Hudry Kristelle, Chetcuti Lacey, Tan Diana Weiting, Clark Alena, Aulich Alexandra, Bent Catherine A, Green Cherie C, Smith Jodie, Fordyce Kathryn, Ninomiya Masaru, Saito Atsushi, Hakoshima Shuji, Whitehouse Andrew J O
Department of Psychology, Counselling and Therapy, School of Psychology and Public Health, La Trobe University, Melbourne, Australia.
Department of Psychiatry and Behavioral Sciences, Stanford University, Stanford, California, USA.
Mol Autism. 2025 Jul 10;16(1):36. doi: 10.1186/s13229-025-00670-4.
Eye-tracking could expedite autism identification/diagnosis through standardisation and objectivity. We tested whether autism assessment, with Classification Algorithm derived from gaze fixation durations, would have good accuracy (area under the curve [AUC] ≥ 0.80) to differentiate 2-4-year-old autistic from non-autistic children.
Community sampling (March 2019-March 2021) of 2:00–4:11 year-olds included children recruited into a diagnosed Autism Group (‘cases’) and Non-Autism ‘Control’ Group (with likely undiagnosed autism minimised). We recruited well beyond minimum necessary sample size to ensure within-group heterogeneity and allow exploratory subgroup analysis. Alongside eye-tracking attempted with all recruited participants, we collected parent-report measures for all children, and clinical/behavioural measures with autistic children.
102 autistic (81.4% male; = 44mths; = 8.8) and 101 non-autistic children (57.4% male; = 40; = 10.5) were recruited and eligible; the former slightly older, proportionately more male, and reflecting greater socio-demographic diversity. autism assessment was completed with 101 non-autistic children ( = 1 returning minimal data), and attempted with 100- and completed with 96 autistic children ( = 2 not attempted following adverse responses to clinical testing; = 4 attempted but unable to calibrate). The Non-Autism Group returned significantly more overall tracking data. The final Classification Algorithm (range 0-100; threshold score = 28.6)—derived from = 196 children’s fixation durations to elements of social/non-social scenes, human face presentations, and referential attention trials—had AUC = 0.82 (sensitivity = 0.82, specificity = 0.70). Compared to those correctly classified, autistic children misclassified as ‘controls’ showed greater overall tracking, and less pronounced autism features and developmental disability. Compared to correctly classified non-autistic children, those misclassified as ‘cases’ were older with lower overall tracking.
Our groups differed on socio-demographic characteristics and overall tracking (included within the Classification Algorithm). We used the ‘Scene 10A’ stimulus set as provided, without update/modification. Industry employees who developed the final Algorithm were non-blinded to child group, and considered only gaze fixation durations. Community sampling and ‘case-control’ design—comparing diagnosed autistic vs. non-autistic children—could be improved via future referral-based recruitment.
Most children tolerated autism assessment, and our Classification Algorithm properties approached those reported from other use and established clinical assessments. Independent replication is required, and research informing the most suitable clinical application of this technology.
ACTRN12619000317190
The online version contains supplementary material available at 10.1186/s13229-025-00670-4.
眼动追踪可通过标准化和客观性加快自闭症的识别/诊断。我们测试了基于注视持续时间推导的分类算法进行自闭症评估，对于区分2至4岁自闭症儿童和非自闭症儿童是否具有良好的准确性（曲线下面积[AUC]≥0.80）。
对2岁0个月至4岁11个月儿童进行社区抽样（2019年3月至2021年3月），包括招募进入已确诊自闭症组（“病例”）和非自闭症“对照”组的儿童（尽量减少可能未被诊断出自闭症的情况）。我们的招募人数远超最低必要样本量，以确保组内异质性并允许进行探索性亚组分析。除了对所有招募的参与者进行眼动追踪外，我们还收集了所有儿童的家长报告测量数据，以及自闭症儿童的临床/行为测量数据。
招募并符合条件的有102名自闭症儿童（81.4%为男性；平均年龄=44个月；标准差=8.8）和101名非自闭症儿童（57.4%为男性；平均年龄=40个月；标准差=10.5）；前者年龄稍大，男性比例更高，反映出更大的社会人口统计学多样性。对101名非自闭症儿童完成了自闭症评估（1名返回的数据极少），对100名自闭症儿童进行了评估，其中96名完成评估（2名因对临床测试有不良反应未进行评估；4名进行了尝试但无法校准）。非自闭症组返回的总体追踪数据明显更多。最终的分类算法（范围0至100；阈值分数=28.6）——基于196名儿童对社交/非社交场景元素、人脸呈现和参照性注意力试验的注视持续时间推导得出——AUC=0.82（敏感性=0.82，特异性=0.70）。与正确分类的儿童相比，被误分类为“对照”的自闭症儿童总体追踪更多，自闭症特征和发育障碍不那么明显。与正确分类的非自闭症儿童相比，被误分类为“病例”的儿童年龄更大，总体追踪更低。
我们的组在社会人口统计学特征和总体追踪方面存在差异（包含在分类算法中）。我们使用了提供的“场景10A”刺激集，未进行更新/修改。开发最终算法的行业员工对儿童分组情况不设盲，且仅考虑注视持续时间。未来通过基于转诊的招募方式，社区抽样和“病例对照”设计（比较已确诊的自闭症儿童与非自闭症儿童）可能会得到改进。
大多数儿童耐受自闭症评估，我们的分类算法特性接近其他自闭症评估及既定临床评估所报告的特性。需要进行独立复制研究，并开展相关研究以确定该技术最适合的临床应用。
ACTRN12619000317190
在线版本包含可在10.1186/s13229-025-00670-4获取的补充材料。
