A Novel Framework for Segmentation of Secretory Granules in Electron Micrographs |
| |
| Affiliation: | 1. Visual Information Laboratory, University of Bristol, Merchant Venturers Building, Woodland Road, Bristol BS8 1UB, UK;2. Wolfson Bioimaging Facility, University of Bristol, Medical Sciences, University Walk, Bristol BS8 1TD, UK;3. School of Biochemistry, University of Bristol, Medical Sciences, University Walk, Bristol BS8 1TD, UK;4. School of Physiology and Pharmacology, University of Bristol, Medical Sciences, University Walk, Bristol BS8 1TD, UK;1. Primary Ciliary Dyskinesia Centre, NIHR Biomedical Research Centre, University Hospital Southampton NHS Foundation Trust, and Academic Unit of Clinical and Experimental Medicine, University of Southampton, Southampton, England;2. Primary Ciliary Dyskinesia Centre, NIHR Biomedical Research Centre, University Hospital Southampton, Southampton, England;3. Department of Pediatric Pulmonology, VU University Medical Center, Amsterdam, the Netherlands;4. Department of Infection, Immunity and Inflammation, Centre for PCD Diagnosis and Research, University of Leicester, Leicester, England;5. Departments of Paediatrics and Paediatric Respiratory Medicine, Primary Ciliary Dyskinesia Centre, Imperial College and Royal Brompton Hospital, London, England;6. Danish PCD & chILD Centre, CF Centre Copenhagen Paediatric Pulmonary Service, and Department of Paediatrics and Adolescent Medicine, Copenhagen University Hospital, Southampton, England;7. Department of Pediatrics, University Hospital Muenster, Muenster, Germany;8. ENT Department, APHP, Bicetre University Hospital, Paris, France;9. PCD service, Department of Respiratory and Sleep Medicine, Royal Children’s Hospital, Melbourne, Australia;10. Departments of Paediatrics and Paediatric Respiratory Medicine, Primary Ciliary Dyskinesia Centre, Imperial College and Royal Brompton Hospital, and School of Medicine, University of Dundee, Dundee, Scotland;1. Acoustic Research Laboratory, Tropical Marine Science Institute, National University of Singapore, 119227 Singapore;2. Department of Electrical and Computer Engineering, Block E4, 4 Engineering Drive 3, 117583 Singapore;1. KU Leuven and UZ Leuven, Department of Pediatric Nephrology & Growth and Regeneration, Leuven, Belgium;2. Cairo University, Faculty of Medicine, Department of Clinical and Chemical Pathology, Cairo, Egypt;3. KU Leuven, Laboratory of Molecular and Cellular Signalling, Department of Cellular and Molecular Medicine, Leuven, Belgium;4. Radboud University Medical Center, Department of Pediatric Nephrology, Nijmegen, The Netherlands |
| |
| Abstract: | It is still a standard practice for biologists to manually analyze transmission electron microscopy images. This is not only time consuming but also not reproducible and prone to induce subjective bias. For large-scale studies of insulin granules inside beta cells of the islet of Langerhans, an automated method for analysis is essential. Due to the complex structure of the images, standard microscopy segmentation techniques cannot be applied. We present a new approach to segment and measure transmission electron microscopy images of insulin granule cores and membranes from beta cells of rat islets of Langerhans. The algorithm is separated into two broad components, core segmentation and membrane segmentation. Core segmentation proceeds through three steps: pre-segmentation using a novel level-set active contour, morphological cleaning and a refining segmentation on each granule using a novel dual level-set active contour. Membrane segmentation is achieved in four steps: morphological cleaning, membrane sampling and scaling, vector field convolution for gap filling and membrane verification using a novel convergence filter. We show results from our algorithm alongside popular microscopy segmentation methods; the advantages of our method are demonstrated. Our algorithm is validated by comparing automated results to a manually defined ground truth. When the number of granules detected is compared to the number of granules in the ground truth a precision of 91% and recall of 87% is observed. The average granule areas differ by 13.35% and 6.08% for core and membranes respectively, when compared to the average areas of the ground truth. These results compare favorably to previously published data. |
| |
| Keywords: | Transmission electron microscopy Active contours Convergence filters Image segmentation |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
