izpis_h1_title_alt

Zaznavanje vnetij sklepov rok iz popravljenih RGB slik
Pikl, Kristina (Author), Milanič, Matija (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (6,66 MB)
MD5: A724946BAC2CF887C99388533D609A1C

Abstract
Vnetne bolezni sklepov (npr. revmatoidni artritis) prizadenejo sklepe v človeškem telesu, kar najprej vodi v boleče sklepe, po daljšem času, brez zdravljenja, pa v trajne poškodbe sklepov. Bolniki postanejo gibalno ovirani in težje opravljajo vsakodnevne dejavnosti. V nekaterih primerih lahko pride do nezmožnosti opravljanja dejavnosti. Kakovost njihovega življenja je bistveno zmanjšana. Za preprečevanje trajnih posledic in čim večje izboljšanje kakovosti življenja, je potrebno pričeti z ustreznim zdravljenjem kar se le da zgodaj. Diagnostiko v Sloveniji se opravlja najpogosteje s kliničnim in ultrazvočnim pregledom. Ker je tovrsten pregled počasen in zahteva izkušenega zdravnika, obstaja potreba po novih, hitrejših in natančnejših diagnostičnih tehnikah. V okviru te diplomske naloge smo zato raziskali možnost uporabe RGB (angl. red - green - blue) slik rok za odkrivanje artritisa malih sklepov rok. Posnete RGB slike rok pacientov smo najprej popravili z algoritmom za korekcijo krivin, s čimer z RGB slik odstranimo artefakte osvetlitve, kot so sence na predelih z večjimi krivinami. V nadaljevanju slike pretvorimo v CIEXYZ prostor barv, XYZ slike pa povežemo s fiziološkimi lastnostmi, kot sta količina krvi in nasičenost s kisikom. Na zbirki slik 29 bolnikov z revmatoidnim in psoriartričnim artritisom smo ocenili, da tovrsten pristop še ni uporaben za diagnostiko vnetja malih sklepov rok. Predlagamo pa tudi način izboljšanja pristopa, da bi bi bilo lahko zaznavanje mogoče.

Language:Slovenian
Keywords:RGB, artritis, diagnostika, fiziološke lastnosti
Work type:Final paper (mb12)
Tipology:2.11 - Undergraduate Thesis
Organization:FMF - Faculty of Mathematics and Physics
Year:2020
COBISS.SI-ID:31569155 This link opens in a new window
Views:97
Downloads:68
Metadata:XML RDF-CHPDL DC-XML DC-RDF
 
Average score:(0 votes)
Your score:Voting is allowed only to logged in users.
:
Share:AddThis
AddThis uses cookies that require your consent. Edit consent...

Secondary language

Language:English
Title:Detection of joint inflammation in hands from corrected RGB images
Abstract:
Inflammatory joint diseases (e.g. rheumatoid arthritis) affect joints in human body leading first to painful joints and after a long time without treatment to permanent joint damage. Patients become physically deprived and find it more difficult to perform daily activities. In some cases, a complete inability to perform activities may occur. Therefore quality of their lives is significantly reduced. To prevent long-term consequences and improve their quality of life as much as possible, it is necessary to start with appropriate treatment as early as possible. Diagnosis in Slovenia is most often performed by clinical and ultrasound examination. Because this type of examination is slow and requires an experienced physician, there is a need for newer, faster, and more accurate diagnostic techniques. Within this thesis, we therefore explored the possibility of using RGB (red - green - blue) hand images to detect arthritis of the small joints of hands. The captured RGB images of patients' hands were first corrected with a curvature correction algorithm to remove illumination artifacts from RGB images, such as shadows on the areas with larger curvatures. Next, the images are converted to the CIEXYZ color space, and finally XYZ images are associated with physiological properties such as blood volume fraction and blood oxygen saturation. Analysing a collection of images of 29 patients with rheumatoid and psoriatic arthritis, we estimated that this approach is not yet useful for the diagnosis of inflammation of the small joints of hands. However, we suggest a way to improve the approach so that the dissease detection could be made.

Keywords:RGB, arthritis, diagnostics, physiological properties

Similar documents

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

Comments

Leave comment

You have to log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back