The project aims to study the feasibility of applying the synergy of state-of-the-art machine learning and radiomics methods to detect brain lesions caused by small vessel disease (SVD). Diagnosis of this disease is crucial in preventing cognitive impairment. It involves the evaluation of six brain lesions, but due to their characteristics, this project will involve the detection and quantitative evaluation of three of them. During studies, a system for microbleed detection and segmentation of leukoaraiosis and lacunes will be designed. Further, the MRI images will be analyzed for radiomic features to identify those that will allow detection of individual lesions. In the final step, the most favorable way to combine the two methods will be studied. In one case, the radiomic features will be given as an additional input to the neural network enriching the input information, which can provide a more accurate response. In the other case, fuzzy reasoning will be applied on the basis of independent information: the output of the network and the radiomic features. This approach increases the reliability of the system, and thus the probability of practical usage. For the implementation of the project, cooperation with the Medical University of Gdansk is essential - this is important for the Gdansk University of Technology due to inter-university cooperation. In addition, the project involves the use of the most recent detection methods that have not yet been used to solve this problem. Its implementation will also allow the proposal of interesting topics for Bachelor's and Master's theses.