Efficient Monitoring of the Development and Treatment of Brain Diseases

The automatic MR image analysis development at the Faculty of Electrical Engineering of the University of Ljubljana enables very detailed monitoring of development of brain diseases and the course of their treatment.

Author: Žiga Špiclin

Brain diseases are among the leading causes of chronic disability. The development of injuries or lesions in the white-matter and/or increased atrophy of the whole brain-matter and smaller brain centres are characteristic for the majority of neurological and cerebrovascular diseases. The radiologic tomographic 3D MR imaging technique is currently the most efficient method for displaying healthy and pathological brain structures.

The main challenge when monitoring the progression of diseases and responses to treatment with MR is detecting and assesing small short-term changes of the lesions and atrophies in 3D images, which are fairly unreliable and thus at best qualitative if done by the naked eye. The automatic image analysis allows us to define these changes quantitatively as well as in a reliable and reproducible manner. This is what Žiga Špiclin and his colleagues were able to achieve.  

The paper in the journal NeuroImage presented the new method for cross-cutting volumetric analysis of multi-sequential MR images, which was later upgraded for the monitoring of longitudinal atrophy and lesion changes for the article in the journal Neuroinformatics . This method allows the MR images to acquire quantitative measurements, which are used as prediction factors in clinical environment and as alternative indicators of the course of the disease and for treatment efficacy.

Sources: A. Galimzianova, F. Pernuš, B. Likar, and Ž. Špiclin, Stratified mixture modeling for segmentation of white-matter lesions in brain MR images, Neuroimage, vol. 124, no. Pt A, p. 1031–1043, 2016.

Ž. Lesjak, F. Pernuš, B. Likar, and Ž. Špiclin, Validation of white-matter lesion change detection methods on a novel publicly available MRI image database, Neuroinformatics, vol. 14, no. 4, p. 403–420, 2016.