[论文解读] The Intelligent ICU Pilot Study: Using Artificial Intelligence Technology for Autonomous Patient Monitoring

Anis Davoudi, Kumar Rohit Malhotra+9 more|arXiv (Cornell University)|Apr 25, 2018

Intensive Care Unit Cognitive Disorders参考文献 58被引用 12

一句话总结

这项试点研究证明了使用人工智能驱动的普适传感技术——结合可穿戴设备、环境传感器和基于视觉的传感器与深度学习——对重症监护病房（ICU）患者及其环境进行自主、精细化监测的可行性。该系统在面部识别（mAP=0.80）、姿势检测（F1=0.94）方面表现出高精度，并识别出谵妄患者与非谵妄患者在面部表情、活动特征以及环境因素（光照、声音、探访）方面存在显著差异（p<0.05）。

ABSTRACT

Currently, many critical care indices are repetitively assessed and recorded by overburdened nurses, e.g. physical function or facial pain expressions of nonverbal patients. In addition, many essential information on patients and their environment are not captured at all, or are captured in a non-granular manner, e.g. sleep disturbance factors such as bright light, loud background noise, or excessive visitations. In this pilot study, we examined the feasibility of using pervasive sensing technology and artificial intelligence for autonomous and granular monitoring of critically ill patients and their environment in the Intensive Care Unit (ICU). As an exemplar prevalent condition, we also characterized delirious and non-delirious patients and their environment. We used wearable sensors, light and sound sensors, and a high-resolution camera to collected data on patients and their environment. We analyzed collected data using deep learning and statistical analysis. Our system performed face detection, face recognition, facial action unit detection, head pose detection, facial expression recognition, posture recognition, actigraphy analysis, sound pressure and light level detection, and visitation frequency detection. We were able to detect patient's face (Mean average precision (mAP)=0.94), recognize patient's face (mAP=0.80), and their postures (F1=0.94). We also found that all facial expressions, 11 activity features, visitation frequency during the day, visitation frequency during the night, light levels, and sound pressure levels during the night were significantly different between delirious and non-delirious patients (p-value&lt;0.05). In summary, we showed that granular and autonomous monitoring of critically ill patients and their environment is feasible and can be used for characterizing critical care conditions and related environment factors.

研究动机与目标

• 评估使用普适传感与人工智能实现对重症监护病房（ICU）患者自主化、精细化监测的可行性。
• 识别非言语ICU患者中与谵妄相关的细微行为与环境指标。
• 通过自动化患者状态与环境条件的重复性评估，减轻护理人员的工作负担。
• 开发一种多模态传感系统，能够持续、实时监测患者生理、行为及ICU环境。
• 探索人工智能通过整合分析面部表情、姿势、声音、光照及探访模式，检测谵妄早期迹象的潜力。

提出的方法

• 在ICU中部署包含可穿戴传感器、高分辨率摄像头、光照传感器和声压级（SPL）传感器在内的多模态传感器套件。
• 通过非侵入性、普适传感方式，长期连续采集患者及其环境的数据。
• 应用深度学习模型进行人脸检测（mAP=0.94）、人脸识别（mAP=0.80）、面部动作单元检测及头部姿态估计。
• 结合统计分析与体动记录法进行姿势识别与活动模式检测（F1=0.94）。
• 利用传感器数据量化环境因素，包括声压级、环境光照强度及白天/夜间探访频率。
• 整合多模态数据流，并应用分类技术比较谵妄与非谵妄患者的特征差异。

实验结果

研究问题

• RQ1 人工智能驱动的自主监测系统能否在ICU环境中准确检测并追踪患者的面部表情与姿势？
• RQ2 谵妄患者与非谵妄患者在环境因素（光照、声音、探访）方面是否存在可测量的差异？
• RQ3 结合深度学习的多模态传感器数据能否识别出非言语患者中谵妄的早期行为标志？
• RQ4 自动化系统在多大程度上可减轻ICU护士手动监测任务的负担？
• RQ5 使用普适传感与人工智能实现的患者行为与ICU环境的连续、实时监测，在粒度与可靠性方面如何？

主要发现

• 系统在人脸检测方面表现优异，平均精度（mAP）达到0.94。
• 患者面部识别的mAP达到0.80，表明识别精度较高。
• 姿势识别的F1得分达到0.94，表明活动监测具有高度可靠性。
• 所有面部表情、11项活动特征、白天探访频率、夜间探访频率、夜间声压级及夜间光照强度在谵妄与非谵妄患者之间均存在显著差异（p<0.05）。
• 研究证实，谵妄患者在夜间噪音和光照暴露等环境因素方面显著升高。
• 通过人工智能整合行为与环境数据，可检测出与谵妄相关的细微但具有临床意义的模式。