Classification of perimetric data for supporting glaucoma diagnosis

(master's thesis)

Janja Belinc (Author), Milan Zorman (Mentor)

Abstract

The aim of the study: Glaucoma is a chronic, progressive and asymptomatic retinal disease which results in an irreversible visual field loss. The main objective of this Master’s thesis work was to study the applicability of classification techniques for supporting glaucoma diagnosis. Research Methodology: In this study perimetric data was obtained by SPARK strategy implemented in Oculus perimeters and provided by medical experts from the Hospital Universitario de Canarias (HUC). This data was used for constructing the feature vectors for the classification problem. Feature vectors of 66 values and feature vectors of 6 values were tested in the experiments. The proposed classification study attempted to: a) demonstrate that the studied classifiers were able to distinguish between “healthy” and “glaucomatous” eyes using only perimetric data, and b) analyse which feature vector design was the most suitable to accomplish this task. Results: The classification results showed that classifiers performed better on 6 than on 66 perimetry values, which demonstrated the suitability of the 6 points selected by the SPARK strategy and supported its use in medical field. Conclusion: In this study two remarkable findings for pattern recognition in perimetric data were obtained. Firstly, that reducing the dataset improved the efficiency of the studied classifier, and secondly, that simple pattern recognition models types were more efficient than complex ones.

Keywords

eye disease;visual field;SPARK perimetry;pattern recognition;machine learning;supervised learning;ROC analysis;

Data

Language:	English
Year of publishing:	2018
Typology:	2.09 - Master's Thesis
Organization:	UM FZV - Faculty of Health Sciences
Publisher:	[J. Belinc]
UDC:	617.7-007.681:004.8(043.2)
COBISS:	2428580
Views:	741
Downloads:	73
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	Slovenian
Secondary title:	Podpora diagnoze glavkoma z uporabo klasifikacijskih algoritmov na perimetričnih meritvah
Secondary abstract:	Izhodišča, namen: Glavkom je kronična, progresivna in asimptomatska bolezen mrežnice, ki povzroči nepopravljivo izgubo vidnega polja. Glavni cilj te magistrske naloge je bil preučiti uporabnost klasifikacijskih algoritmov za podporo diagnoze glavkoma. Raziskovalne metode: Perimetrični podatki, uporabljeni v tej študiji, so bili pridobljeni s strategijo SPARK, ki se izvaja v perimetrih proizvajalca Oculus. Perimetrične podatke so zagotovili zdravstveni strokovnjaki iz bolnišnice Hospital Universitario de Canarias, uporabili pa so se za izdelavo vektorjev funkcij pri problemu klasifikacije. V eksperimentih so bili testirani vektorji funkcij 66 vrednosti in vektorji funkcij 6 vrednosti. Predlagana klasifikacijska študija je poskušala: a) pokazati, da lahko preiskovani klasifikatorji razlikujejo med »zdravimi« in »obolelimi« očmi zgolj z uporabo perimetričnih podatkov in b) analizirati, katera zasnova vektorjev je najbolj primerna za izvedbo te naloge. Rezultati: Rezultati klasifikacije so pokazali, da klasifikatorji delujejo bolje na 6 kot na 66 perimetričnih vrednostih, kar dokazuje ustreznost 6 točk vidnega polja, ki jih je izbrala strategija SPARK, in podpirajo uporabo te strategije na medicinskem področju. Diskusija in zaključek: V tej študiji je prišlo do dveh izjemnih ugotovitev za prepoznavanje vzorcev na perimetričnih podatkih: prve, da zmanjšanje nabora podatkov izboljša učinkovitost preučevanega klasifikatorja, in druge, da enostavne vrste modelov prepoznavajo vzorce bolj učinkovito kot kompleksnejše.
Secondary keywords:	očesna bolezen;vidno polje;perimetrija SPARK;prepoznavanje vzorcev;strojno učenje;nadzorovano učenje;analiza ROC;Glaucoma;Perimetry;Regression analysis;Glavkom;Perimetrija;Regresijska analiza;
URN:	URN:SI:UM:
Type (COBISS):	Master's thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za zdravstvene vede
Pages:	VI, 42 str., 45 str. pril.
ID:	10938154