Accurate monitoring of meat composition at all production stages is one of the top priorities of the agri-food industry. In medicine, early detection of biochemical changes in skeletal muscles could greatly improve the diagnosis and treatment of various myopathies. At present, there is no non-invasive method to determine physiological and pathophysiological changes in the structure and chemical composition of skeletal muscles in live animals. There is a great deal of evidence to suggest that ultrasonographic characteristics of skeletal muscles (i.e., echotexture or echointensity) can be used to determine their chemical composition. Therefore, the main objective of the experiments described in this thesis was to investigate whether quantitative (computer-assisted) image analysis of muscle ultrasonograms is an appropriate method to estimate the content of their chemical/biochemical constituents. We have designed and written code called r-Algo to identify the specific ranges of echointensity values with the strongest correlation to chemical/biochemical constituents of ultrasonographically examined muscles. In the first set of studies, ultrasonograms of the pectoralis major muscle of broiler chickens were recorded just before slaughter. Standard laboratory procedures were used to analyze the chemical composition of muscle samples, including fat, protein, and moisture content as well as fatty acid profile. By detecting a specific range of pixel intensities (corresponding to each chemical ingredient), the r-Algo-assisted analyses yielded significant correlations between mean numerical pixel values and all chemical constituents studied. All software-estimated values were compared with those obtained using approved laboratory standards, and all accuracies/percent errors for estimated values were found satisfactory. In the second set of studies, we used a previously validated rat model of neurogenic inflammation (lumbar facet injury). Ultrasonograms of neurosegmentally linked muscles were obtained from both surgery and sham-operated groups of animals just before surgical procedures and then weekly until 21 days after experimental interventions. Mean echointensity values for r-Algo-detected pixel ranges associated with the expression of affected inflammatory regulators differed significantly between the two groups of animals. The present results indicate that the use of r-Algo provides a novel and effective method of quantitative image analysis, with multiple potential applications in biomedical research, medicine and industry.