diplomsko delo
Aljoša Koren (Author), Matej Kristan (Mentor)

Abstract

V diplomski nalogi naslavljamo problem avtomatske diagnoze malarije. V okviru naloge smo razvili orodje medicinskim tehnikom, ki detektira število s paraziti okuženih celic ter zdravih celic v slikah krvnih razmazov. Orodje za diagnostiko, ki smo ga razvili, bistveno pospeši proces diagnostike. Okužene celice so dodatno klasificirane v štiri različne življenjske cikle parazitov. V okviru naloge smo razvili modela MD_FCOS_c6 in MD_FCOS_c2, ki uporabljata FCOS in model MD_FCOS_DNet_c6, ki uporablja FCOS in DenseNet. Modele smo analizirali na prostodostopni bazi anotiranih slik krvnih razmazov, ki vsebuje 1364 slik. Modela MD_FCOS_c6 in MD_FCOS_DNet_c6 detektirata šest razredov celic, medtem ko model MD_FCOS_c2 detektira zgolj zdrave in okužene celice. Model MD_FCOS_c6, ki smo ga razvili, dosega F1 metriko nad 0.7 za polovico razredov, za drugo polovico pa nad 0.3. Model MD_FCOS_c2, ki je specializiran za zgolj razlikovanje med okuženimi in zdravimi celicami, dosega F1 za zdrave celice nad 0.95 in F1 za okužene celice nad 0.8. Tretji model, ki smo ga razvili, pa dosega boljše rezultate na slabo zastopanih razredih in slabše na boljše zastopanih razredih v podatkovni zbirki. Model MD_FCOS_DNet_c6 s 98.8% verjetnostjo prepozna sliko okuženega krvnega razmaza kot okuženo. V okviru naloge smo razvili tudi spletno aplikacijo, ki omogoča uporabo našega modela zdravstvenim tehnikom. Aplikacija s preprostim uporabniškim vmesnikom omogoča nalaganje slik, na katerih model nato naredi predikcije. Diagnoza se nato v manj kot 30 sekundah pojavi na zaslonu uporabnika, kjer lahko vidi z različnimi barvami anotirane celice na naloženi sliki.

Keywords

detekcija objektov;diagnosticiranje;računalništvo in informatika;univerzitetni študij;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [A. Koren]
UDC: 004.932.72'1(043.2)
COBISS: 76822275 Link will open in a new window
Views: 284
Downloads: 81
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: A tool for diagnosing malaria from blood smears
Secondary abstract: We addressed the problem of automatic malaria diagnosis. In particular, we designed a tool for medical technicians that detects and counts the number of infected and healthy cells from blood smear images and also classifies cells into a particular life cycle of malaria infection. The tool provides a major improvement in the speed of the diagnosis process. We developed two models; MD_FCOS_c6 and MD_FCOS_c2, which both use FCOS, and model MD_FCOS_DNet_c6 which uses FCOS and DenseNet. We analyzed our models on an open-sourced dataset of blood smears which contains 1364 images. MD_FCOS_c6 and MD_FCOS_DNet_c6 models detect six classes of cells, meanwhile MD_FCOS_c2 model detects only healthy and infected cells. The MD_FCOS_c6 model achieves a F1 score above 0.7 for half of the classes and above 0.3 for the other half. MD_FCOS_c2 model, specialized in detecting the difference between infected and healthy cells, has F1 score above 0.95 for healthy cells and above 0.8 for infected cells. MD_FCOS_DNet_c6 model is better at detecting classes that are underrepresented in the database, but also worse than MD_FCOS_c6 at detecting well represented classes. MD_FCOS_DNet_c6 model recognizes an infected blood smear as infected with the probability of 98.8%. We also developed a web application that enables the usage of our model to medical technicians. The application provides a simple user interface that shows the annotated cells on the input image, providing a tool to technicians that shows a diagnosis in less than 30 seconds.
Secondary keywords: malaria;object detection;diagnostics;computer science;computer and information science;diploma;Malarija;Sistemi za prepoznavanje vzorcev;Računalništvo;Univerzitetna in visokošolska dela;
Type (COBISS): Bachelor thesis/paper
Study programme: 1000468
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 47 str.
ID: 13345763