심혈관계 환자 관리 개선을 위한 심박 및 투약 정보 동시 모니터링이 가능한 스마트 수액 폴대 프로토타입 개발
Development of a Smart IV Pole Prototype for Simultaneous Monitoring of Heart-beat and Drug Infusion to Improve the Safety of Patients with Cardiovascular Diseases
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
To improve the care of critically ill inpatients with an unstable cardiovascular status, it is beneficial to simultaneously monitor the occurrence of abnormal events in electrocardiogram (ECG) signal and the status of intra- venous (IV) drug administration in real-time to help the timely medical treatment and post-verification of the cause of such ECG anomalies. In this study, a novel smart IV pole that can detect the anomalies in ECG pattern and heart rate using deep learning techniques was proposed, which can also detect the identification codes of patient and in- bag medication using three pole-embedded camera modules and an on-bag QR marker whenever the ECG anomalies were detected by the deep learning model. The experimental results demonstrated that the average accuracy of the pattern classification was over 96% in the off-line test and over 85% in the on-line test. The average %error rate of the heart rate detection was less than 2% for all test conditions. The implemented smart IV pole could successfully extract the patient/drug information from the image of QR marker-attached pole-hung IV bag regardless of the size of the bag (100, 500, and 1,000 mL) and the position of the on-bag QR marker. During the long-term operation for several hours, the pole controller could detect overall abnormal ECG events and transmit appropriate alarms to the monitoring computer. Based on these results, we concluded that the proposed smart IV pole has the potential to improve the safety of critically ill patients with an unstable cardiovascular status.
Abstract
To improve the care of critically ill inpatients with an unstable cardiovascular status, it is beneficial to simultaneously monitor the occurrence of abnormal events in electrocardiogram (ECG) signal and the status of intra- venous (IV) drug administration in real-time to help the timely medical treatment and post-verification of the cause of such ECG anomalies. In this study, a novel smart IV pole that can detect the anomalies in ECG pattern and heart rate using deep learning techniques was proposed, which can also detect the identification codes of patient and in- bag medication using three pole-embedded camera modules and an on-bag QR marker whenever the ECG anomalies were detected by the deep learning model. The experimental results demonstrated that the average accuracy of the pattern classification was over 96% in the off-line test and over 85% in the on-line test. The average %error rate of the heart rate detection was less than 2% for all test conditions. The implemented smart IV pole could successfully extract the patient/drug information from the image of QR marker-attached pole-hung IV bag regardless of the size of the bag (100, 500, and 1,000 mL) and the position of the on-bag QR marker. During the long-term operation for several hours, the pole controller could detect overall abnormal ECG events and transmit appropriate alarms to the monitoring computer. Based on these results, we concluded that the proposed smart IV pole has the potential to improve the safety of critically ill patients with an unstable cardiovascular status.