Machine learning to infer neurocognitive testing scores among adolescents and young adults with congenital heart disease.
Authors
Affiliations (27)
Affiliations (27)
- Department of Pediatrics, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA.
- Departments of Neurology and Pediatrics, University of Rochester School of Medicine and Dentistry, Rochester, NY, USA.
- Department of Pediatrics, University of Toronto, Holland Bloorview Kids Rehabilitation Hospital, Toronto, ON, Canada.
- Department of Neurology and Department of Psychiatry, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA.
- Departments of Genetics and Pediatrics, Yale University School of Medicine, New Haven, CT, USA.
- Departments of Surgery and Pediatrics, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA.
- Mindich Child Health and Development Institute and Department of Pediatrics, Icahn School of Medicine at Mount Sinai, New York, NY, USA.
- Division of Cardiology, Children's Hospital of Philadelphia, Department of Pediatrics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA.
- Center for Multimodal Imaging and Genetics, and Department of Radiology, School of Medicine, University of California San Diego, La Jolla, CA, USA.
- Department of Radiology, Children's Hospital of Philadelphia, University of Pennsylvania, Philadelphia, PA, USA.
- Department of Pediatrics, Benioff Children's Hospital, University of California, San Francisco, San Francisco, CA, USA.
- Department of Pediatrics, Primary Children's Hospital, University of Utah, Salt Lake City, UT, USA.
- Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, NY, USA.
- Department of Pediatrics, University of Rochester Medical Center, Rochester, NY, USA.
- Department of Psychiatry, The Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA.
- Department of Neuroscience, University of Rochester Medical Center, Rochester, NY, USA.
- School of Medicine, University of California, San Francisco, San Francisco, CA, USA.
- Department of Systems Biology & Department of Biomedical Informatics, Columbia University, New York, NY, USA.
- Department of Cardiology, Boston Children's Hospital, Boston, MA, USA.
- Department of Pediatrics, Harvard Medical School, Boston, MA, USA.
- Department of Radiology, Harvard Medical School, Boston, MA, USA.
- Department of Pediatrics, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA. [email protected].
- Department of Pediatrics, Harvard Medical School, Boston, MA, USA. [email protected].
- Department of Pediatrics, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA. [email protected].
- Department of Pediatrics, Harvard Medical School, Boston, MA, USA. [email protected].
- Department of Radiology, Harvard Medical School, Boston, MA, USA. [email protected].
- Computational Health Informatics Program, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA. [email protected].
Abstract
Congenital heart disease (CHD) affects about 1% of births and is linked to differences in thinking and learning. Understanding how birth, genetic, clinical, and environmental factors together explain cognitive variability can inform monitoring and care. This study builds a multivariate model predicting cognition across multiple domains in adolescents and young adults with CHD. We studied 89 adolescents and young adults (AYAs; mean age 16 years) with CHD who completed structural and diffusion MRI and fifteen neurocognitive tests across seven domains. Using an enhanced forward-inclusion and backward-elimination strategy with cross-validation, we built multivariate models incorporating biological, socioeconomic, clinical, genetic, and brain imaging features. Performance was evaluated using Pearson correlation (<math xmlns="http://www.w3.org/1998/Math/MathML"><mi>r</mi></math>) between observed and inferred scores, mean absolute error (MAE), and inverse inferability score (IIS). Here we show that models infer scores with moderate accuracy (<math xmlns="http://www.w3.org/1998/Math/MathML"><mi>r</mi></math> = 0.245-0.648; MAE = 1.6-12.0 points; mean MAE = 6.3). Highest correlations include Digit Span (<math xmlns="http://www.w3.org/1998/Math/MathML"><mi>r</mi></math> = 0.65; p < 0.001), Verbal Comprehension Index (<math xmlns="http://www.w3.org/1998/Math/MathML"><mi>r</mi></math> = 0.594; p < 0.001), and Matrix Reasoning (<math xmlns="http://www.w3.org/1998/Math/MathML"><mi>r</mi></math> = 0.574; p < 0.001). Domain ranking by IIS shows the best (lowest) scores for general intelligence (0.0886), followed by working memory (0.7100), and a higher (worse) score for perceptual reasoning (1.9199). A multivariate approach combining brain imaging with genetic, clinical, and environmental factors provides clinically meaningful inference of individual cognitive performance in AYAs with CHD. These findings suggest complementary roles of brain, genetic, and contextual factors in shaping cognitive variability and motivate validation in larger cohorts.