The ambulance service will use artificial intelligence to make four-week forecasts to improve patient transport capacity, after which such forecasts will be made for life-saving missions, it was announced at a press conference in Budapest on Thursday.
Gabor Ksato, Director General of the National Ambulance Service (OMSZ), emphasized that for a few decades, no one believed that the now 134-year-old ambulance service would be so advanced in innovation and would also use artificial intelligence to help as many people as possible.
He explained that they have 1.2 million cases annually, covering about 40 million km, and this number has doubled during the pandemic. For several years now, data has also been recorded digitally with a timestamp, making it immutable. Thus, the data sets are suitable for analysis, and they are assisted in this free work, i.e. free of charge by Dyntell Magyarország Kft. In Debrecen.
As mentioned
In recent years, OMSZ has been at the forefront of developments, releasing the community app HeartCity, and their Life-Saving app will also allow video communication to be created for the Fall Alert location.
These improvements are visible, but at the same time business intelligence developments have been implemented, which have helped them survive the coronavirus pandemic. For example, he mentioned that with these improvements, they can see for a minute exactly how many cases they have, how many devices, their protective equipment, and where they need to be replaced.
Pál Győrfi, a spokesperson for OMSZ, spoke about the fact that the software is used to plan what we believe to be unplanned, since the work of ambulances is to respond to unforeseen situations. He explained that last year Debrecen ranked sixth in the Deloitte Technology Fast 50 Central-Europe ranking, which presents the 50 fastest growing technology companies in the Central European region.
said Peter Salga, Managing Director of Dyntell Magyarország Kft. Two methods are used. One uses statistical methods and algorithms that are able to “learn” patterns in past data. On the other hand, correlations are looked for in other data sets, such as weather or traffic, for directions similar to ambulances.
He pointed out that this was discovered during the initial research work
Rescue missions have a specific internal pattern of time and place, as well as seasonality, that is, based on past data, future missions can be predicted to some extent.
Datasets related to the evolution of case numbers were searched, that is, the forecast takes into account not only weather forecasts, traffic or hospital data, but also the positions of the planets and the moon, for example. Based on a question from MTI, their current predictions are based on an analysis of data from the past two years, and correlations have been found with the interplanetary positions of Mars and Venus.
He also spoke about the fact that the AI technology developed by Dyntell has been performing well for years in industrial tasks where it is also used to predict human behavior as well as orders and other business data. For example, when a machine is running, it is expected to fail, and downtime can drastically reduce the downtime caused by failure. And their algorithm to help plan the sale of perishable foods, already in use in the United States, helps determine the optimal amount of food on shelves with over 90 percent accuracy.
The effectiveness of these prosthetic methods in HUF, EUR and USD, which are performed in the ambulance service, can be measured in human life.
– said Peter Salga.
From the end of July, the company plans to send out monthly and weekly forecasts to all ambulance stations in the country, not only for patient transport but also for life-saving missions.
Cover image illustration.
