Application of intelligent early warning system in pre- and in-hospital connection of stroke patients
10.16781/j.0258-879x.2018.09.0970
- Author:
Hong-Jian ZHANG
1
Author Information
1. Stroke Center, Changhai Hospital, Navy Medical University (Second Military Medical University)
- Publication Type:Journal Article
- Keywords:
Artificial intelligence;
Informatization;
Pre-hospital rescue;
Stroke;
Warning system
- From:
Academic Journal of Second Military Medical University
2018;39(9):970-976
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the value of intelligent early warning system in pre- and in-hospital connection in the treatment of stroke. Methods The clinical data of the patients with suspected stroke, who received treatment and prehospital warning with intelligent early warning system in Stroke Center of Changhai Hospital of Navy Medical University (Second Military Medical University) between Nov. 2017 and Jun. 2018, were retrospectively analyzed. Pre-hospital warning was defined as 120 alerting stroke teams using software before transferring patients to the target hospital. According to the in-hospital treatment methods, the patients were divided into non-stroke group, conservative treatment group, thrombolysis group, thrombectomy group, bridging group and cerebral hemorrhage group. The efficiency in each link of the intelligent early warning system, the ability of emergency doctor diagnosing stroke, the ability of emergency doctor diagnosing severe stroke using gaze-face arm speech time (G-FAST) scale, the speed of first aid response of the attending physician and the door-in-to-door-out (DIDO) time in primary stroke center were evaluated in each group. Results A total of 248 patients were included in this study, including 24 non-stroke patients and 224 stroke patients (101 patients in the conservative treatment group, 23 patients in the thrombolysis group, 32 patients in the thrombectomy group, 22 patients in the bridging group and 46 patients in the cerebral hemorrhage group). The median primary screening time, early warning time and response time of 248 patients were 28.0 (13.0, 92.5), 11.0 (7.3, 15.3) and 19.0 (13.0, 35.0) s, respectively. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy of 6 pre-hospital emergency physicians diagnosing severe stroke using G-FAST scale were 84%, 71%, 83%, 72% and 79%, respectively. Eight patients with severe stroke underwent whole-process monitoring of pre-hospital treatment, and had a median transport time of 113 (82, 142) min and a median DIDO time of 84 (12, 125) min. Conclusion Intelligent early warning system can realize the automation of data acquisition for transfer time points, making the data more comprehensive and reliable. It is helpful to analyze the efficiency of emergency physicians, drivers, stroke teams and primary stroke center in pre- and in-hospital connection in the treatment of stroke, so as to shorten the pre-hospital treatment time and improve the pre-hospital treatment efficiency.
