Klasifikacija bolesti temeljem rendgenskih snimki prsnog koša korištenjem metoda strojnog učenja

undergraduate thesis

• Title (english): Classification of diseases based on chest X-rays using machine learning methods
• Author: Andrea Milanović
• Mentor: Alan Jović (mentor)
• Committee member: Alan Jović (predsjednik povjerenstva)
• Committee member: Hrvoje Nuić (član povjerenstva)
• Committee member: Eda Jovičić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-02, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovaj rad prikazuje primjenu konvolucijskih neuronskih mreža za klasifikaciju i detekciju
bolesti temeljem rendgenskih snimki prsnog koša. Prikazana je izgradnja dva
klasifikacijska te dva detekcijska modela i uspoređena je njihova uspješnost. Analizirane
su metrike točnosti, preciznosti i konfuzijske matrice za izgrađene modele. Za izradu
rada korišten je programski jezik Python te modeli ResNet, EfficientNet i YOLOv8.
Model ResNet postigao je točnost od 50 %. Model YOLO ukupno postiže preciznost od
20.5 % i odziv od 18.5 %, no za neka oboljenja poput kardiomegalije te vrijednosti prelaze
70 %.

Abstract (english)

This paper presents the application of convolutional neural networks for the classification
and detection of diseases based on chest X-ray images. The construction of
two classification and two detection models is demonstrated, and their performance is
compared. Accuracy metrics, precision metrics and confusion matrices are analyzed for
the constructed models. Python programming language was used for the development
of the paper, along with ResNet, EfficientNet, and YOLOv8 models. The ResNet model
achieved an accuracy of 50 %. The YOLO model achieved an overall precision of 20.5 %
and a recall of 18.5 %, but for some diseases such as cardiomegaly, these values exceed
70 %.

• Keywords: strojno učenje; konvolucijske neuronske mreže; klasifikacija slika; detekcija na slikama; rendgen prsnog koša
• Keywords (english): machine learning; convolutional neural networks; image classification; image detection; chest X-ray
• Language: croatian
• URN:NBN: urn:nbn:hr:168:556377
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2024-10-12 21:35:50