Drone in disaster victim identification: optimising and characterising the efficiency, and DNA profiling in drone-assisted versus conventional procedures

Post-disaster environments present significant challenges for Disaster Victim Identification (DVI) process. Searching for dead bodies caused by disasters could be a challenging and time-consuming operation, causing delays in retrieving and transporting dead bodies to the mortuary which can result in body decomposition, leading to DNA degradation and adversely impacting the DVI process. Drones have been widely deployed in disaster operations due to their multidimensional capabilities, but there has been limited research on their applications in assisting the DVI process. To address this gap, this study investigates the capabilities of drones in DVI operations by comparing the efficiency of drone-assisted versus conventional procedures in collecting and analysing the primary (e.g., DNA) and secondary identifiers (e.g., personal belongings). This study aims to determine the practical drone heights and camera lens specifications to capture images of secondary identifiers onsite, using two drone models: DJI Mavic 2 Zoom (M2Z) and DJI Matrice 300 RTK (M300). The quality of DNA samples transported by drone and analysed using RapidHIT ID (RHID) onsite was also studied. The study design of this research is a comparative cross- sectional study. The Secondary Identifier DVI Scoring Card (SIDSC) was introduced to collect secondary identifier data in drone-assisted procedures. The data was assessed between different settings, drones, observers and comparable parameters of other forms. DNA samples processed by RHID were compared to those analysed using conventional methods based on the number of alleles observed versus expected to determine the kinship and concordance rate. The time taken for these procedures were also evaluated. Results of this study have shown that 5 m height with a 48 mm lens for M2Z drone, and 50 m height with a 20x zoom for M300 drone produced optimal image quality and orientations. The application of SIDSC was found reliable in drone-assisted procedures, as demonstrated by intra-operator and inter-operator precision analyses (p>0.05). When comparing data between drone images and the external post-mortem (PM) examination, no significant differences (p>0.05) were observed. DNA samples collected onsite and transported via drone to the mobile lab for processing by RHID yielded 100% complete profiles (n=20), meanwhile, samples that were collected at the hospital mortuary produced 80% complete profiles (n=16). The kinship matching rate between the victim and their relative profiles generated by RHID is 89.77% compared to conventional method with 79.62%. RHID also had a concordance rate of 88.24% compared to the conventional procedure. Drone-assisted procedures were also reported to be generally faster than the conventional procedures conducted at the mortuaries. In conclusion, drone-assisted procedures were not only quicker in collecting and analysing primary and secondary identifiers but also produced results comparable to those of conventional procedures. Images from the M300 drone have better quality than the M2Z drone but both were considered comparable with external PM examination at the mortuary. Additionally, drone-assisted procedures enable data collection to be conducted early onsite before evidence is lost or destroyed. Lastly, the combination of drones and RHID can expedite data collection and analysis, mitigating the impact of DNA degradation due to body decomposition.