This work originates from the growing interest of the medical imaging community in the application of machine learning techniques and, from deep learning to improve the accuracy of cancerscreening. The thesis is structured into two different tasks. In the first part, magnetic resonance images were analysed in order to support clinical experts in the treatment of patients with brain tumour metastases (BM). The main topic related to this study was to investigate whether BM segmentation may be approached successfully by two supervised ML classifiers belonging to feature-based and deep learning approaches, respectively. SVM and V-Net Convolutional Neural Network model are selected from the literature as representative of the two approaches. The second task related to this thesisis illustrated the development of a deep learning study aimed to process and classify lesions in mammograms with the use of slender neural networks. Mammography has a central role in screening and diagnosis of breast lesions. Deep Convolutional Neural Networks have shown a great potentiality to address the issue of early detection of breast cancer with an acceptable level of accuracy and reproducibility. A traditional convolution network was compared with a novel one obtained making use of much more efficient depth wise separable convolution layers. As a final goal to integrate the system developed in clinical practice, for both fields studied, all the Medical Imaging and Pattern Recognition algorithmic solutions have been integrated into a MATLAB® software package

Investigating the role of machine learning and deep learning techniques in medical image segmentation / Gloria Gonella , 2020. 33. ciclo, Anno Accademico 2019/2020.

Investigating the role of machine learning and deep learning techniques in medical image segmentation

gonella gloria
2020-01-01

Abstract

This work originates from the growing interest of the medical imaging community in the application of machine learning techniques and, from deep learning to improve the accuracy of cancerscreening. The thesis is structured into two different tasks. In the first part, magnetic resonance images were analysed in order to support clinical experts in the treatment of patients with brain tumour metastases (BM). The main topic related to this study was to investigate whether BM segmentation may be approached successfully by two supervised ML classifiers belonging to feature-based and deep learning approaches, respectively. SVM and V-Net Convolutional Neural Network model are selected from the literature as representative of the two approaches. The second task related to this thesisis illustrated the development of a deep learning study aimed to process and classify lesions in mammograms with the use of slender neural networks. Mammography has a central role in screening and diagnosis of breast lesions. Deep Convolutional Neural Networks have shown a great potentiality to address the issue of early detection of breast cancer with an acceptable level of accuracy and reproducibility. A traditional convolution network was compared with a novel one obtained making use of much more efficient depth wise separable convolution layers. As a final goal to integrate the system developed in clinical practice, for both fields studied, all the Medical Imaging and Pattern Recognition algorithmic solutions have been integrated into a MATLAB® software package
2020
Deep Learning; CNN; Mammograms; Screening; medical software, MRI; brain segmentation; brain metastases; machine learning; features extraction; convolutional neural network
Investigating the role of machine learning and deep learning techniques in medical image segmentation / Gloria Gonella , 2020. 33. ciclo, Anno Accademico 2019/2020.
File in questo prodotto:
File Dimensione Formato  
PhD_Thesis_Gloria_Gonella_pdfa_compresso_pdfa.pdf

accesso aperto

Tipologia: Tesi di dottorato
Licenza: Dominio pubblico
Dimensione 2.8 MB
Formato Adobe PDF
2.8 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11383/2115045
 Attenzione

L'Ateneo sottopone a validazione solo i file PDF allegati

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact