A Zoning Earthquake Casualty Prediction Model Based on Machine Learning

The evaluation of mortality in earthquake-stricken areas is vital for the emergency response during rescue operations. Hence, an effective and universal approach for accurately predicting the number of casualties due to an earthquake is needed. To obtain a precise casualty prediction method that can be applied to regions with different geographical environments, a spatial division method based on regional differences and a zoning casualty prediction method based on support vector regression (SVR) are proposed in this study. This study comprises three parts: (1) evaluating the importance of influential features on seismic fatality based on random forest to select indicators for the prediction model; (2) dividing the study area into different grades of risk zones with a strata fault line dataset and WorldPop population dataset; and (3) developing a zoning support vector regression model (Z-SVR) with optimal parameters that is suitable for different risk areas. We selected 30 historical earthquakes that occurred in China’s mainland from 1950 to 2017 to examine the prediction performance of Z-SVR and compared its performance with those of other widely used machine learning methods. The results show that Z-SVR outperformed the other machine learning methods and can further enhance the accuracy of casualty prediction.

Assessment of earthquake casualties and comparison of accuracy of five injury triage methods: evidence from a retrospective study

ObjectiveThe use of an injury triage method among earthquake injury patients can facilitate the reasonable allocation of resources, but the various existing injury triage methods need further confirmation. This study aims to assess the accuracy of several injury triage methods, namely, the Simple Triage and Rapid Treatment (START) technique; CareFlight Injury Triage (CareFlight); Rapid Emergency Medicine Score (REMS); Triage Revised Trauma Score (T-RTS) and Triage Early Warning Score (TEWS), based on their effects on earthquake injury patients.DesignData in the Huaxi Earthquake Casualty Database were analysed retrospectively.SettingThis study was conducted in China.ParticipantsData on 29 523 earthquake casualties were separately evaluated using the START technique, CareFlight, REMS, T-RTS and TEWS, with these being the five types of injury triage studied.Primary outcome measureThe receiver operating characteristic (ROC) curves for the five injury triages were calculated based on hospital deaths, injury severity scores greater than 15 points, and whether casualties stayed in the intensive care unit.ResultsThe ROC curve areas of the START technique, CareFlight, REMS, T-RTS and TEWS were 0.750, 0.737, 0.835, 0.736 and 0.797, respectively. Among the five injury triages, the most accurate in predicting hospital deaths was REMS, with an average area under the curve (AUC) of 0.835, with this due to the inclusion of more evaluation indicators.ConclusionAll methods had an effect on the triage of earthquake mass casualties. Among them, the REMS injury triage method had the largest AUC of the five triage methods. Except for REMS, no obvious difference was found in the effect of the other four injury triage methods.

Assessing the impact of Syrian refugees on earthquake casualty estimations in southeast Turkey

The influx of millions of Syrian refugees into Turkey has rapidly changed the population distribution along the Dead Sea Rift and East Anatolian fault zones. In contrast to other countries in the Middle East where refugees are accommodated in camp environments, the majority of displaced individuals in Turkey are integrated into local cities, towns, and villages – placing stress on urban settings and increasing potential exposure to strong earthquake shaking. Yet, displaced populations are often unaccounted for in the census based population models used in earthquake casualty estimations. Accordingly, this study constructs a refugee inclusive gridded population model and analyzes its impact on semi-empirical casualty estimations across southeast Turkey. Daytime and nighttime fatality estimates were calculated for five geographically distributed fault zones at earthquake magnitudes 5.8, 6.4, and 7.0. Total casualty estimates ranged from 28–7723 individuals, with the contribution of refugees varying from 1 %–26 % of total estimated casualties. On average, these percentages correspond to casualty underestimations of tens to hundreds of individuals. These findings communicate the necessity of incorporating refugee statistics into earthquake risk analyses in southeast Turkey and the ongoing importance of placing environmental hazards in their appropriate regional and temporal context.

Prediction of Seismic Casualties Based on Urban Land Types

A probability model based on urban land types and probability reliability was put forward to predict seismic casualties. The casualty probabilities of different urban land types were got by classifying the urban land types, combining with the historical earthquake casualty statistics and statistic residence status in the architecture of different urban land types, and then the upper value of earthquake casualties in certain probability might be estimated. Meanwhile, an example was analyzed with the proposed method. Results show the feasibility of the proposed method.