Back to Search
Start Over
Automated Algorithm for Standardized Quantification on Liver Fibrosis using Second Harmonic Generation Microscopy
- Source :
- 2008 IEEE PhotonicsGlobal@Singapore.
- Publication Year :
- 2008
- Publisher :
- IEEE, 2008.
-
Abstract
- Determining the extent of liver fibrosis has clinically been difficult due to the lack of a simple, objective method that can accurately quantify the amount of collagen in the diseased tissue. Second harmonic generation (SHG) microscopy has been shown to produce bright and robust signals from non-centrosymmetric fibrillar collagen. We designed a SHG system that can objectively measure collagen presence in livers. In addition, we have developed a fully automated algorithm to quantify liver fibrosis in an efficient, standardized and reproducible manner. SHG microscopy was performed on livers harvested from bile duct ligated Wistar rats using a confocal microscope with a mode-locked Ti:Sapphire laser. Images acquired were later analyzed with Otsu method and a custom-developed, fully automated algorithm to measure area of fibers. Both the qualitative and quantitative progression of collagen over time was observed using SHG microscopy. We have compared the modified quantification algorithm with Otsu segmentation algorithm. With the modified algorithm, we have maintained greater amount of collagen after segmentation. We have also shown modified algorithm was able to identify the presence of finer collagen which was completely removed with the normal Otsu method. We have built an imaging system using state-of-the-art SHG microscopy. In addition, a fully-automated algorithm was developed to quantify collagen content in tissue to measure the presence of collagen in high accuracy. By combining SHG microscopy and our quantification algorithm, we can provide sensitive measurement for liver fibrosis which accurately reflects the progression of liver fibrosis, especially in early stages.
Details
- Database :
- OpenAIRE
- Journal :
- 2008 IEEE PhotonicsGlobal@Singapore
- Accession number :
- edsair.doi...........c98265316786a6890f75c4094c01909f
- Full Text :
- https://doi.org/10.1109/ipgc.2008.4781383