The aim of the doctoral thesis is to develop approaches, methods and algorithms that allow to create digital medical diagnostic systems. Medical image pre-processing, segmentation and analysis are the most important aspects of research. This work investigates a problem of pattern recognition, geometric and texture features analysis. Stages of human body fluids and tissues analysis are described in this work. There are many stages of this analysis: medical image pre-processing; fluids and tissues detection and localization; geometric and texture features calculation; defect or lesion detection. Knee joint anatomy, stages of knee osteoarthritis, medical image processing, magnetic resonance imaging are the most important aspects of research and are described in the theoretical part of the thesis. Six new modules have been developed and are described in the practical part. These new modules have a lot of possibilities: relaxation times calculation; human body fluids and tissues segmentation, features calculation and defect detection; medical image pre-processing and visualization; optimization of medical image processing instruction; textual analysis of patient's information. Geometric, texture and topology features of healthy and injured knee joint fluids and tissues have been defined in the experimental part of the doctoral thesis. Pre-processing methods, segmentation methods, author's pattern recognition modes and image process optimization methods have been compared in several experiments. As a result of the doctoral thesis, the new author's MRI medical image processing system has been developed. This system consists of several new modules. These modules can be useful for knee joint diagnostics and allow starting treatment earlier and therefore reducing the risk of cartilage destruction. Main results of research were reported on in 11 international conferences as well as published in 10 international publications. The work consists of introduction, 5 chapters, conclusions and 12 appendixes. It contains 219 pages, 134 figures, 45 tables and 116 references.