| Developers: | The pennsylvanian university - University of Pennsylvania |
| Date of the premiere of the system: | October, 2020 |
| Branches: | Pharmaceutics, medicine, health care |
2020: The announcement of the application for diagnosis of a stroke
At the end of October, 2020 researchers from the Pennsylvanian university and Methodist hospital of Houston developed the application which helps doctors to diagnose a stroke quickly. Developers hoped that this application on the basis of AI will be able to accelerate diagnosis of a stroke and to lower number of heavy complications which it is possible to avoid at a timely initiation of treatment.
Fast diagnostics and early start of treatment at a stroke are extremely important for prevention of long term damage. However in certain cases doctors can not notice almost inaudible symptoms, so, treatment is postponed indefinitely. At a heavy stroke it is almost impossible to pass symptoms even at primary survey, but researches show that in in many cases with easy or moderate symptoms the diagnosis is set several watch later, and it limits treatment options.
The new application analyzes record of the speech and movements of the person of the patient using an algorithm of machine learning and within several minutes yields result on accuracy comparable to the diagnosis of the emergency doctor. This system uses AI for identification of such scarcely noticeable symptoms of a stroke as muffled pronunciation or an obvisaniye of facial muscles on the one hand. Researchers trained an algorithm at data of more than 80 patients with stroke symptoms.
The new technology provides to 79% the accuracy of diagnosis of a stroke that is similar to the accuracy of emergency doctors. As a system gives the answer in only several minutes, it can become useful addition to tool kit of any doctor of primary link. Moreover, researchers remind that if for training of a system to use larger data set, it is possible to reach a bigger precision rate.
By the end of October, 2020 medical institutions can receive the application on demand to developers.[1]
Notes
| The site content is translated by machine translation software powered by PROMT. The machine-translated articles are not always perfect and may contain errors in vocabulary, syntax or grammar. Read original article If you find inaccuracies or errors in the results of machine translation, please write to editor@tadviser.ru. We will make every effort to correct them as soon as possible. |