, USA., Westlye LT; Department of Psychology, University of Oslo, Oslo, Norway.; NORMENT, Division of Mental Health and Addiction, Oslo University Hospital, Oslo, Norway., Winstein CJ; Division of Biokinesiology and Physical Therapy, Herman Ostrow School of Dentistry, University of Southern California, Los Angeles, CA, USA.; Department of Neurology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA., Wittenberg GF; Departments of Neurology, Bioengineering, Physical Medicine & Rehabilitation, University of Pittsburgh, Pittsburgh, PA, USA.; GRECC, HERL, Department of Veterans Affairs Pittsburgh Healthcare System, Pittsburgh, PA, USA., Wong KA; Department of Physical Medicine & Rehabilitation, Dell Medical School, University of Texas at Austin, Austin, TX, USA., Thompson PM; Imaging Genetics Center, Mark and Mary Stevens Neuroimaging and Informatics Institute, Keck School of Medicine, University of Southern California, Charleston, SC, USA., Liew SL; Stevens Neuroimaging and Informatics Institute, University of Southern California, Los Angeles, CA, USA., Kuceyeski AF; Department of Radiology, Weill Cornell Medicine, New York City, New York, USA.
Jazyk:	angličtina
Zdroj:	BioRxiv : the preprint server for biology [bioRxiv] 2023 Sep 01. Date of Electronic Publication: 2023 Sep 01.
DOI:	10.1101/2023.06.19.545638
Abstrakt:	Chronic motor impairments are a leading cause of disability after stroke. Previous studies have predicted motor outcomes based on the degree of damage to predefined structures in the motor system, such as the corticospinal tract. However, such theory-based approaches may not take full advantage of the information contained in clinical imaging data. The present study uses data-driven approaches to predict chronic motor outcomes after stroke and compares the accuracy of these predictions to previously-identified theory-based biomarkers. Using a cross-validation framework, regression models were trained using lesion masks and motor outcomes data from 789 stroke patients (293 female/496 male) from the ENIGMA Stroke Recovery Working Group (age 64.9±18.0 years; time since stroke 12.2±0.2 months; normalised motor score 0.7±0.5 (range [0,1]). The out-of-sample prediction accuracy of two theory-based biomarkers was assessed: lesion load of the corticospinal tract, and lesion load of multiple descending motor tracts. These theory-based prediction accuracies were compared to the prediction accuracy from three data-driven biomarkers: lesion load of lesion-behaviour maps, lesion load of structural networks associated with lesion-behaviour maps, and measures of regional structural disconnection. In general, data-driven biomarkers had better prediction accuracy - as measured by higher explained variance in chronic motor outcomes - than theory-based biomarkers. Data-driven models of regional structural disconnection performed the best of all models tested (R 2 = 0.210, p < 0.001), performing significantly better than predictions using the theory-based biomarkers of lesion load of the corticospinal tract (R 2 = 0.132, p< 0.001) and of multiple descending motor tracts (R 2 = 0.180, p < 0.001). They also performed slightly, but significantly, better than other data-driven biomarkers including lesion load of lesion-behaviour maps (R 2 =0.200, p < 0.001) and lesion load of structural networks associated with lesion-behaviour maps (R 2 =0.167, p < 0.001). Ensemble models - combining basic demographic variables like age, sex, and time since stroke - improved prediction accuracy for theory-based and data-driven biomarkers. Finally, combining both theory-based and data-driven biomarkers with demographic variables improved predictions, and the best ensemble model achieved R 2 = 0.241, p < 0.001. Overall, these results demonstrate that models that predict chronic motor outcomes using data-driven features, particularly when lesion data is represented in terms of structural disconnection, perform better than models that predict chronic motor outcomes using theory-based features from the motor system. However, combining both theory-based and data-driven models provides the best predictions. Competing Interests: Competing Interests S.C.C. serves as a consultant for Abbvie, Constant Therapeutics, BrainQ, Myomo, MicroTransponder, Neurolutions, Panaxium, NeuExcell, Elevian, Helius, Omniscient, Brainsgate, Nervgen, Battelle, and TRCare. B.H. has a clinical partnership with Fourier Intelligence. N.J.S. is an inventor for a patent US 10,071,015 B2. C.J. W. is a consultant for Microtransponder, BrainQ, and MedRhythm. G.F.W. sits on Advisory Boards for Myomo and Neuro-innovators.
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu