izpis_h1_title_alt

AVTOMATSKA ZAZNAVA SPREMEMB BELE MOŽGANOVINE V MAGNETNO RESONANČNIH SLIKAH
ID LESJAK, ŽIGA (Author), ID Vrtovec, Tomaž (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (6,36 MB)
MD5: 7C38D09BA20E7C106C13C56DB745697E
PID: 20.500.12556/rul/d7754ed7-486f-496d-b54d-ed91c2144234

Abstract
Možganske bolezni so eden od vodilnih vzrokov telesne in duševne invalidnosti v sodobni družbi in imajo zelo velik socialnoekonomski učinek. Pri teh boleznih se patologija pogosto odraža v obliki lezij bele možganovine (LBM), ki imajo velik diagnostični in prognostišni pomen. Za kvantifikacijo prostornine in števila LBM, ki predstavljata pomemben biomarker za spremljanje in napovedovanje poteka pri mnogih možganskih boleznih kot na primer MS, so bili razviti številni avtomatski postopki. Objektivna in temeljita validacija teh postopkov je na kliničnih MR slikah zelo težavna, saj potrebujemo referenčne obrise LBM. Referenčni obrisi morajo biti čim bolj natančni in zanesljivi, pridobimo pa jih lahko le z ročnim obrisovanjem. S tem namenom smo razvili protokol za izdelavo referenčnih obrisov z medsebojnim konsenzom večih obrisovalcev in pokazali, da so na ta način pridobljeni obrisi natančnejši in zanesljivejši kot obrisi posameznih izkušenih obrisovalcev. Z omenjenim protokolom smo ustvarili referen čne obrise LBM za zbirko MR slik 30 MS bolnikov, ki je bila zajeta na 3T MR napravi in je vsebovala T1- in T2-utežene ter FLAIR slike. Omenjeni protokol smo uporabili tudi za izgradnjo referenčnih obrisov sprememb LBM longitudinalne MR zbirke 20 MS bolnikov, zajete na 1,5T Philips MR napravi. Longitudinalna zbirka je poleg referenčnih obrisov sprememb vsebovala za posameznega bolnika vsaj dve MR preiskavi, vsaka od preiskav pa T1- in T2-utežene ter FLAIR slike. S pomočjo longitudinalne zbirke slik s konsenznimi referenčnimi obrisi smo primerjali in validirali uveljavljene in najnovejše postopke za zaznavo sprememb LBM iz longitudinalnih MR slik glave, ki temeljijo na primerjavi sivinskih vrednosti. Za validacijo smo poleg standardnih metrik, ki se pojavljajo v znanstveni literaturi, uporabili na novo razvito metriko Regijskega Dice-ovega koeficienta, ki nam je omogočala primerjavo zmožnosti zaznave sprememb v odvisnosti od njihove velikosti. Rezultati validacije so precej odstopali od izvirnih rezultatov drugih avtorjev in kažejo na visoko odvisnost med zmožnostjo zaznave sprememb in uporabljeno validacijsko zbirko slik ter natančnostjo referenčnih obrisov. Z namenom olajšanja validacije uveljavljenih in novo razvitih postopkov, smo omenjeni zbirki s konsenznimi referenčnimi obrisi tudi javno objavili. Postopki strojnega učenja omogočajo primerjavo in določitev optimalnih značilnic za zaznavo sprememb LBM ter s tem tudi identifikacijo slikovne informacije, ki ključno vpliva na zmožnost zaznave sprememb. Z uporabo razvrščevalnika z naključnimi gozdovi smo ovrednotili in primerjali pomembnost značilnic za zaznavo LBM, določili nabor optimalnih značilnic, zanesljivost ocene pomembnosti značilnic in občutljivost na izbrani postopek razvrščanja. Izbor značilnic določa za zaznavo sprememb pomembne MR sekvence in njihove nastavitve, zato lahko na ta način optimiziramo tudi postopek zajema MR slik in po potrebi iz protokola zajema odstranimo nepotrebne MR sekvence. To lahko posledično skrajša ˇcas zajema MR preiskave ter tako zniža stroške preiskave in obremenitve tako bolnikov kot tudi medicinskega osebja. S predlagano metodologije izbire optimalnih značilnic lahko načrtamo postopke zaznave sprememb z višjo občutljivostjo in specifičnostjo, kar v klinični praksi omogoča natančnejše spremljanje in boljše razumevanje poteka bolezni, predvsem pa hitrejše ukrepanje v primerih, ko kopičenje sprememb kaže na neučinkovitost trenutne terapije.

Language:Slovenian
Work type:Doctoral dissertation
Organization:FE - Faculty of Electrical Engineering
Year:2017
PID:20.500.12556/RUL-91174 This link opens in a new window
COBISS.SI-ID:11731028 This link opens in a new window
Publication date in RUL:24.03.2017
Views:12281
Downloads:850
Metadata:XML RDF-CHPDL DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:AUTOMATED DETECTION OF BRAIN WHITE MATTER CHANGES IN MAGNETIC RESONANCE IMAGES
Abstract:
Neurological diseases are among the most common causes of mental and physical disabilities in the modern society and as such present a huge socio-economic impact. At the onset and during the progression of these diseases, for instance in Multiple Sclerosis, the underlying pathological processes often result in the occurence of scar tissue or lesions in the white matter. These lesions are best observed by Magnetic Resonance imaging. Quantification of white matter lesion volume and count from the MR images therefore represents an important biomarker for diagnosis, prognosis and treatment followup of neurological diseases. So far many automated methods for the MR image quantification have been developed. However, an objective an extensive validation of such methods on real, clinical image dataset is a challenging task since it requires accurate and reliable reference delineations of white matter lesions, which can only be acquired by manual delineation of each lesion. To facilitate the creation of such reference delineations we proposed a protocol, in which multiple raters created a consensus based reference delineations of white matter lesions. Validation of the proposed protocol shows that it results in a more accurate and reliable reference delineations of lesions in comparison with the delineations made by any single rater. The protocol was used to create reference delineations of white matter lesions for a database of 30 MS patients, which consisted of T1- and T2-weighted and FLAIR images MR images acquired on a 3T MR machine. The same protocol was also employed to build a second, longitudinal database of 20 MS patients with two MR studies per patient and corresponding reference white matter lesion changes delineations. Each study consisted of T1- and T2-weighted and FLAIR images MR images acquired on 1,5T Philips MR machine. Using the longitudinal MR database with reference lesion change delineations we validated and compared state-of-the-art methods for lesion change detection based on longitudinal analysis of intensity variations. Validation was performed using standard metrics found in the literature with the addition of a new metric – regional Dice coecient, which allowed the analysis of methods’ performance with respect to the size of each particular lesion change. Obtained results were not nearly as good as the ones reported by the original authors, which suggests that the performances of the evaluated change detection methods might either be very dependent on the MR image acquisition or dependent on the accuracy of ground truth segmentation. In order to facilitate validation of existing and newly developed change detection methods the aforementioned datasets were publicly disseminated. Machine learning algorithms can be used to determine the set of optimal features required for white matter lesion change detection. Identification of such features helps us to better understand, which imaging information is the most important for change detection. Using a random forest classifier we evaluated and compared the importance of various features for lesion change detection, assessed the reliability of the estimated feature importance and, finally, determined the ability to generalize feature importance estimation to other classifiers. The selection of optimal features may influence the setup of MR acquisition parameters and sequences needed for lesion change detection and as such can be used to optimize MR acquisition, e.g. to remove MR sequences unnecessary for change detection. Such optimizations could potentially decrease the cost and time required to acquire an MR patient study, which would benefit both the patients and the medical personnel. The proposed feature selection method can also be used to develop change detection methods with higher accuracy and specificity, which, in turn, would enable clinicians a better insight into disease progression, increase their understanding of the underlying pathology and allow them to make timely and fully-informed decisision in case of ineective treatments.


Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back