Machine learning based analysis for intellectual disability in Down syndrome

Q: What is the goal of the machine learning analysis in this study?

To apply ML techniques to clinical records of individuals with Down syndrome and analyze their association with intellectual disability, using age equivalent scores as the outcome indicator.

Q: Which machine learning models are used and what inputs are analyzed?

The study uses random forest and gradient boosting machine models, analyzing 109 features across 106 DS subjects.

Q: How do the results help interpret factors linked to cognitive impairment in Down syndrome?

The models show low prediction error and acceptable R2, while feature ranking highlights variables related to hearing, gastrointestinal alterations, thyroid status, immune system, and vitamin B12 as candidates for focused care pathway improvement.

Aug 6, 2026

Upload by Logic

Down syndrome (DS) and trisomy 21 are the most frequent genetic causes of intellectual disability (ID), yet a clear pathogenic mechanism remains unidentified. The study leverages the growing availability of big data and electronic health records to apply machine learning for analyzing clinical records of DS subjects. Two tree-based models—random forest and gradient boosting—assess 109 variables across 106 individuals, using the age equivalent score as an indicator of intellectual functioning. Model tuning includes Boruta-based feature selection, data augmentation, and age-effect mitigation.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Abstract

Introduction

Heliyon 9 (2023) e19444 Contents lists available at ScienceDirect Heliyon journal [homepage:](homepage: www.cell.com/heliyon)[ www.cell.com/heliyon](homepage: www.cell.com/heliyon) Machine learning based analysis for intellectual disability in Down syndrome Federico Baldoa, 1, Allison Piovesan b, 1, Marijana Rakvina, Giuseppe Ramacieric, Chiara Locatellid, Silvia Lanfranchie, Sara Onnivelloe, Francesca Pulinae, Maria Caracausib, Francesca Antonarosb, Michele Lombardia, **, Maria Chiara Pellerib, * a Department of Computer Science and Engineering, University of Bologna, Viale Risorgimento 2, 40136, Bologna, BO, Italy b Department of Biomedical and Neuromotor Sciences (DIBINEM), University of Bologna, Via Massarenti 9, 40138, Bologna, BO, Italy c Department of Medical and Surgical Sciences (DIMEC), University of Bologna, Via Massarenti 9, 40138, Bologna, BO, Italy d Neonatology Unit, IRCCS University General Hospital Sant’Orsola Polyclinic, Via Massarenti 9, 40138, Bologna, BO, Italy e Department of Developmental Psychology and Socialisation, University of Padova, Via Venezia 8, 35131, Padua, PD, Italy A R T I C L E I N F O Keywords: Down syndrome Intellectual disability Data mining Machine learning A B S T R A C T Down syndrome (DS) or trisomy 21 is the most common genetic cause of intellectual disability (ID), but a pathogenic mechanism has not been identified yet. Studying a complex and not monogenic condition such as DS, a clear correlation between cause and effect might be difficult to find through classical analysis methods, thus different approaches need to be used. The increased availability of big data has made the use of artificial intelligence (AI) and in particular machine learning (ML) in the medical field possible. The purpose of this work is the application of ML techniques to provide an analysis of clinical records obtained from subjects with DS and study their association with ID. We have applied two tree-based ML models (random forest and gradient boosting machine) to the research question: how to identify key features likely associated with ID in DS. We analyzed 109 features (or variables) in 106 DS subjects. The outcome of the analysis was the age equivalent (AE) score as indicator of intellectual functioning, impaired in ID. We applied several methods to configure the models: feature selection through Boruta framework to minimize random correlation; data augmentation to overcome the issue of a small dataset; age effect mitigation to take into account the chronological age of the subjects. The results show that ML algorithms can be applied with good accuracy to identify variables likely involved in cognitive impairment in DS. In particular, we show how random forest and gradient boosting machine produce results with low error (MSE <0.12) and an acceptable R2 (0.70 and 0.93). Interestingly, the ranking of the variables point to several features of interest related to hearing, gastrointestinal alterations, thyroid state, immune system and vitamin B12 that can be considered with particular attention for improving care pathways for people with DS. * Corresponding author.Department of Biomedical and Neuromotor Sciences (DIBINEM), University of Bologna, via Massarenti, 9, 40138, Bologna, BO, Italy. ** Corresponding author. Department of Computer Science and Engineering, University of Bologna, Viale Risorgimento 2, 40136, Bologna, BO, Italy. [E-mail addresses:](E-mail addresses: michele.lombardi2@unibo.it)[ michele.lombardi2@unibo.it](E-mail addresses: michele.lombardi2@unibo.it) (M. Lombardi), [mariachiara.pelleri2@unibo.it](mariachiara.pelleri2@unibo.it) (M.C. Pelleri). 1 These authors contributed equally to this work. [https://doi.org/10.1016/j.heliyon.2023.e19444](https://doi.org/10.1016/j.heliyon.2023.e19444) Received 2 February 2023; Received in revised form 19 July 2023; Accepted 23 August 2023 Available online 27 August 2023 2405-8440/© 2023 The Authors. Published

FAQs

What is the goal of the machine learning analysis in this study?

To apply ML techniques to clinical records of individuals with Down syndrome and analyze their association with intellectual disability, using age equivalent scores as the outcome indicator.

Which machine learning models are used and what inputs are analyzed?

The study uses random forest and gradient boosting machine models, analyzing 109 features across 106 DS subjects.

How do the results help interpret factors linked to cognitive impairment in Down syndrome?

The models show low prediction error and acceptable R2, while feature ranking highlights variables related to hearing, gastrointestinal alterations, thyroid status, immune system, and vitamin B12 as candidates for focused care pathway improvement.

Aug 6, 2026

Upload by Logic