Cognoa Inc., Palo Alto, CA, USA.
Departments of Pediatrics, Biomedical Data Science and Psychiatry and Behavioral Sciences, Stanford University, Stanford, CA, USA.
Sci Rep. 2020 Mar 19;10(1):5014. doi: 10.1038/s41598-020-61213-w.
Autism has become a pressing healthcare challenge. The instruments used to aid diagnosis are time and labor expensive and require trained clinicians to administer, leading to long wait times for at-risk children. We present a multi-modular, machine learning-based assessment of autism comprising three complementary modules for a unified outcome of diagnostic-grade reliability: A 4-minute, parent-report questionnaire delivered via a mobile app, a list of key behaviors identified from 2-minute, semi-structured home videos of children, and a 2-minute questionnaire presented to the clinician at the time of clinical assessment. We demonstrate the assessment reliability in a blinded, multi-site clinical study on children 18-72 months of age (n = 375) in the United States. It outperforms baseline screeners administered to children by 0.35 (90% CI: 0.26 to 0.43) in AUC and 0.69 (90% CI: 0.58 to 0.81) in specificity when operating at 90% sensitivity. Compared to the baseline screeners evaluated on children less than 48 months of age, our assessment outperforms the most accurate by 0.18 (90% CI: 0.08 to 0.29 at 90%) in AUC and 0.30 (90% CI: 0.11 to 0.50) in specificity when operating at 90% sensitivity.
自闭症已成为一个紧迫的医疗保健挑战。用于辅助诊断的工具既耗时又昂贵，且需要经过培训的临床医生来操作，这导致高风险儿童需要长时间等待。我们提出了一种基于多模块、机器学习的自闭症评估方法，该方法包含三个互补模块，可获得具有诊断级可靠性的统一结果：一个 4 分钟的、家长报告的问卷，通过移动应用程序提供；从 2 分钟的、儿童半结构化家庭视频中识别出的一组关键行为列表；以及在临床评估时向临床医生提供的 2 分钟问卷。我们在美国进行的一项针对 18-72 个月大的儿童（n = 375）的盲法、多地点临床研究中证明了该评估的可靠性。与儿童基线筛查器相比，该评估在 AUC 中高出 0.35（90%CI：0.26 至 0.43），在特异性中高出 0.69（90%CI：0.58 至 0.81），当灵敏度为 90%时。与评估年龄小于 48 个月的儿童的基线筛查器相比，我们的评估在 AUC 中高出 0.18（90%CI：0.08 至 0.29，灵敏度为 90%），在特异性中高出 0.30（90%CI：0.11 至 0.50）。
