University of Thessaly (UTH)
Ανάπτυξη συστημάτων τεχνητής νοημοσύνης για την υποβοήθηση διαφορικής διάγνωσης νόσων εγκεφάλου με χρήση μαγνητικής φασματοσκοπίας και τεχνικών διάχυσης και αιμάτωσης σε υψηλό πεδίο 3T
Abstract
dc:descriptionMagnetic Resonance Spectroscopy (MRS) has been studied for more than a decade as a promising diagnostic tool for a variety of pathologies. Coupled with the morphological features provided by Magnetic Resonance Imaging (MRI) techniques, it can provide accurate identification and quantification of biologically important compounds in soft tissue. Diffusion Weighted Imaging (DWI) is a rapid technique evaluating the movement of water molecules inside tissues, a property which is altered under pathological conditions in the brain. Diffusion Tensor Imaging (DTI) represents a further development of DWI, taking into account the directionality of water diffusion in a certain voxel due to the microstructure of the underlying tissue. Dynamic, contrast-enhanced perfusion weighted MR imaging (PWI) of the brain provides hemodynamic information that complements the anatomic information available from conventional MR imaging. PWI is a fast and robust imaging technique that is increasingly being used as a research tool to evaluate and understand intracranial disease processes and as a clinical tool for diagnosis, management, and understanding of intracranial mass lesions, brain tumors in particular. Last years the continuous development of imaging techniques has transformed these modalities from conventional imaging to high-level metric systems, which provide a quite large amount of quantitative information. This large amount of numeric MR data often remains unexploited during the clinical routine. The reason is that the simultaneous analysis and evaluation of multiple parameters is a time consuming process and requires specific expertise. The aforementioned MRI techniques in order to transit from the process of experimental evaluation into clinical routine, the implementation of a classification model it is necessary, so as to automate the process of data assessment and characterization of pathology based on the metabolic findings and the parameters of the diffusion and perfusion techniques. The machine learning methods provide the mathematical and computational mechanisms and take advance of the underlying pathology in order to distinguish lesions under evaluation. An innovative decision support system FA.S.M.A (Fast Spectroscopic Multiple Analysis) was developed that constitutes the first clinical decision support system that evaluates the metabolic, diffusion and perfusion intratumoral and peritumoral data in high magnetic field 3T. The embedded classification system allows an easy and immediate classification of a new clinical case in one of the brain tumor types that have been included in the system. FA.S.M.A is user friendly and easy-to-use system that permits the visual and quantitative evaluation of data during all the individual steps that take place during a full examination MRI protocol. By this way the system’s autonomy as a “supportive” tool during the clinical decision and differentiation procedures is ensured in order to serve as a non - invasive assistant tool during clinical routine.
Degree
thesis:*- Grantor dc:publisher
- University of Thessaly (UTH)
- Year dc:date
- 2013
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Tsolaki, Evangelia
- Τσολάκη, Ευαγγελία
Subjects
dc:subject × 24- Τεχνητή νοημοσύνη
- Μηχανική μάθηση
- Σύστημα υποβοήθησης κλινικής διάγνωσης
- Νόσοι κεντρικού νευρικού συστήματος
- Φασματοσκοπία πρωτονίου μαγνητικού συντονισμού
- Απεικόνιση βαρύτητας διάχυσης
- Απεικόνιση τανυστή διάχυσης
- Απεικόνιση δυναμικής αιμάτωσης
- Artificial intelligence
- Machine learning
- Clinical decision support systems
- Central nervous lesions
- Magnetic resonance spectroscopy
- Diffusion weighted imaging
- Diffusion Tensor Imaging (DTI)
- Perfusion weighted imaging
- Φυσικές Επιστήμες
- Φυσική
- Ιατρική και Επιστήμες Υγείας
- Κλινική Ιατρική
- Natural Sciences
- Physical Sciences
- Medical and Health Sciences
- Clinical Medicine
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/31559