Objective: To address the challenge of assessing sedation status in critically ill patients in the intensive care unit (ICU), we aimed to develop a non-contact automatic classifier of agitation using artificial intelligence and deep learning. Methods: We collected the video recordings of ICU patients and cut them into 30-second (30-s) and 2-second (2-s) segments. All of the segments were annotated with the status of agitation as “Attention” and “Non-attention”. After transforming the video segments into movement quantification, we constructed the models of agitation classifiers with Threshold, Random Forest, and LSTM and evaluated their performances. Results: The video recording segmentation yielded 427 30-s and 6405 2-s segments from 61 patients for model construction. The LSTM model achieved remarkable accuracy (ACC 0.92, AUC 0.91), outperforming other methods. Conclusion: Our study proposes an advanced monitoring system combining LSTM and image processing to ensure mild patient sedation in ICU care. LSTM proves to be the optimal choice for accurate monitoring. Future efforts should prioritize expanding data collection and enhancing system integration for practical application.

Background: Agitation and sedation management is critical in intensive care as it affects patient safety. Traditional nursing assessments suffer from low frequency and subjectivity. Automating these assessments can boost intensive care unit (ICU) efficiency, treatment capacity, and patient safety. Objectives: The aim of this study was to develop a machine-learning based assessment of agitation and sedation. Methods: Using data from the Taichung Veterans General Hospital ICU database (2020), an ensemble learning model was developed for classifying the levels of agitation and sedation. Different ensemble learning model sequences were compared. In addition, an interpretable artificial intelligence approach, SHAP (Shapley additive explanations), was employed for explanatory analysis. Results: With 20 features and 121,303 data points, the random forest model achieved high area under the curve values across all models (sedation classification: 0.97; agitation classification: 0.88). The ensemble learning model enhanced agitation sensitivity (0.82) while maintaining high AUC values across all categories (all >0.82). The model explanations aligned with clinical experience. Conclusions: This study proposes an ICU agitation-sedation assessment automation using machine learning, enhancing efficiency and safety. Ensemble learning improves agitation sensitivity while maintaining accuracy. Real-time monitoring and future digital integration have the potential for advancements in intensive care.

Intensive care units (ICUs) are a crucial part of the health-care system. Patients in ICUs may experience agitation due to delirium, treatment discomfort, and environment-related reasons. This may lead to various clinical safety problems, such as the self-removal of catheters and assault of nursing personnel by patients. Hospitals in Taiwan generally use the Richmond Agitation–Sedation Scale (RASS) to assess patient agitation and adjust sedative dosages. However, the RASS has certain limitations, such as the requirement of subjective assessment, differences in patient evaluation standards among medical personnel, and low frequency of assessment. These limitations may result in the insufficient or excessive sedation of patients. Therefore, in this study, we developed an ensemble learning model by combining two machine learning models. The proposed model classified patients into oversedation, maintain range, and agitation categories. The proposed classification system may effectively assist medical personnel in performing RASS assessment, mitigate patient agitation–related clinical safety risks in ICUs, and increase the medical capacity of ICUs.

Background: Intensive care units (ICUs) are crucial in healthcare, but internal factors, like patient agitation due to delirium, treatment discomfort, and the ICU environment, can compromise care and lead to safety issues. In Taiwan, the Richmond Agitation–Sedation Scale (RASS) is used for agitation assessment, but it has limitations, including subjectivity and infrequent assessments. Methods: To enhance ICU care, we developed a machine learning-based patient agitation and sedation assessment tool. We used an ensemble learning model, combining two machine learning models to classify patients into three categories: oversedation, optimal sedation, and agitation. Results: The RandomForest model achieved the highest average accuracy for sedation (ACC = 0.92, AUC = 0.97), while addressing class imbalance increased agitation classification accuracy (ACC = 0.77, AUC = 0.88). The model's results, based on key features identified, can guide sedative dosage adjustments, enabling more precise patient care. Conclusions: Our study demonstrated the effectiveness of machine learning in classifying patient agitation and sedation. We recommend incorporating image-based features in patient agitation assessment. Our classification system can assist medical professionals in RASS assessments, mitigating safety risks related to patient agitation in ICUs and improving overall ICU capacity. Trial Registration: We obtained access to the critical care database (AI-111010) of the AI Center of Taichung Veterans General Hospital (TCVGH) from the Institutional Review Board (approval number: CE22484A), retrospectively registered.