Multimodal Artificial Intelligence for Detecting the Psychological State of Cancer Patients
试验速览
- 阶段
- 不适用
- 状态
- Enrolling By Invitation
- 发起方
- 入组人数
- 1,500
- 试验地点
- 1
- 主要终点
- Area Under the Receiver Operating Characteristic Curve (AUC) of the Multimodal Machine Learning Model for Anxiety and Depression Screening
研究概览
简要总结
Artificial intelligence (AI) technology is expected to assist clinical doctors in promptly identifying cancer patients at risk of developing psychological issues and to develop preemptive management plans, thereby enhancing their quality of life. Computer vision technology can directly capture and extract subtle changes in skin color from facial images in videos, assess heart rate using signal processing algorithms, and also extract facial expressions to evaluate psychological conditions through facial expression change signal processing algorithms. The accuracy rate can exceed 88%. By leveraging the capabilities of computer vision technology, it can accurately capture subtle movements and expressions of the human body, thereby understanding the internal psychological state and obtaining relevant psychological information
研究设计
- 研究类型
- Observational
- 观察模型
- Cohort
- 时间视角
- Prospective
入排标准
- 年龄范围
- 18 Years 至 —（Adult, Older Adult）
- 性别
- All
- 接受健康志愿者
- 否
入选标准
- •Age ≥18 years old
- •Patients with tumors diagnosed by magnetic resonance imaging or contrast-enhanced ultrasound
- •The patient has self-awareness and is able to cooperate with the research
- •Patients who voluntarily undergo psychological assessment tests
排除标准
- •Age ≤18 years old
- •Not diagnosed as a tumor patient
- •Lack of autonomy and inability to conduct cooperative research
结局指标
主要结局
Area Under the Receiver Operating Characteristic Curve (AUC) of the Multimodal Machine Learning Model for Anxiety and Depression Screening
时间窗: Data collected at two time points: 1 day pre-operatively and at ≤7 days post-operatively or at discharge, whichever came first
The AUC quantifies the overall discriminative ability of the final multimodal machine learning model to distinguish between patients with positive vs. negative anxiety/depression status. The AUC will be calculated on an independent test set that is strictly separated from the training and validation sets and will not be used in any model training or hyperparameter tuning.
次要结局
未报告次要终点
研究者
Ping Liang
Principal Investigator, Chief Physician
Chinese PLA General Hospital