Ruđer Bošković Institute Research Data Repository
| Title (english) | ColonCancerHSI_Data |
| Author | Ivica Kopriva Ruđer Bošković Institute |
| Scientific / art field, discipline and subdiscipline | TECHNICAL SCIENCES Computing Data Processing |
| Abstract | This directory contains 27 hyerspectral images (HSIs) of hematoxylin-end-eosin stained pathohistological specimens collected intraoperatively from 14 patients with diagnosis adenocarcinoma of the colon in the liver. Directory also contains corresponding (co-registered) pseudo-RGB color images. Stored data can also be used for semi-supervised learning (SSL) classifiers in combination with two kinds of 3D spectral spatial features. In this case all data are equal, i.e., there are no train and test sets. MATLAB scripts for two SSL classifiers are available at: https://github.com/ikopriva/ColonCancerHSI. Hyperspectral image analysis was conducted as a part of research funded by the Croatian science foundation grant IP-2022-10-6403. |
| Methods | Hyperspectral images were acquired at Photon etc's hyperspectral microscope (https://www.photonetc.com) in the range 450nm to 800nm with a spectral resolution of 1nm. The specimens were gathered through a clinical study funded by the Croatian science foundation grant IP-2016--06-5235, conducted between March 2017 and February 2020 at the Department of Pathology and Cytology, Clinical Hospital Dubrava, Zagreb, Croatia.This dataset is a part of larger cohort of 19 patients with the same diagnosis, for which 82 RGB images with pixel-wise annotations are publicly available at: https://cocahis.irb.hr. The Institutional Review Board of Clinical Hospital Dubrava approved the collection of samples on May 24, 2016. All patients provided informed consent, and the data were anonymized. |
| Keywords (english) | |
| Language | english |
| URN:NBN | urn:nbn:hr:241:134948 |
| Publication | 2024-11-21 |
| Version | 1.1 |
| Geolocation | Zagreb, Bijenička cesta 54, Croatia |
| Project | Number: IP-2022-10-6403 Title (english): Development of algorithms for data clustering, regression and feature extraction with applications in pathology, ophtalmology and metabolomics Acronym: PHATOME Leader: Ivica Kopriva Jurisdiction: Croatia Funding stream: Istraživački projekti |
| External related object | Is cited by: DOI 10.48550/arXiv.2411.06969 |
| Publisher | Institut Ruđer Bošković Ruđer Bošković Institute |
| Access conditions | Open access |
| Terms of use |  |
| Public note (english) | If you intend to use these data for you research please do not forget to cite this reference and the following: Ivica Kopriva, Dario Sitnik, Laura-Isabelle Dion-Bertrand, Marija Milković Periša, Mirko Hadžija, Marijana Popović Hadžija, "A Hyperspectral Imaging Dataset and Methodology for Intraoperative Pixel-Wise Classification of Metastatic Colon Cancer in the Liver," http://arxiv.org/abs/2411.06969 |
| Created on | 2024-11-21 08:11:49 |
