Avtomatska segmentacija anizotropnih podatkov pridobljenih z elektronskim mikroskopom

MRVAR, GAL (author), Marolt, Matija (mentor)

Mitohondrij je pomemben organel evkariontskih celic. Njegova znotrajcelična segmentacija predstavlja izziv, saj se mitohondriji kvalitativno in kvantitativno razlikujejo glede na celico, v kateri se nahajajo. Klasično se mitohondrije segmentira ročno, kar je časovno neučinkovito in izpostavljeno človeški napaki ter interpretaciji. V sklopu tega diplomskega dela zato predstavljamo postopek, ki omogoča avtomatsko segmentacijo mitohondrijev nad anizotropnimi volumetričnimi podatki pridobljenimi z elektronskim mikroskopom (EM). Predlagamo postopek oz. cevovod, ki zajema operacije nad volumetričnimi podatki oz. obdelavo le teh in avtomatsko segmentacijo z uporabo konvolucijske nevronske mreže. Rezultate avtomatske segmentacije smo evalvirali nad testnima množicama, s čimer smo pridobili zelo dobre rezultate – klasifikacijska točnost je predstavljala 99 % nad obema testnima množicama.

• Language: Slovenian
• Keywords: konvolucijske nevronske mreže, avtomatska segmentacija, mitohondriji
• Work type: Bachelor thesis/paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2021
• PID: 20.500.12556/RUL-124807
• COBISS.SI-ID: 52402179
• Publication date in RUL: 19.02.2021

MRVAR, GAL, 2021, Avtomatska segmentacija anizotropnih podatkov pridobljenih z elektronskim mikroskopom [online]. Bachelor’s thesis. [Accessed 26 March 2025]. Retrieved from: https://repozitorij.uni-lj.si/IzpisGradiva.php?lang=eng&id=124807

Secondary language

• Language: English
• Title: Automatic segmentation of anisotropic electron microscope data

Abstract

Mitochondria is an important organelle of eukaryotic cells. Its intracellular segmentation presents a challenge as mitochondria differs qualitatively and quantitatively according to the cell in which they are located. The classic approach of mitochondria segmentation is manual, which is time consuming and prone to human error and interpretation. As a part of this diploma thesis, we are presenting a procedure that enables automatic segmentation of mitochondria over anisotropic volumetric data obtained with an electronic microscope (EM). We propose a procedure or a pipeline, which includes operations on volumetric data, and automatic segmentation using a convolutional neural network. We have evaluated the results of automatic segmentation using two different test sets. The evaluation showed very promising results – the classification accuracy was 99 % on both test sets.

• Keywords: Convolutional neural networks, automatic segmentation, mitochondria