The presence of leukoaraiosis or white matter lesions (WML) in the brain of elderly individuals is linked to increased risk of stroke, cognitive impairment, dementia and ultimately, death. Magnetic resonance imaging (MRI) is by far the most sensitive modality for detecting WMLs and MRI is consequently a very central diagnostic procedure in the elderly population. In this project they will contribute to the development of an automated WML quantification and analysis tool using data from large on-going prospective MR studies in patients with different neurodegenerative diseases. The algorithms developed as part of the project are also likely to have application for other types of focal brain lesions like multiple sclerosis and brain tumors. The project management, located at the Intervention Centre at Oslo University Hospital has a strong and proven track record in MR physics, image processing and large clinical trial execution. The combined expertise of both the Norwegian and Spanish groups will provide a unique setting for developing, testing and implementing large-scale WML analysis in clinically relevant patient groups.
Summary of project results
The presence of leukaraisosis or white matter lesions (WML) in the brain of elderly individuals is linked to increased risk of stroke, cognitive impairment, dementia and ultimately, death. Magnetic resonance imaging (MRI) is by far the most sensitive modality for detecting WMLs and MRI is consequently a very central diagnostic procedure in the elderly population. Manual WML segmentation is very time-consuming and prone to user-bias, which has resulted in several attempts to generating automated analysis tools for WML segmentation. The overall aim of the project was to provide automated assessment, diagnosis, monitoring and prognosis of neurodegenerative diseases and other focal brain disorders. In this project partners developed an automated WML quantification and analysis tool using expert's knowledge and available MRI data from large on-going prospective MR studies in patients with different neurodegenerative diseases. The tool is developed using novel object based machine learning algorithms. Preliminary results, validated against radiological gold-standard, indicate that the tool outperforms other available WML segmentation tools currently available. The algorithms developed as part of the project have also application for other types of focal brain lesions like multiple sclerosis and brain tumours, which can be considered and described as amorphous objects. The tool partners are developing is called AMOS, which comes from "Amorphous Object Segmentation". Partners have prepared at least 9 publications on the tool and its results that are published, submitted or in preparation, at the closure of the NILS funding. Partners have contacted various Norwegian and Spanish institutions that are interested in using the segmentation system in their studies. The list includes Hospital Clínico (Madrid), Hospital de Alcorcón (Madrid), Laboratorio de Análisis de Imagen Médica y Biometría (University Rey Juan Carlos, Madrid), Department of Psychology (University of Oslo), Institute of clinical medicine (University of Oslo), Faculty of Medicine (University of Castilla La Mancha). Companies interested in the system include Deimos Space (Madrid, Spain) and NordicNeuroLab AS (Bergen). They have also established contacts with other institutions such as the School of Clinical Medicine (Canberra Hospital, Canada) and the Clinical Ophtalmology and Eye Health Central Clinical School (at the University of Sidney, Australia).
Summary of bilateral results
The development of the integrated semantic framework for multicentre neurological multidimensional data analysis opens the door to further collaboration due to the complementary expertise of partners. The automatic analysis of clinical data sets can be extended to other kind of pathologies, not only neurological, but to any clinical study that uses imagine technology. The automatic segmentation of amorphous objects is also very attractive for running large-scale studies because it reduces significantly costs by simplifying delineation of objects of interest, which is very time consuming task. Around these two axes partners are establishing contacts with other research groups and institutions to collaborate on joint projects as, in the present era of “big data”, the prevailing trend is to design interesting multidisciplinary and multidimensional studies taking advantage of the large amount of data stored in digital form. Partners have foreseen various cooperation opportunities both at the academic and at the research level: - Cooperation between doctorate programs combining medical imaging for data generation and intelligent systems for automatic data analysis. - Exchange of master students through the European Erasmus program. - Participate as partners in international projects on development and analysis of clinically relevant large-scale studies that include imaging data. Within this frame, partners are preparing a project proposal to apply for funding to national and international research call and they are in contact with private companies and research groups interested in large-scale neuroimage studies that include analysis of amorphous focal brain lesions. They plan to establish a comprehensive collaboration for use of the integrated semantic framework for neurological multicentre and multidimensional data analysis.