Drones can be used to provide dispatch centres with on-site photos before arrival of EMS in time critical incidents.

BACKGROUND: Drones are able to deliver automated external defibrillators in cases of out-of-hospital cardiac arrest (OHCA) but can be deployed for other purposes. Our aim was to evaluate the feasibility of sending live photos to dispatch centres before arrival of other units during time-critical incidents. METHODS: In this retrospective observational study, the regional dispatch centre implemented a new service using five existing AED-drone systems covering an estimated 200000 inhabitants in Sweden. Drones were deployed automatically over a 4-month study period (December 2022-April 2023) in emergency calls involving suspected OHCAs, traffic accidents and fires in buildings. Upon arrival at the scene, an overhead photo was taken and transmitted to the dispatch centre. Feasibility of providing photos in real time, and time delays intervals were examined. RESULTS: Overall, drones were deployed in 59/440 (13%) of all emergency calls: 26/59 (44%) of suspected OHCAs, 20/59 (34%) of traffic accidents, and 13/59 (22%) of fires in buildings. The main reasons for non-deployment were closed airspace and unfavourable weather conditions (68%). Drones arrived safely at the exact location in 58/59 cases (98%). Their overall median response time was 3:49 min, (IQR 3:18-4:26) vs. emergency medical services (EMS), 05:51 (IQR: 04:29-08:04) p-value for time difference between drone and EMS = 0,05. Drones arrived first on scene in 47/52 cases (90%) and the largest median time difference was found in suspected OHCAs 4:10 min, (IQR: 02:57-05:28). The time difference in the 5/52 (10%) cases when EMS arrived first the time difference was 5:18 min (IQR 2:19-7:38), p = NA. Photos were transmitted correctly in all 59 alerts. No adverse events occurred. CONCLUSION: In a newly implemented drone dispatch service, drones were dispatched to 13% of relevant EMS calls. When drones were dispatched, they arrived at scene earlier than EMS services in 90% of cases. Drones were able to relay photos to the dispatch centre in all cases. Although severely affected by closed airspace and weather conditions, this novel method may facilitate additional decision-making information during time-critical incidents.

National coverage of out-of-hospital cardiac arrests using automated external defibrillator-equipped drones - A geographical information system analysis.

BACKGROUND: Early defibrillation is essential for increasing the chance of survival in out-of-hospital-cardiac-arrest (OHCA). Automated external defibrillator (AED)-equipped drones have a substantial potential to shorten times to defibrillation in OHCA patients. However, optimal locations for drone deployment are unknown. Our aims were to find areas of high incidence of OHCA on a national level for placement of AED-drones, and to quantify the number of drones needed to reach 50, 80, 90 and 100% of the target population within eight minutes. METHODS: This is a retrospective observational study of OHCAs reported to the Swedish Registry for Cardiopulmonary Resuscitation between 2010-2018. Spatial analyses of optimal drone placement were performed using geographical information system (GIS)-analyses covering high-incidence areas (>100 OHCAs in 2010-2018) and response times. RESULTS: 39,246 OHCAs were included. To reach all OHCAs in high-incidence areas with AEDs delivered by drone or ambulance within eight minutes, 61 drone systems would be needed, resulting in overall OHCA coverage of 58.2%, and median timesaving of 05:01 (min:sec) [IQR 03:22-06:19]. To reach 50% of the historically reported OHCAs in <8 min, 21 drone systems would be needed; for 80%, 366; for 90%, 784, and for 100%, 2408. CONCLUSIONS: At a national level, GIS-analyses can identify high incidence areas of OHCA and serve as tools to quantify the need of AED-equipped drones. Use of only a small number of drone systems can increase national coverage of OHCA substantially. Prospective real-life studies are needed to evaluate theoretically optimized suggestions for drone placement.

The use of drones and a machine-learning model for recognition of simulated drowning victims-A feasibility study.

BACKGROUND: Submersion time is a strong predictor for death in drowning, already 10 min after submersion, survival is poor. Traditional search efforts are time-consuming and demand a large number of rescuers and resources. We aim to investigate the feasibility and effectiveness of using drones combined with an online machine learning (ML) model for automated recognition of simulated drowning victims. METHODS: This feasibility study used photos taken by a drone hovering at 40 m altitude over an estimated 3000 m2 surf area with individuals simulating drowning. Photos from 2 ocean beaches in the south of Sweden were used to (a) train an online ML model (b) test the model for recognition of a drowning victim. RESULTS: The model was tested for recognition on n = 100 photos with one victim and n = 100 photos with no victims. In drone photos containing one victim (n = 100) the ML model sensitivity for drowning victim recognition was 91% (95%CI 84.9%-96.2%) with a median probability score that the finding was human of 66% (IQR 52-71). In photos with no victim (n = 100) the ML model specificity was 90% (95%CI: 83.9%-95.6%). False positives were present in 17.5% of all n = 200 photos but could all be ruled out manually as false objects. CONCLUSIONS: The use of a drone and a ML model was feasible and showed satisfying effectiveness in identifying a submerged static human simulating drowning in open water and favorable environmental conditions. The ML algorithm and methodology should be further optimized, again tested and validated in a real-life clinical study.