Institutional Repository of Vilnius University
Cancer induced changes in prostate tissue revealed by nonlinear microscopy and ai assisted image analysis /
Abstract
dc:descriptionThis bachelor's thesis is based on investigation of cancer induced changes in prostate tissue using nonlinear microscopy and artificial intelligence tools. Prostate cancer, second most frequently occurring cancer among men, tend to remain undetected for years due to the lack of symptoms in early stages. Traditional diagnostic methods, including PSA testing, and biopsy, have limitations in accuracy and early detection. Nonlinear imaging techniques, such as second harmonic generation (SHG) and third harmonic generation (THG), offer deeper tissue visualization and have shown potential in examining lung, breast and ovaries cancerous tissues. The aim of this work is to use imaging techniques that involve nonlinear processes, polarization and GLCM based texture analysis to get a deeper structural understanding of the histological prostate cancer samples. The research involved intensity and polarimetric DSMP imaging methods, texture analysis of SHG and THG images, and AI-assisted reconstruction of H&E images. Results indicated that SHG intensity was generally low in prostate tissue, limiting the effectiveness of the method. Texture analysis of THG images revealed significant differences between cancerous and healthy tissues, highlighting its potential in cancer diagnosis. The AI algorithm successfully reconstructed H&E images from THG and MPF data, facilitating better visualization for pathologists. Overall, all of the three contrast mechanisms (SHG, THG and MPF) can be described as complimentary, each of them providing different information. SHG analysis did not report any evident collagen changes for cancerous regions whereas THG imaging and demonstrated significant potential in differentiating prostate cancer tissue, offering a promising approach for diagnosis and understanding of prostate cancer.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Bareikytė, Austėja,
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- eng
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD229556881&prefLang=en_US