[an error occurred while processing the directive]
RSS
Логотип
Баннер в шапке 1
Баннер в шапке 2

Medical Research Institute La Fe

Company

width=200px

Content

History

2017: Creation of software with AI for recognition of a disease according to x-rays

Within the congress of the European society of radiologists (ECR-2017) which passed in Vienna from March 1 to March 5 was the software using artificial intelligence for primary detection of diseases according to x-rays is submitted.

Development was created by forces of Medical Research Institute La Fe (Valencia, Spain) and software developer of Quibim under control of organization. It is about the computer system of detection (Computed-aided detection, CAD) constructed on the basis of convolution neural networks. The algorithm is capable to distinguish diseases according to the roentgenogram of a thorax with a high accuracy, considering sensitivity and specificity of an organism.

Example of use of artificial intelligence for determination of diseases according to x-rays

The created system will help radiodiagnosis specialists to perform initial inspection of the patient in the automatic mode and to pay attention to those zones in which with high probability there can be deviations.

File:Aquote1.png
Our research improves or is at least equivalent to the results achieved in the previous research works — Belen Fos-Guarinos, the student trainee of department of researches of biomedical visualization in Medical Research Institute La Fe noted. She presented new technology at a scientific session within ECR 2017.[1]
File:Aquote2.png

It should be noted that the ideas of use of artificial intelligence for processing of medical images were voiced repeatedly. So, at National Institute of health care in Bethesda (the State of Maryland, the USA) in March, 2017 announced the development of the system of deep training also capable to detect diseases according to x-rays of a thorax. Specialists used model of parallel programming Nvidia CUDA and a possibility of graphic processors for training of the neural networks helping to define a disease, its specifics, weight and bodies struck with it.

Notes