Publication details
Examination of the Spatial Structure of Pigs' Melanoma in Tissue Sections Based on Histology and Mass Spectrometry
| Authors | |
|---|---|
| Year of publication | 2019 |
| Type | Article in Proceedings |
| Conference | WORLD CONGRESS ON MEDICAL PHYSICS AND BIOMEDICAL ENGINEERING 2018, VOL 1 |
| MU Faculty or unit | |
| Citation | |
| Web | https://link.springer.com/chapter/10.1007%2F978-981-10-9035-6_46 |
| Doi | http://dx.doi.org/10.1007/978-981-10-9035-6_46 |
| Keywords | Clustering; Image processing; MeLiM; Melanoma 2D maximum overlap wavelet transform |
| Description | We examine the spatial structure of the melanoma in tissue sections. The pigs melanoma was examined in 10 tissue samples obtained from animals of age between 4 and 22 weeks. The tissue sections were measured by light microscopy and Laser Ablation Ion Inductively Coupled Plasma Mass Spectrometry to obtain spatial metal (Cu, Zn) distribution. The exploratory analysis of the tissue sections indicates there is clearly a pattern in the spatial structure. Different projections of the spatial structure of the melanoma are obtained by the different measurement methods. The spectral clustering on the data was utilized to describe the structure in the data. According to the clustering results, there are distinct clusters of observations in the histological data. The spatial elemental distribution of the metals Cu and Zn cannot be clustered -the data form one compact cluster. The clustering of the histological images produces clusters which are related to the annotation of the biological samples in broader terms -the differences between fibrous and cancerous tissue. |