Morphological segmentation of histology cell images

Nedzved, A.; Ablameyko, Sergey; Pitas, Ioannis

doi:10.1109/icpr.2000.905385

Cited by 57 publications

(44 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A comprehensive analysis where architectural features could be seen at low and intermediate magnification (up to 200Â) and cytologic features could be seen at high magnification (400Â) could not be performed. Because of these limitations, working on single images instead, previous studies of image analysis concentrated on specific problems of image analysis like nuclear detection (20), detection of immunopositivity (8,9), quantification of vascularisation (35), mitotic counting (36) in high resolution.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Automated virtual microscopy of gastric biopsies

Ficsor

Varga

Berczi

et al. 2006

Cytometry Part B Clinical

View full text Add to dashboard Cite

Background: Automated virtual microscopy of specimens from gastrointestinal biopsies is based on cytometric parameters of digitized histological sections. To our knowledge, cytometric parameters of gastritis and of adenocarcinoma have yet to be fully characterized. Our objective was to classify gastritis and adenocarcinoma based on cytometric parameters. We hypothesized that automated virtual microscopy using this novel classification can reliably diagnose gastritis and adenocarcinoma.Methods: Routinely processed hematoxylin-and-eosin-stained histological sections from specimens that showed normal mucosa (14 cases), gastritis (35 cases), and adenocarcinoma (30 cases) diagnosed by conventional optical microscopy were scanned and digitized at high resolution. Thirty-eight cytometric parameters based on density and morphometry were applied to glands and superficial epithelium. Twelve cytometric parameters based on cytologic detail were applied to individual cells.Results: Statistically significant differences in cytometric parameters for normal mucosa, gastritis, and adenocarcinoma were found. The most discriminatory parameter was the ratio of the total number of cells to the number of interstitial cells. These differences correctly classified adenocarcinoma at 100% accuracy and overall correctness was 86%.Conclusions: We describe a novel method of analyzing gastric mucosal histology based on cytometric parameters. Automated virtual microscopy can be used to classify gastric mucosa as normal, gastritis, or adenocarcinoma with reasonable accuracy. Further research is necessary to determine whether automated virtual microscopy can subclassify gastric mucosal histology in greater detail. q 2006 International

show abstract

Section: Discussionmentioning

confidence: 99%

“…These artifacts represent noise that automated analysis must ignore during the final interpretation. Despite these obstacles, recent studies demonstrate highly effective automated analysis of histological sections, including the detection of carcinoma (19,20). Most studies have focused on routinely-processed hematoxylin-and-eosin-stained sections.…”

mentioning

confidence: 99%

Automated virtual microscopy of gastric biopsies

Ficsor

Varga

Berczi

et al. 2006

Cytometry Part B Clinical

View full text Add to dashboard Cite

show abstract

“…However, most existing research on cell image segmentation has focused on identifying cells in a noisy background (Garbay et al, 1986;Jiang and Yang, 2002;Nedzved et al, 2000) or solely on nucleus isolation and evaluation (Sammouda et al, 2002;Thiran and Macq, 1996;Schnorrenberg et al, 1997). Sammouda et al (2002) work with cell images similar to ours to detect cancer; however, that work focused only on nucleus segmentation and analysis, whereas we extract information from the entire cell image.…”

Section: Related Researchmentioning

confidence: 99%

A kernelised fuzzy-Support Vector Machine CAD system for the diagnosis of lung cancer from tissue images

Land¹,

McKee²,

Zhukov³

et al. 2008

IJFIPM

View full text Add to dashboard Cite

This research describes a non-interactive process that applies several forms of computational intelligence to classifying biopsy lung tissue samples. Three types of lung cancer evaluated (squamous cell carcinoma, adenocarcinoma, and bronchioalveolar carcinoma) together account for 65-70% of diagnoses. Accuracy achieved supports hypothesis that an accurate predictive model is generated from training images, and performance achieved is an accurate baseline for the process's potential scaling to larger datasets. Tatyana Zhukov is an Assistant Professor, Cancer Prevention and Control Division at the H. Lee Moffitt Cancer Center and Research Institute. He was trained in clinical biochemistry and clinical cytology and has worked for the last ten years to apply these skills in the development of molecular markers for human cancer. Her professional insight into cell image cytometry, proteomics, and biomarker development has led to substantial advances in methods used and proposed for lung and breast cancer screening largely through her participation in the Early Detection Research Network (EDRN).Dansheng Song is a Research Associate at the Department of Interdisciplinary Oncology, College of Medicine, University of South Florida. He has extensive experience over the last ten years in the applications of advanced Computer-Aided Diagnosis (CAD) methods on medical imaging, biomedical imaging and molecular imaging to clinical trials. He has published more than 20 paper in these field.Wei Qian is an Associate Professor, Director of the Biomedical Imaging Program at the Department of Interdisciplinary Oncology and Radiology, College of Medicine, University of South Florida. He has extensive experience over the last ten years in the applications of advanced Computer-Aided Diagnosis (CAD) methods on medical imaging, biomedical imaging and molecular imaging to clinical trials. He has published more than 100 papers in these field.

show abstract

“…In order to solve problem 2 (undesired white pixels in the likelihood image), we clean the image using morphological operators as in [6,7,8]. We apply image reconstruction [9] as expressed in Equation (2), using an eroded version of our image as the marker.…”

Section: B Likelihood Imagementioning

confidence: 99%

Automatic Segmentation of Arteries in Multi-stain Histology Images

et al. 2009

View full text Add to dashboard Cite

Abstract-Atherosclerosis is a very common disease that affects millions of people around the world. Currently most of the studies conducted on this disease use Ultrasound Imaging (IVUS) to observe plaque formation, but these images cannot provide any detailed information of the specific morphological features of the plaque. Microscopic imaging using a variety of stains can provide much more information although, in order to obtain proper results, millions of images must be analyzed. We provide an automatic way to find the Region of Interest (ROI) of these images, where the atherosclerotic plaque is formed. Once the image is well-segmented the amount of fat and other measurements of interest can also be determined automatically. A proper study of these images provides valuable information that can be key to understanding the evolution of atherosclerosis.

show abstract

Morphological segmentation of histology cell images

Cited by 57 publications

References 4 publications

Automated virtual microscopy of gastric biopsies

Automated virtual microscopy of gastric biopsies

A kernelised fuzzy-Support Vector Machine CAD system for the diagnosis of lung cancer from tissue images

Automatic Segmentation of Arteries in Multi-stain Histology Images

Contact Info

Product

Resources

About