Scholarlink

Background <p>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.</p> Methods <p>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 (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(r\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>r</mi> </math></EquationSource> </InlineEquation>) between observed and inferred scores, mean absolute error (MAE), and inverse inferability score (IIS).</p> Results <p>Here we show that models infer scores with moderate accuracy (<InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(r\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>r</mi> </math></EquationSource> </InlineEquation> = 0.245–0.648; MAE = 1.6–12.0 points; mean MAE = 6.3). Highest correlations include Digit Span (<InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(r\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>r</mi> </math></EquationSource> </InlineEquation> = 0.65; p &lt; 0.001), Verbal Comprehension Index (<InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(r\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>r</mi> </math></EquationSource> </InlineEquation> = 0.594; p &lt; 0.001), and Matrix Reasoning (<InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(r\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>r</mi> </math></EquationSource> </InlineEquation> = 0.574; p &lt; 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).</p> Conclusions <p>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.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine learning to infer neurocognitive testing scores among adolescents and young adults with congenital heart disease

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月6日

查看全文
  • Mohammad Arafat Hussain,
  • Sheng He,
  • Heather R. Adams,
  • Evdokia Anagnoustou,
  • David C. Bellinger,
  • Martina Brueckner,
  • Wendy K. Chung,
  • John Cleveland,
  • Bruce D. Gelb,
  • Elizabeth Goldmuntz,
  • Donald J. Hagler Jr.,
  • Hao Huang,
  • Patrick McQuillen,
  • Thomas A. Miller,
  • Ami Norris-Brilliant,
  • George A. Porter Jr.,
  • Nina Thomas,
  • Madalina E. Tivarus,
  • Duan Xu,
  • Yufeng Shen,
  • Jane W. Newburger,
  • P. Ellen Grant,
  • Sarah U. Morton,
  • Yangming Ou

摘要

Background

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.

Methods

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 ( \(r\) r ) between observed and inferred scores, mean absolute error (MAE), and inverse inferability score (IIS).

Results

Here we show that models infer scores with moderate accuracy ( \(r\) r  = 0.245–0.648; MAE = 1.6–12.0 points; mean MAE = 6.3). Highest correlations include Digit Span ( \(r\) r  = 0.65; p < 0.001), Verbal Comprehension Index ( \(r\) r  = 0.594; p < 0.001), and Matrix Reasoning ( \(r\) r  = 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).

Conclusions

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.