Health Center La Vie sp. z o.o. in consortium with the Institute of Bioorganic Chemistry of the Polish Academy of Sciences – Poznan Supercomputing and Networking Center is implementing the project “Development of an innovative tool to support diagnosis and automatic description of radiological images using artificial intelligence tools”.
The aim of the project is to carry out research and development work necessary for the development of a tool to support radiological diagnostics and automate the description of radiological examination based on artificial intelligence mechanisms. As a result of the work, it will be possible to introduce product and technological innovation to the activities of the consortium leader – La Vie Health Center Ltd. From the point of view of the consortium leader, a research unit – the Institute of Bioorganic Chemistry of the Polish Academy of Sciences – Poznan Supercomputing and Networking Center, the implementation of the project will allow the acquisition of new knowledge in the area of application of AI mechanisms (in particular, based on weak annotations) in the field of medical diagnostics and publication of own research results.
The project will create a medical database in the form of a graph, which will allow annotation of radiological images using a tree of selectable options. For the purpose of collecting the learning set, a research tool will be developed – an application that will use the developed database. Based on the dedicated environment created within the project, based on MLOps tools, tests will be carried out on the effectiveness of machine learning algorithms for computed tomography and magnetic resonance images of the human chest area and magnetic resonance images of the head. The obtained results will lead to modification of the annotation method so as to obtain the optimal algorithm generating the description of the radiological image in terms of efficiency, with the optimal (as little time-consuming as possible) workload of the radiologist.
The target group that will benefit from the results of the study will be radiologists who perform examination descriptions and the results of their work will benefit patients and doctors.
The project is divided into 7 R&D tasks. On the side of the Bioorganic Chemistry of the Polish Academy of Sciences – Poznan Supercomputing and Networking Center as a research unit – a member of the consortium are tasks related to the creation of a dedicated environment for the application server, the creation of a dedicated environment based on MLOps tools for the construction, testing and implementation of machine learning models, and testing the effectiveness of machine learning algorithms for computed tomography images and magnetic resonance images of the human chest area.
The total value of the project is PLN 4,876,580.88, and the contribution from the European Funds is PLN 4,155,359.25.
#EUFunds #EuropeanFunds
