Kyoto Tachibana University and Hitachi, Ltd. have started joint research on a system that uses AI to support the determination of the urgency of the injured and sick for dispatchers who receive the 119 call.
The number of emergency dispatches is increasing due to the aging of society. There is concern that the lifesaving rate will decline, as it will take time from the 119 call to arrival at the site and arrival at the hospital, and the arrival time at the site will be delayed.Therefore, the Fire and Disaster Management Agency appropriately judges the necessity of emergency dispatch by calling 119, and recommends the use of "Emergency Security Center (# 7119)" nationwide depending on the degree of urgency.However, it is not easy to identify cases with low urgency and predict the name of the disease, so it cannot be said that it has fully penetrated.
Therefore, Kyoto Tachibana University and Hitachi, Ltd. cooperated with the Toyonaka City Fire Department to utilize the contents of the hearing when calling 119. We will build a prediction model using AI and work on the study and development of a system that predicts the degree of urgency and disease name in real time from the contents of hearing at the time of reporting such as age, gender, symptoms, and medical history.
For many years, Kyoto Tachibana University has been conducting research on improving the accuracy of urgency determination protocols in the fields of correspondence command education and emergency lifesaving.This time, we will collect data not only from urban areas but also from rural areas.Hitachi will utilize the independently developed "AI that can explain the grounds" (Explainable AI) to support the judgment of the communication commander while presenting the prediction grounds.
After the system is developed, we will jointly verify the accuracy and usefulness of the system by comparing the judgment contents by AI with the diagnosis result of the delivery destination.In addition, we plan to continuously improve the accuracy of the system by letting AI learn the urgency and disease name predicted by the system and the judgment contents such as the diagnosis at the emergency services and the hospital to which the patient is transported.
