Institutional Repository of Vilnius University
Development of nonlinear microscopy methods for biological imaging /
Abstract
dc:descriptionCollagen is an important mammalian structural protein which is involved in various cellular processes. As its structure changes during various diseases, investigation of these differences can give information on the processes occurring in tissue during disease development. In this work, polarimetric second harmonic generation (SHG) microscopy was used to investigate collagen structure in gastrointestinal stromal tumor (GIST) samples. Double Stokes-Mueller Polarimetry (DSMP) imaging was done in normal and cancerous sample regions. Polarimetric and textural analysis was performed on the acquired data, and the resulting parameters were used to compare normal and cancerous sample areas. Statistical analysis showed SHG intensity, SHGCD, and C ratio polarimetric parameters as the most efficient in differentiating between normal and cancerous regions. Tissue response to nearby tumor was not identical in all samples, with collagen fibers being either broken down or rearranged into more orderly structures, the latter resulting in more intense SHG. Successful tissue differentiation with SHGCD and C-ratio indicated that the out of-plane collagen fiber orientation distribution near tumor was altered. Finally, a logistic regression classifier was used to distinguish normal and cancerous tissue regions and was shown to successfully identify tissue closer to tumor as cancerous in one sample, however, due to a small training dataset, classification of another sample yielded ambiguous results. The results obtained in this work could be used to gain further insight into the processes occurring during tumorigenesis.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Rupšys, Gustas,
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:ELABAETD158213028&prefLang=en_US