A cell level automated approach for quantifying antibody staining in immunohistochemistry images. A structural approach for quantifying antibody staining in colonic cancer spheroid images by integrating image processing and machine learning towards the implementation of computer aided scoring of cancer markers.
View/ Open
Rema PhD-Thesis(after-viva-2).pdf (3.883Mb)
Download
Publication date
2013-12-09Author
Khorshed, Reema A.A.Supervisor
Jiang, JianminPhillips, Roger M.
Holton, Robert
Keyword
Image processingCancer
Immunohistological
Nuclei segmentation
Nuclei classification
Feature extraction
Automation
Antibody quantification
Antibody staining
Colonic cancer spheroid images
Cancer markers
Computer aided scoring
Machine learning
Rights
The University of Bradford theses are licenced under a Creative Commons Licence.
Institution
University of BradfordDepartment
Department of Computing, School of Computing, Informatics and MediaAwarded
2013
Metadata
Show full item recordAbstract
Immunohistological (IHC) stained images occupy a fundamental role in the pathologist¿s diagnosis and monitoring of cancer development. The manual process of monitoring such images is a subjective, time consuming process that typically relies on the visual ability and experience level of the pathologist. A novel and comprehensive system for the automated quantification of antibody inside stained cell nuclei in immunohistochemistry images is proposed and demonstrated in this research. The system is based on a cellular level approach, where each nucleus is individually analyzed to observe the effects of protein antibodies inside the nuclei. The system provides three main quantitative descriptions of stained nuclei. The first quantitative measurement automatically generates the total number of cell nuclei in an image. The second measure classifies the positive and negative stained nuclei based on the nuclei colour, morphological and textural features. Such features are extracted directly from each nucleus to provide discriminative characteristics of different stained nuclei. The output generated from the first and second quantitative measures are used collectively to calculate the percentage of positive nuclei (PS). The third measure proposes a novel automated method for determining the staining intensity level of positive nuclei or what is known as the intensity score (IS). The minor intensity features are observed and used to classify low, intermediate and high stained positive nuclei. Statistical methods were applied throughout the research to validate the system results against the ground truth pathology data. Experimental results demonstrate the effectiveness of the proposed approach and provide high accuracy when compared to the ground truth pathology data.Type
ThesisQualification name
PhDCollections
Related items
Showing items related by title, author, creator and subject.
-
Novel Ran-RCC1 inhibitory peptide-loaded nanoparticles have anti-cancer efficacy in vitro and in vivoHaggag, Y.A.; Matchett, K.B.; Falconer, Robert A.; Isreb, Mohammad; Jones, Jason; Faheem, A.; McCarron, P.; El-Tanani, Mohamed (2019-02)The delivery of anticancer agents to their subcellular sites of action is a significant challenge for effective cancer therapy. Peptides, which are integral to several oncogenic pathways, have significant potential to be utilised as cancer therapeutics due to their selectivity, high potency and lack of normal cell toxicity. Novel Ras protein-Regulator of chromosome condensation 1 (Ran-RCC1) inhibitory peptides designed to interact with Ran, a novel therapeutic target in breast cancer, were delivered by entrapment into polyethylene glycol-poly (lactic-co-glycolic acid) PEG-PLGA polymeric nanoparticles (NPs). A modified double emulsion solvent evaporation technique was used to optimise the physicochemical properties of these peptide-loaded biodegradable NPs. The anti-cancer activity of peptide-loaded NPs was studied in vitro using Ran-expressing metastatic breast (MDA-MB-231) and lung cancer (A549) cell lines, and in vivo using Solid Ehrlich Carcinoma-bearing mice. The anti-metastatic activity of peptide-loaded NPs was investigated using migration, invasion and colony formation assays in vitro. A PEG-PLGA-nanoparticle encapsulating N-terminal peptide showed a pronounced antitumor and anti-metastatic action in lung and breast cancer cells in vitro and caused a significant reduction of tumor volume and associated tumor growth inhibition of breast cancer model in vivo. These findings suggest that the novel inhibitory peptides encapsulated into PEGylated PLGA NPs are delivered effectively to interact and deactivate Ran. This novel Ran-targeting peptide construct shows significant potential for therapy of breast cancer and other cancers mediated by Ran overexpression.
-
Systematic associations between germ-line mutations and human cancersAl-Shammari, Mohamad H.; Tobin, Desmond J.; Peng, Yonghong (2016)The revolution in Big Data has opened the gate for new research challenges in biomedical science. The aim of this study was to investigate whether germ-line gene mutations are a significant factor in 29 major primary human cancers. Using data obtained from multiple biological databases, we identified 424 genes from 8879 cancer mutation records. By integrating these gene mutation records a human cancer map was constructed from which several key results were obtained. These include the observations that missense/nonsense and regulatory mutations might play central role in connecting cancers/genes, and tend to be distributed in all chromosomes. This suggests that, of all mutation classes missense/nonsense and regulatory mutation classes are over-expressed in human genome and so are likely to have a significant impact on human cancer aetiology and pathomechanism. This offers new insights into how the distribution and interconnections of gene mutations influence the development of cancers.
-
Stem Cell Organoids in Primary Cultures of Human Non-Malignant and Malignant ColonTariq, S.; Tahseen, M.; Hassan, M.; Masood, M.A.; Khattak, S.; Syed, A.A.; Ahmad, A.H.; Hussain, M.; Yusuf, M.A.; Sutton, Chris W. (2017-05-28)A sub-population of cells named cancer stem cells (CSCs) that initiate and promote tumour growth have been demonstrated to exist in several malignancies including colon carcinoma. The objective of our pilot study was to isolate CD133+CD26+CD44+ CSCs from patient colon tumours, culture spheres or organoids and observe their proliferation in primary cultures. Parallel cultures of non-cancer controls from colon normal lining and nonadenomatous polyps were set up. Magnetic activated cell sorting was used to isolate CD133+CD26+CD44+ cell populations followed by primary cell culturing under stem cell culture conditions. Number, cells/organoid and daughter generations of organoids were calculated using phase contrast microscope. Trypan blue exclusion method was used to test the viability of the cells. Both colon tumour and colon non-adenomatous polyp formed floating organoids in suspension; however non-adenomatous polyp cultures did not show self-renewal properties for more than 1 passage. Normal colon singlecell suspension did not create organoids. Metastatic colon tumours rapidly produce cancer cell organoids in less than 24 hours in larger numbers compared to non-metastatic colon tumours (1-3 weeks). Metastatic colon tumour organoids have the ability for proliferation for upto five daughter generations in primary culture compared to three generations for those grown from non-metastatic tumours. This in vitro CSC organoid model will help study colon cancer biology, in particular providing a valuable source of primary cell-derived tissue for studying personalized molecular profiling using ‘omics strategies to direct therapeutic intervention.