polito.it
Politecnico di Torino (logo)

Automatic segmentation algorithm and Texture Analysis for tumor tissues on Mammary Magnetic Resonance Imaging (MMRI)

Stefania Zara

Automatic segmentation algorithm and Texture Analysis for tumor tissues on Mammary Magnetic Resonance Imaging (MMRI).

Rel. Filippo Molinari. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Biomedica, 2018

[img]
Preview
PDF (Tesi_di_laurea) - Tesi
Document access: Anyone
Licenza: Creative Commons Attribution Non-commercial No Derivatives.

Download (2MB) | Preview
Abstract:

The research activity of this thesis project has led to the development of a Computer Aided Diagnosis (CAD) system for the analysis of breast cancer lesions using Mammary Magnetic Resonance images. The main objective of this system is the tumor lesions segmentation and their characterization. The process consists of five main parts: image pre-processing, regions of interest contouring, lesions segmentations, heuristic analysis of the results and parametric characterization of the selected regions.A MATLAB algorithm was therefore created for:1. Image pre-processing, images loaded are processed to highlight lesions.2. Breast segmentation, the breasts are identified to prevent false positives (FP) due to structures such as heart and vessels that are not of clinical interest in this application.3. Contouring of the tumor lesions, a double-threshold segmentation algorithm was created.4. False positives reduction, the suspicious areas are analyzed and evaluated, if considered FP then they are discarded.5. Computation of lesion features, geometrical information and clinical features are extrapolated to characterize segmented regions.The algorithm was built on 5 patients and subsequently validated on a further 5 patients of which several post-treatment time scans were available, overall 20. The images used were provided by the Borgo Trento Hospital in Verona in collaboration with Tecnologie Avanzate S.r.l.

Relators: Filippo Molinari
Academic year: 2017/18
Publication type: Electronic
Number of Pages: 61
Subjects:
Corso di laurea: Corso di laurea magistrale in Ingegneria Biomedica
Classe di laurea: New organization > Master science > LM-21 - BIOMEDICAL ENGINEERING
Aziende collaboratrici: Tecnologie Avanzate
URI: http://webthesis.biblio.polito.it/id/eprint/8013
Modify record (reserved for operators) Modify record (reserved for operators)