Abstract：
BACKGROUND: Depression is prevalent, chronic, and burdensome. Due to limited screening access, depression often remains undiagnosed. Artificial intelligence (AI) models based on spoken responses to interview questions may offer an effective, efficient alternative to other screening methods.
OBJECTIVE: The primary aim was to use a demographically diverse sample to validate an AI model, previously trained on human-administered interviews, on novel bot-administered interviews, and to check for algorithmic biases related to age, sex, race, and ethnicity.
METHODS: Using the Aiberry app, adults recruited via social media (N = 393) completed a brief bot-administered interview and a depression self-report form. An AI model was used to predict form scores based on interview responses alone. For all meaningful discrepancies between model inference and form score, clinicians performed a masked review to determine which one they preferred.
RESULTS: There was strong concurrent validity between the model predictions and raw self-report scores (r = 0.73, MAE = 3.3). 90 % of AI predictions either agreed with self-report or with clinical expert opinion when AI contradicted self-report. There was no differential model performance across age, sex, race, or ethnicity.
CONCLUSIONS: Limitations include access restrictions (English-speaking ability and access to smartphone or computer with broadband internet) and potential self-selection of participants more favorably predisposed toward AI technology.
CONCLUSIONS: The Aiberry model made accurate predictions of depression severity based on remotely collected spoken responses to a bot-administered interview. This study shows promising results for the use of AI as a mental health screening tool on par with self-report measures.
OBJECTIVE: The primary aim was to use a demographically diverse sample to validate an AI model, previously trained on human-administered interviews, on novel bot-administered interviews, and to check for algorithmic biases related to age, sex, race, and ethnicity.
METHODS: Using the Aiberry app, adults recruited via social media (N = 393) completed a brief bot-administered interview and a depression self-report form. An AI model was used to predict form scores based on interview responses alone. For all meaningful discrepancies between model inference and form score, clinicians performed a masked review to determine which one they preferred.
RESULTS: There was strong concurrent validity between the model predictions and raw self-report scores (r = 0.73, MAE = 3.3). 90 % of AI predictions either agreed with self-report or with clinical expert opinion when AI contradicted self-report. There was no differential model performance across age, sex, race, or ethnicity.
CONCLUSIONS: Limitations include access restrictions (English-speaking ability and access to smartphone or computer with broadband internet) and potential self-selection of participants more favorably predisposed toward AI technology.
CONCLUSIONS: The Aiberry model made accurate predictions of depression severity based on remotely collected spoken responses to a bot-administered interview. This study shows promising results for the use of AI as a mental health screening tool on par with self-report measures.
摘要：
背景：抑郁症很普遍，慢性，和负担。由于筛选通道有限，抑郁症往往无法确诊。基于对面试问题的口头回答的人工智能(AI)模型可能会提供一种有效的、有效替代其他筛选方法。
目标：主要目的是使用人口统计学上不同的样本来验证AI模型，以前接受过人类管理的采访，在新颖的机器人管理的采访中，并检查与年龄相关的算法偏差，性别,种族,和种族。
方法：使用Aiberry应用程序，通过社交媒体招募的成年人（N=393）完成了简短的机器人管理访谈和抑郁症自我报告表。使用AI模型仅根据面试回答来预测表格分数。对于模型推断和形式评分之间的所有有意义的差异，临床医生进行了一项隐蔽审查,以确定他们更喜欢哪一项.
结果：模型预测和原始自我报告得分之间存在很强的并发有效性（r=0.73，MAE=3.3）。当AI与自我报告相矛盾时，90%的AI预测要么与自我报告一致，要么与临床专家意见一致。不同年龄的模型性能没有差异，性别,种族,或种族。
结论：限制包括访问限制（讲英语的能力以及使用宽带互联网的智能手机或计算机）以及对AI技术更有利的参与者的潜在自我选择。
结论：Aiberry模型根据远程收集的对机器人管理的访谈的口头反应，对抑郁症的严重程度做出了准确的预测。这项研究显示，在使用AI作为与自我报告措施相当的心理健康筛查工具方面，有希望的结果。
目标：主要目的是使用人口统计学上不同的样本来验证AI模型，以前接受过人类管理的采访，在新颖的机器人管理的采访中，并检查与年龄相关的算法偏差，性别,种族,和种族。
方法：使用Aiberry应用程序，通过社交媒体招募的成年人（N=393）完成了简短的机器人管理访谈和抑郁症自我报告表。使用AI模型仅根据面试回答来预测表格分数。对于模型推断和形式评分之间的所有有意义的差异，临床医生进行了一项隐蔽审查,以确定他们更喜欢哪一项.
结果：模型预测和原始自我报告得分之间存在很强的并发有效性（r=0.73，MAE=3.3）。当AI与自我报告相矛盾时，90%的AI预测要么与自我报告一致，要么与临床专家意见一致。不同年龄的模型性能没有差异，性别,种族,或种族。
结论：限制包括访问限制（讲英语的能力以及使用宽带互联网的智能手机或计算机）以及对AI技术更有利的参与者的潜在自我选择。
结论：Aiberry模型根据远程收集的对机器人管理的访谈的口头反应，对抑郁症的严重程度做出了准确的预测。这项研究显示，在使用AI作为与自我报告措施相当的心理健康筛查工具方面，有希望的结果。
