Automatic Assessment of Neuropsychiatric Symptoms at Home Predictive of Mental Health and Cognitive Disorders Using Non-Intrusive Ambient Intelligence Technologies
试验速览
- 阶段
- 不适用
- 状态
- 进行中（未招募）
- 入组人数
- 25
- 试验地点
- 2
- 主要终点
- develop a personalized and clinically interpretable NPS assessment system
研究概览
简要总结
Neuropsychiatric symptoms (NPS) refer to a range of mental and emotional issues that can be observed through how patients move, perform daily tasks, and express feelings on their faces. In this study, the investigators want to find ways to accurately and unobtrusively track these symptoms in people's homes over time. Our goals are to note when these symptoms happen, predict potential problems, and gather clear data to help doctors make accurate diagnoses.
To do this, the investigators will first collect information from participants who have in-home sensors. the investigators will then use special computer programs that can recognize everyday activities and identify features that connect to scores from the Mild Behavioral Impairment Checklist (MBI-C). These scores will be compared to a questionnaire (NPIQ) filled out by caregivers or family members, along with any relevant information from doctors about the patients' symptoms. The investigators aim to see how these features can help differentiate between types of NPS, such as mood changes and agitation.
Finally, the investigators will create a dashboard for doctors that summarizes the patterns of these symptoms in patients, making it easier to monitor and manage their mental health.
详细描述
What the investigators would need from participants:
- Collect the participants' own data of daily activities from a room in participants home such as the participant's living room via a small camera the investigators provide for no less than 7 days and up to 30 days. The participant will have full control over the camera to turn on and off.
- The participant and a loved one will regularly answer short surveys asking about the participant's mood and daily functioning.
- At the end of agreed observation period, return the memory card of the participant's collected data from the camera in person or pre-paid FedEx envelope.
研究设计
- 研究类型
- Observational
- 观察模型
- Other
- 时间视角
- Other
入排标准
- 年龄范围
- 65 Years 至 —（Older Adult）
- 性别
- All
- 接受健康志愿者
- 是
入选标准
- •Age 65+ years old
- •Any age and ethnicity.
- •With or without, or at risk of mild to moderate dementia and healthy controls.
- •Availability of a study partner (informant) who has contact with the patient at least once a week (this can also be the surrogate/LAR but does not necessarily have to be)
排除标准
- •Those with severe mobility impairments
- •participants in non-home settings (skilled nursing facilities, etc.).
结局指标
主要结局
develop a personalized and clinically interpretable NPS assessment system
时间窗: 2 year
Assessing NPS relies primarily on clinician observation \& interviews with caregivers, usually within brief \& infrequent clinical visits. Precise assessment and monitoring methods would incorporate objective, patient- \& caregiver-friendly, real-time tools and provide reliable \& frequently captured data. We aim to create a system that can monitoring individual and subsyndromes of NPS; and detect early changes in NPS by automatically quantifying \& detecting relevant behavioral abnormalities based on both intra- \& inter-individual comparisons. Our approach is a video sensor, information processing, \& alert system that utilizes in-home cameras \& ambient sensors with video features that can recognize \& track behaviors securely, pre-defines domains involved in a behavior; personalizes the results by assessing the individual's behavioral norms and the home environment; and analyzes, learns, and provides explainable items from different modalities of data: time, location, video, and audio.
NPS Clinical Dashboard
时间窗: 1 year
Overall, the investigators envision the clinical dashboard as a transformative tool that delivers clear and comprehensive data on each individual. Therefore enhancing the quality of diagnosis of NPS for individuals, leading to better patient interventions and more effective long term management strategies.
次要结局
- generate Data and support for a large-scale study or clinical trial within the context of an R01.(2 years)
研究者
Ehsan Adeli
Asst. Professor
Stanford University