基于语音的人工智能检测丹麦患者的痴呆症 (DetectAI)
通过人工智能自动语音识别(detectai)的神经退行性疾病分类的非侵入性丹麦语言和语音生物标志物的开发和验证
这项观察性研究的目的是开发和测试人工智能(AI)模型,该模型可以检测痴呆症的迹象和语音记录中相关条件。 主要的问题是,基于语音的AI模型是否可以正确地告诉患有正常记忆的人,并从认知障碍的人身上思考。
该研究还将探索AI是否可以将痴呆症与抑郁症区分开,不同的痴呆症亚型,并确定哪些轻度认知障碍者(MCI)可能会发展为痴呆症。
参与者将在记录时完成短暂的记忆和语音任务。 AI模型将分析这些记录,以学习与不同诊断相关的模式。 在研究结束时,将对新参与者进行测试。
研究概览
地位
详细说明
背景痴呆症是日益增长的公共卫生挑战,早期,准确的诊断对于有效的护理和潜在的未来疾病改良治疗至关重要。 当前的诊断途径是资源密集的,并且与较长的等待时间有关。 语音反映了认知功能,最近的国际研究表明,AI可以以有希望的准确性检测语音记录中与痴呆相关的模式。 这项研究旨在在丹麦环境中开发和验证基于语音的AI模型,从而为初级保健提供一种非侵入性且可扩展的筛选工具。
阶段第一,该协议描述了我们研究的第一阶段,该阶段预计将在两个单独的阶段完成。
在第一阶段,我们试图培训AI模型,以分析具有认知障碍的参与者的语音数据,并将其与来自健康控制参与者的语音数据进行比较,这是该协议所详述的。 如果该方法已验证,我们将继续进行第二阶段。
未来的第二阶段工作我们希望进行外部验证。 AI模型分析将对参考痴呆症评估的初级保健部门的200名参与者进行。 AI分析的结果将与最终的黄金标准共识诊断进行比较。
第二阶段将具有一个单独的协议,该协议将根据第一阶段的结果来制定。
详细说明时间视角其他:混合设计。 大多数参与者将被包括在横断面病例对照研究(单语音记录)中。 对于患有MCI的参与者,将在研究期内收集随访数据,以评估对痴呆症的进展,从而评估该模型将进行性渐进性与非促进性MCI区分开的能力。
研究类型
注册 (估计的)
联系人和位置
学习联系方式
- 姓名:Sofie J Vængebjerg, MD
- 电话号码:+4530294621
- 邮箱:sova@regsj.dk
研究联系人备份
- 姓名:Peter Høgh, MD, PhD, Assoc Prof
- 电话号码:+45 22526698
- 邮箱:phh@regionsjaelland.dk
学习地点
-
-
Region Sjælland
-
Roskilde、Region Sjælland、丹麦、4000
- Zealand University Hospital
-
接触:
- Sofie J Vængebjerg, MD, PhD student
- 电话号码:+45 30294621
- 邮箱:sova@regsj.dk
-
接触:
- Peter Høgh, MD, PhD, Assoc Prof
- 邮箱:phh@regionsjaelland.dk
-
-
参与标准
资格标准
适合学习的年龄
- 成人
- 年长者
接受健康志愿者
取样方法
研究人群
描述
纳入标准:
- 年龄> 50岁
- 流利的丹麦语
- 至少需要7年的教育
对于后续队列的参与者:
- 在入学前6个月内建立的AD,VAD,LBD,FTD,MCI或抑郁症的AD,VAD,LBD,FTD,MCI或抑郁症的共识诊断
适用于健康控制队列的参与者:
- 没有已知的认知障碍或情感障碍
排除标准:
- 视力或听力大大受损(在某种程度上,参与者无法参加语言AI分析)
- 参与者无法同意
后续和新推荐队的参与者:
- MMSE分数<16
- 有多次诊断的参与者(例如。 混合痴呆或同时抑郁症的AD)
适用于新推荐队的参与者:
- 参与者属于研究中包括的六个类别(AD,VAD,LBD,FTD,MCI,抑郁症)
- 在基线很明显的参与者不会属于上述类别(可以在给出临床共识诊断之前排除)
适用于健康控制队列的参与者:
- MMSE <26和Ace <90
- GDS分数表示抑郁(6或更高)
- 可能影响认知功能的临床,实验室或神经放射学发现
学习计划
研究是如何设计的?
设计细节
队列和干预
团体/队列 |
干预/治疗 |
|---|---|
|
Cognitively Healthy Control Participants for Model A
We seek to enroll 40 age-matched cognitively healthy control participants for the training of model A.
|
参与者将在测试期间记录,以便允许AI学习和分析语音模式。
其他名称:
参与者将在测试期间记录,以便允许AI学习和分析语音模式。
其他名称:
Participants will be asked to describe the Cookie Theft Picture from the Boston Diagnostic Aphasia Examination.
The task will take 2 minutes.
Participants will be recorded during the speech task in order to allow the AI to learn and analyze the speech patterns.
Participants will be asked to recall the picture shown in the previous speech task "Picture Narrative".
This task will take 2 minutes.
Participants will be recorded during the test in order til allow the AI to learn and analyze speech patterns.
For healthy controls an MRI will be conducted to provide comparable imaging and as part of screening to ensure they do not meet exclusion criteria (neuroradiological findings that could affect cognitive functions). For patient participants, imaging will be performed as part of the standard diagnostic battery and results will be obtained from the electronic journal. Healthy control participants will undergo a standard blood test panel commonly used in dementia diagnostics. The panel includes complete blood counts, inflammatory markers, kidney- and liver function markers, thyroid-stimulating hormone (TSH), vitamine B12 and folate. These tests are performed to exclude underlying medical conditions that could mimic cognitive impairment. For patient participants, blood sampling will be performed as part of the standard diagnostic battery and results will be obtained from the electronic journal. Performed on healthy controls to rule out depression using either the geriatric depression scale (GDS) for patients > 65 year of age or the Major Depression Index (MDI) for patiens <65 year of age. For patient participants, depression screening will be performed as part of the standard diagnostic battery and results will be obtained from the electronic journal. Healthy controls will undergo a standard somatic and neurological examination to exclude conditions that may affect cognition. This includes basic neurological assessment and clinical evaluation of general health status. For patient participants, a somatic and neurological examination will be performed as part of the standard diagnostic battery and results will be obtained from the electronic journal
The participant is asked to tell a brief story based on a culturally neutral picture.
This task will take approximately 2 minutes.
Participants will be recorded during the speech task in order to allow the AI to learn and analyze the speech patterns
|
|
Patient Participants for Model A
We seek to retrospectively enroll patients from the ZUH memory clinic with a diagnosis of either Alzheimer's Disease (AD, n=50) or MCI (n=50), made within 6 months prior to enrollment.
These participants will be used for the training of model A.
|
参与者将在测试期间记录,以便允许AI学习和分析语音模式。
其他名称:
参与者将在测试期间记录,以便允许AI学习和分析语音模式。
其他名称:
Participants will be asked to describe the Cookie Theft Picture from the Boston Diagnostic Aphasia Examination.
The task will take 2 minutes.
Participants will be recorded during the speech task in order to allow the AI to learn and analyze the speech patterns.
Participants will be asked to recall the picture shown in the previous speech task "Picture Narrative".
This task will take 2 minutes.
Participants will be recorded during the test in order til allow the AI to learn and analyze speech patterns.
The participant is asked to tell a brief story based on a culturally neutral picture.
This task will take approximately 2 minutes.
Participants will be recorded during the speech task in order to allow the AI to learn and analyze the speech patterns
|
|
Patient Participants for Model B
We will prospectively recruit newly referred patients for the memory clinic at ZUH. Enrollment happens at first patient visit.
At this time, diagnosis is not yet known, but assumed present.
|
参与者将在测试期间记录,以便允许AI学习和分析语音模式。
其他名称:
参与者将在测试期间记录,以便允许AI学习和分析语音模式。
其他名称:
Participants will be asked to describe the Cookie Theft Picture from the Boston Diagnostic Aphasia Examination.
The task will take 2 minutes.
Participants will be recorded during the speech task in order to allow the AI to learn and analyze the speech patterns.
Participants will be asked to recall the picture shown in the previous speech task "Picture Narrative".
This task will take 2 minutes.
Participants will be recorded during the test in order til allow the AI to learn and analyze speech patterns.
The participant is asked to tell a brief story based on a culturally neutral picture.
This task will take approximately 2 minutes.
Participants will be recorded during the speech task in order to allow the AI to learn and analyze the speech patterns
|
研究衡量的是什么?
主要结果指标
结果测量 |
措施说明 |
大体时间 |
|---|---|---|
|
Model A: Primary measure is the AUC-ROC of the model in distinguishing between MCI and AD as well as between MCI and cognitively healthy control participants.
大体时间:At baseline (speech recording)
|
We will measure the AUR-ROC of AI predictions compared to clinical consensus diagnosis. Metrics will be presented including uncertainty estimates. Model performance will be measured on an independent test-set consisting of patients from the model B training population. |
At baseline (speech recording)
|
次要结果测量
结果测量 |
措施说明 |
大体时间 |
|---|---|---|
|
痴呆症与抑郁症的准确性
大体时间:基线(语音记录)
|
与临床共识诊断相比,使用参与者的基线语音记录,通过敏感性,特异性,AU预测的AUR AUR-ROC衡量。
在研究完成时,将在数据库锁定后测量模型性能。
|
基线(语音记录)
|
|
轻度认知障碍(MCI)的亚分类为渐进式与非促进性
大体时间:在基线(语音记录)和入学后最多12个月(确定进展)
|
与临床共识诊断相比,使用参与者的基线语音记录,通过敏感性,特异性,AU预测的AUR AUR-ROC衡量。 在研究完成时,将在数据库锁定后测量模型性能。 进展定义为在研究期间的新痴呆症诊断。 |
在基线(语音记录)和入学后最多12个月(确定进展)
|
|
痴呆亚型的分类(AD,VAD,LBD,FTD)
大体时间:基线(语音记录)
|
与临床共识诊断相比,使用参与者的基线语音记录,通过敏感性,特异性,AU预测的AUR AUR-ROC衡量。
在研究完成时,将在数据库锁定后测量模型性能。
|
基线(语音记录)
|
|
与已建立的生物标志物进行比较
大体时间:基线或生物标志物测试时如果在基线之后进行
|
AI预测与痴呆诊断的最新生物标志物之间的诊断准确性差异
|
基线或生物标志物测试时如果在基线之后进行
|
|
特征重要性分析
大体时间:基线(语音记录)
|
特征重要性将使用可解释性分析进行评估(例如
置换重要性,特征组的塑形值和/或消融),以量化声学和语言特征对模型预测的贡献。
|
基线(语音记录)
|
其他结果措施
结果测量 |
措施说明 |
大体时间 |
|---|---|---|
|
最佳准确性所需的任务数量
大体时间:基线(语音记录)
|
评估减少的语音任务是否提供了与完整测试电池相当的准确性。
|
基线(语音记录)
|
|
单个语音任务对AI模型性能的贡献
大体时间:基线(语音记录)
|
将通过比较模型性能(例如,
在语音任务子集(记忆测试、故事回忆、图片描述)上进行训练和测试时,准确度、灵敏度、特异性、AUC-ROC)。
这将确定哪些任务提供最强的诊断信号。
|
基线(语音记录)
|
合作者和调查者
调查人员
- 首席研究员:Peter Høgh, MD, PhD, Assoc Prof、Zealand University Hospital
出版物和有用的链接
一般刊物
- Nicholas LE, Brookshire RH. A system for quantifying the informativeness and efficiency of the connected speech of adults with aphasia. J Speech Hear Res. 1993 Apr;36(2):338-50. doi: 10.1044/jshr.3602.338.
- Buderer NM. Statistical methodology: I. Incorporating the prevalence of disease into the sample size calculation for sensitivity and specificity. Acad Emerg Med. 1996 Sep;3(9):895-900. doi: 10.1111/j.1553-2712.1996.tb03538.x.
- Dargaud L, Partal A, Birn A, & Detlefsen S. N. (2023). Developing a Spontaneous Speech-based Artificial Intelligence for Alzheimer's Disease Detection. Transatlantic Telehealth Research Network (TTRN) International Scientific Conference 2023, Journal of the International Society for Telemedicine and eHealth.
- Lanzi AM, Saylor AK, Fromm D, Liu H, MacWhinney B, Cohen ML. DementiaBank: Theoretical Rationale, Protocol, and Illustrative Analyses. Am J Speech Lang Pathol. 2023 Mar 9;32(2):426-438. doi: 10.1044/2022_AJSLP-22-00281. Epub 2023 Feb 15.
- Li J, Song K, Zheng B, Li D, Wu X, Meng H. Leveraging Pretrained Representations with Task-related Keywords for Alzheimer's Disease Detection. arXiv preprint. 2023.
- Luz S, Haider F, de la Fuente Garcia S, Fromm D, MacWhinney B. Detecting cognitive decline using speech only: The ADReSSo challenge. arXiv preprint 2021.
- Luz S, Haider F, Fromm D, Lazarou I, Kompatsiaris I, Macwhinney B. An Overview of the ADReSS-M Signal Processing Grand Challenge on Multilingual Alzheimer's Dementia Recognition Through Spontaneous Speech. IEEE Open J Signal Process. 2024;5:738-749. doi: 10.1109/ojsp.2024.3378595. Epub 2024 Mar 18.
- Bex T. Comprehensive Guide to Multiclass Classification With Sklearn. Towards Data Science. 2021.
- Chen J, Ye J, Tang F, Zhou J. Automatic Detection of Alzheimer's Disease Using Spontaneous Speech Only. Interspeech. 2021 Aug-Sep;2021:3830-3834. doi: 10.21437/interspeech.2021-2002.
- Agbavor F, Liang H. Predicting dementia from spontaneous speech using large language models. PLOS Digit Health. 2022 Dec 22;1(12):e0000168. doi: 10.1371/journal.pdig.0000168. eCollection 2022 Dec.
- de la Fuente Garcia S, Ritchie CW, Luz S. Artificial Intelligence, Speech, and Language Processing Approaches to Monitoring Alzheimer's Disease: A Systematic Review. J Alzheimers Dis. 2020;78(4):1547-1574. doi: 10.3233/JAD-200888.
研究记录日期
研究主要日期
学习开始 (估计的)
初级完成 (估计的)
研究完成 (估计的)
研究注册日期
首次提交
首先提交符合 QC 标准的
首次发布 (实际的)
研究记录更新
最后更新发布 (实际的)
上次提交的符合 QC 标准的更新
最后验证
更多信息
与本研究相关的术语
关键字
其他相关的 MeSH 术语
其他研究编号
- SJ-1107
计划个人参与者数据 (IPD)
计划共享个人参与者数据 (IPD)?
药物和器械信息、研究文件
研究美国 FDA 监管的药品
研究美国 FDA 监管的设备产品
此信息直接从 clinicaltrials.gov 网站检索,没有任何更改。如果您有任何更改、删除或更新研究详细信息的请求,请联系 register@clinicaltrials.gov. clinicaltrials.gov 上实施更改,我们的网站上也会自动更新.