This Small Business Innovation Research Phase I project is will investigate the feasibility of developing an objective agitation and sedation assessment algorithm for patients in the intensive care unit (ICU). Use of machine learning to identify clinically relevant information to characterize the agitation and sedation state of the patients is investigated. Furthermore, a plan to ?calibrate? and validate the agitation and sedation score provided by the algorithm will be designed. Agitation and sedation assessment is a challenging problem for patients undergoing critical care. Agitation, which is primarily characterized by excessive gross motor movement, is experienced by 74% of adults during ICU stay. Agitated patients may do physical harm to themselves by dislodging vital life support and monitoring devices with excessive musculoskeletal activity. Oversedation increases risk to the patient since liberation from mechanical ventilation may not be possible due to a diminished level of consciousness and respiratory depression from sedative. Currently, the assessment is performed by the clinical staff and no technology exists for such assessment. It is anticipated that through this research, novel algorithms for reliable detection of a patient?s agitation and sedation state using their physiological signals will be developed.
The broader impact/commercial potential of this project includes reduction in clinical staff workload and healthcare costs. Current clinical practice in patient critical care requires the nursing staff to assess the patient's agitation and sedation state and provide sedatives to ameliorate the patient's agitation. The process relies on subjective assessments and may result in oversedation, which in turn increases the number of interventions, length of mechanical ventilation, and duration of stay in the ICU, and hence, increases healthcare costs. Development of an objective agitation and sedation assessment system can have a great impact on the quality of care in a critical care setting. Such a system can enable continuous patient monitoring and increase quality of care. Currently, clinical staff need to attend to multiple patients and continuous monitoring of patients is not feasible. In addition, in the absence of an automated agitation and sedation assessment algorithm, early indications of undersedation or oversedation can be overlooked due to the complex nature of the patient critical care problem and clinical staff's workload.
Patients in intensive care units who require mechanical ventilation also frequently require the administration of sedative agents. Currently, no technology exists for agitation and sedation assessment. The current clinical standard in the ICU for assessing the level of agitation and sedation is based on subjective assessments by the clinical staff. Agitated patients are at risk of accidental device removal, which amounts to an annual cost of $400 million in the US per year. The primary consequence of oversedation is the prolongation of mechanical ventilation and ICU stay. Considering the fact that each year 5 million patients are admitted to the ICU in the US, oversedation is potentially present in 40-60% of mechanically ventilated patients (approximately 40% of sedated patients), and the average cost of ICU with mechanical ventilation is $4,000 to $5,500 per day reducing one day of mechanical ventilation in oversedated patients can result in a $4 billion annual savings in the US alone. In this Phase I NSF SBIR project, we investigated and established the feasibility of developing a multi-modality agitation and sedation assessment system predicated on machine learning techniques for patients in the intensive care unit (ICU). Furthermore, we designed a protocol for validation of the agitation and sedation assessment system. The emphasis of the approach was on an agitation and sedation assessment system which can be utilized for monitoring sedated patients in the ICU. The proposed agitation and sedation assessment system will assist in realizing an improved distribution of ICU workload by allowing the medical staff to focus on patients in need of immediate attention. It is expected that the proposed technology will create jobs in the healthcare information technology and medical device industries through the company and potential partners and suppliers. Finally, it is expected that this technology will also improve the quality of care in military intensive care units. Combat critical care facilities have been experiencing an increase in the number of traumatically injured soldiers as a result of the war on terror. Combat critical care will benefit from reduced workload and improved quality of care given the severe limitations of the available resources and the disparate nature of the battlefield environment.
