The potential of evaluating shape drawing using machine learning for predicting high autistic traits

Aug 4, 2026

The study evaluates whether shape drawing can be assessed objectively with machine learning to predict children with high autistic traits. Drawing deficits are treated as an important factor for social adaptability, and movement-derived features from digital pen data are used to build predictive models. Seventy boys and 63 girls completed ordered drawing tasks, while pen kinematics and eye movements were recorded and analyzed. Support vector machine classifiers achieved accuracy, sensitivity, and specificity above 85% with high specificity, including perfect specificity in key models. Results support the approach as a promising screening tool, motivating future work with broader shape sets.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

Abstract

Background

Method

Data and Results

Conclusions

Introduction

OPEN ACCESS Citation: Ohmoto Y, Terada K, Shimizu H, Imamura A, Iwanaga R, Kumazaki H (2025) The potential of evaluating shape drawing using machine learning for predicting high autistic traits. PLoS ONE 20(4): e0320770 . [https://doi](https://doi). org/10 . 1371/journal. pone.0320770 Editor: Kenji Hashimoto, Chiba Daigaku, JAPAN Received: December 16, 2024 Accepted: February 24, 2025 Published: April 9, 2025 Copyright: © 2025 Ohmoto et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data availability statement: All relevant data are within the paper and Supporting Information files. Funding: This work was supported in part by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (22H03911 and 23H04358), JST MIRAI, and JST's CREST funding program. RESEARCH ARTICLE The potential of evaluating shape drawing using machine learning for predicting high autistic traits Yoshimasa Ohmoto1, Kazunori Terada2, Hitomi Shimizu3, Akira Imamura4, Ryoichiro Iwanaga4, Hirokazu Kumazaki3* 1 Faculty of Informatics, Department of Behavior Informatics, Shizuoka University, Shizuoka, Japan, 2 Faculty of Engineering, Department of Electrical, Electronic, and Computer Engineering, Gifu University, Gifu, Japan, 3 Department of Neuropsychiatry, Graduate School of Biomedical Sciences, Nagasaki University, Nagasaki, Japan, 4 Unit of Medical Science, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan * [kumazaki@tiara.ocn.ne.jp](kumazaki@tiara.ocn.ne.jp) Abstract Background Children with high autistic traits often exhibit deficits in drawing, an important skill for social adaptability. Machine learning is a powerful technique for learning predictive models from movement data, so drawing processes and product characteristics can be objectively evaluated. This study aimed to assess the potential of evaluating shape drawing using machine learning to predict high autistic traits. Method Seventy boys (5.03 ± 0.16) and 63 girls (5.06 ± 0.18) from the general population participated in the study. Participants were asked to draw shapes in the following order: equilateral triangle, inverted equilateral triangle, square, and the sun. A model for classifying participants as likely to have high autistic traits was developed using a support vector machine algorithm with a linear kernel utilizing 16 variables. A 16-inch liquid crystal display pen tablet was used to acquire data on hand-finger fine motor activity while the participants drew each shape. The X andY coordinates of the pen tip, pen pressure, pen orientation, pen tilt, and eye movements were recorded to determine whether the participants had any problems with this skill. Eye movements were assessed using a webcam. These data and eye movements were used to identify the variables for the support vector machine model. Data and Results For each shape, a model support vector machine was created to classify the high and low autistic trait groups, with accuracy, sensitivity, and specificity all above 85%. The specificity values across all models were 100% . In the inverted equilateral triangle model, specificity, accuracy, and sensitivity values were 100% . Competing interests: The authors declare no competing interests. Conclusions These results demonstrate the potential of assessing shape characteristics using machine learning to predict high levels of autistic traits. Future studies with a wider variety of shapes are warranted to establish further the potential efficacy of drawing skills for screening for autism spectrum conditions. Introduction Autism spectrum condition (ASC) is a developmental condition that affects social communication, social interactions, and repetitive restricted behavior. The signs and

FAQs

What does the study aim to predict using machine learning?

The study aims to predict high autistic traits using features extracted from shape drawing processes and product characteristics.

How were drawing and behavioral data collected?

Participants drew specific shapes with a liquid crystal display pen tablet, while pen tip coordinates, pressure, orientation, tilt, and eye movements were recorded using a webcam.

How accurate were the resulting predictive models?

For each shape, support vector machine models classified high versus low autistic trait groups with accuracy, sensitivity, and specificity all above 85%, and specificity reached 100% across models, including 100% in the inverted equilateral triangle model.

Aug 4, 2026

Upload by Evangeline