Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Methods for Processing and Analysis of Biomedical TEM Images
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Computerized Image Analysis and Human-Computer Interaction. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Visual Information and Interaction. (Quantitative Microscopy Wählby Lab)
2019 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Transmission Electron Microscopy (TEM) has the high resolving capability and high clinical significance; however, the current manual diagnostic procedure using TEM is complicated and time-consuming, requiring rarely available expertise for analyzing TEM images of the biological specimen. This thesis addresses the bottlenecks of TEM-based analysis by proposing image analysis methods to automate and improve critical time-consuming steps of currently manual diagnostic procedures. The automation is demonstrated on the computer-assisted diagnosis of Primary Ciliary Dyskinesia (PCD), a genetic condition for which TEM analysis is considered the gold standard.

The methods proposed for the automated workflow mimic the manual procedure performed by the pathologists to detect objects of interest – diagnostically relevant cilia instances – followed by a computational step to combine information from multiple detected objects to enhance the important structural details. The workflow includes an approach for efficient search through a sample to identify objects and locate areas with a high density of objects of interest in low-resolution images, to perform high-resolution imaging of the identified areas. Subsequently, high-quality objects in high-resolution images are detected, processed, and the extracted information is combined to enhance structural details.

This thesis also addresses the challenges typical for TEM imaging, such as sample drift and deformation, or damage due to high electron dose for long exposure times. Two alternative paths are investigated: (i) different strategies combining short exposure imaging with suitable denoising techniques, including conventional approaches and a proposed deep learning based method, are explored; (ii) conventional interpolation approaches and a proposed deep learning based method are analyzed for super-resolution reconstruction using a single image. For both explored directions, in the best case scenario, the processing time is nearly 20 times faster as compared to the acquisition time for a single long exposure high illumination image. Moreover, the reconstruction approach (ii) requires nearly 16 times lesser data (storage space) and overcomes the need for high-resolution image acquisition.

Finally, the thesis addresses critical needs to enable objective and reliable evaluation of TEM image denoising approaches. A method for synthesizing realistic noise-free TEM reference images is proposed, and a denoising benchmark dataset is generated and made publicly available. The proposed dataset consists of noise-free references along with masks encompassing the critical diagnostic structures. This enables performance evaluation based on the capability of denoising methods to preserve structural details, instead of merely grading them based on the signal to noise ratio improvement and preservation of gross structures.

Place, publisher, year, edition, pages
Uppsala: Acta Universitatis Upsaliensis, 2019. , p. 49
Series
Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology, ISSN 1651-6214 ; 1807
Keywords [en]
image analysis, image processing, deep learning, transmission electron microscopy, denoising, super-resolution reconstruction, registration, detection
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Computerized Image Processing
Identifiers
URN: urn:nbn:se:uu:diva-381800ISBN: 978-91-513-0653-7 (print)OAI: oai:DiVA.org:uu-381800DiVA, id: diva2:1305491
Public defence
2019-06-05, Room 2446, ITC, Lägerhyddsvägen 2, Uppsala, 10:15 (English)
Opponent
Supervisors
Available from: 2019-05-15 Created: 2019-04-17 Last updated: 2019-06-18
List of papers
1. Automated detection of cilia in low magnification transmission electron microscopy images using template matching
Open this publication in new window or tab >>Automated detection of cilia in low magnification transmission electron microscopy images using template matching
Show others...
2016 (English)In: Biomedical Imaging (ISBI), 2016 IEEE 13th International Symposium on, IEEE, 2016, p. 386-390Conference paper, Published paper (Other academic)
Abstract [en]

Ultrastructural analysis using Transmission Electron Microscopy (TEM) is a common approach for diagnosing primary ciliary dyskinesia. The manually performed diagnostic procedure is time consuming and subjective, and automation of the process is highly desirable. We aim at automating the search for plausible cilia instances in images at low magnification, followed by acquisition of high magnification images of regions with detected cilia for further analysis. This paper presents a template matching based method for automated detection of cilia objects in low magnification TEM images, where object radii do not exceed 10 pixels. We evaluate the performance of a series of synthetic templates generated for this purpose by comparing automated detection with results manually created by an expert pathologist. The best template achieves a detection at equal error rate of 47% which suffices to identify densely populated cilia regions suitable for high magnification imaging.

Place, publisher, year, edition, pages
IEEE, 2016
Series
IEEE International Symposium on Biomedical Imaging, ISSN 1945-7928
Keywords
Image resolution, Transmission Electron Microscopy, Object detection, Shape, Image analysis, Template matching
National Category
Computer Vision and Robotics (Autonomous Systems)
Research subject
Computerized Image Processing; Computerized Image Analysis
Identifiers
urn:nbn:se:uu:diva-308090 (URN)10.1109/ISBI.2016.7493289 (DOI)000386377400093 ()9781479923496 (ISBN)9781479923502 (ISBN)
Conference
IEEE 13th International Symposium on Biomedical Imaging (ISBI), 2016
Available from: 2016-11-23 Created: 2016-11-23 Last updated: 2019-04-17Bibliographically approved
2. Convolutional neural networks for false positive reduction of automatically detected cilia in low magnification TEM images
Open this publication in new window or tab >>Convolutional neural networks for false positive reduction of automatically detected cilia in low magnification TEM images
Show others...
2017 (English)In: Image Analysis: Part I, Springer, 2017, p. 407-418Conference paper, Published paper (Refereed)
Abstract [en]

Automated detection of cilia in low magnification transmission electron microscopy images is a central task in the quest to relieve the pathologists in the manual, time consuming and subjective diagnostic procedure. However, automation of the process, specifically in low magnification, is challenging due to the similar characteristics of non-cilia candidates. In this paper, a convolutional neural network classifier is proposed to further reduce the false positives detected by a previously presented template matching method. Adding the proposed convolutional neural network increases the area under Precision-Recall curve from 0.42 to 0.71, and significantly reduces the number of false positive objects.

Place, publisher, year, edition, pages
Springer, 2017
Series
Lecture Notes in Computer Science, ISSN 0302-9743 ; 10269
National Category
Computer Vision and Robotics (Autonomous Systems)
Research subject
Computerized Image Processing
Identifiers
urn:nbn:se:uu:diva-334218 (URN)10.1007/978-3-319-59126-1_34 (DOI)000454359300034 ()978-3-319-59125-4 (ISBN)
Conference
SCIA 2017, June 12–14, Tromsø, Norway
Funder
VINNOVA, 2016-02329
Available from: 2017-05-19 Created: 2017-11-21 Last updated: 2019-04-17Bibliographically approved
3. Enhancement of cilia sub-structures by multiple instance registration and super-resolution reconstruction
Open this publication in new window or tab >>Enhancement of cilia sub-structures by multiple instance registration and super-resolution reconstruction
Show others...
2017 (English)In: Image Analysis: Part II, Springer, 2017, p. 362-374Conference paper, Published paper (Refereed)
Abstract [en]

Ultrastructural analysis of cilia cross-sectional images using transmission electron microscopy (TEM) assists the pathologists to diagnose Primary Ciliary Dyskinesia, a genetic disease. The current diagnostic procedure is manual and difficult because of poor signal-to-noise ratio in TEM images. In this paper, we propose an automated multi-step registration approach to register many cilia cross-sectional instances. The novelty of the work is in the utilization of customized weight masks at each registration step to achieve good alignment of the specific cilium regions. Registration is followed by super-resolution reconstruction to enhance the substructural information. Landmarks matching based evaluation of registration results in pixel alignment error of 2.35±1.82" role="presentation">2.35±1.82 pixels, and the subjective analysis of super-resolution reconstructed cilium shows a clear improvement in the visibility of the substructures such as dynein arms, radial spokes, and central pair.

Place, publisher, year, edition, pages
Springer, 2017
Series
Lecture Notes in Computer Science, ISSN 0302-9743 ; 10270
National Category
Medical Image Processing
Research subject
Computerized Image Processing
Identifiers
urn:nbn:se:uu:diva-334225 (URN)10.1007/978-3-319-59129-2_31 (DOI)000454360300031 ()978-3-319-59128-5 (ISBN)
Conference
SCIA 2017, June 12–14, Tromsø, Norway
Available from: 2017-05-19 Created: 2017-11-21 Last updated: 2019-04-17Bibliographically approved
4. Automated Transmission Electron Microscopy to Assist Primary Ciliary Dyskinesia Diagnosis
Open this publication in new window or tab >>Automated Transmission Electron Microscopy to Assist Primary Ciliary Dyskinesia Diagnosis
(English)In: Article in journal (Refereed) Submitted
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:uu:diva-381799 (URN)
Available from: 2019-04-13 Created: 2019-04-13 Last updated: 2019-04-25
5. Denoising of short exposure transmission electron microscopy images for ultrastructural enhancement
Open this publication in new window or tab >>Denoising of short exposure transmission electron microscopy images for ultrastructural enhancement
Show others...
2018 (English)In: Proc. 15th International Symposium on Biomedical Imaging, IEEE, 2018, p. 921-925Conference paper, Published paper (Refereed)
Place, publisher, year, edition, pages
IEEE, 2018
National Category
Medical Image Processing
Research subject
Computerized Image Processing
Identifiers
urn:nbn:se:uu:diva-367040 (URN)10.1109/ISBI.2018.8363721 (DOI)000455045600210 ()978-1-5386-3636-7 (ISBN)
Conference
ISBI 2018, April 4–7, Washington, DC
Available from: 2018-11-27 Created: 2018-11-27 Last updated: 2019-04-17Bibliographically approved
6. Denoising Benchmark Dataset for Transmission Electron Microscopy Images
Open this publication in new window or tab >>Denoising Benchmark Dataset for Transmission Electron Microscopy Images
(English)In: Article in journal (Refereed) Submitted
Abstract [en]

Benchmark datasets that present the typical characteristics

of the imaging modality and provide noise-free references

are critical for meaningful evaluation and selection of suitable

denoising methods. Image denoising is a well-studied problem

in consumer cameras, and many denoising benchmark datasets

are available enhancing the development and evaluation of new

methods for natural scene photography. For biomedical applications

in general, and Transmission Electron Microscopy (TEM)

in particular, very few publicly available denoising benchmark

datasets exist. In TEM, it is difficult to acquire real noisefree

reference images; thus this paper presents an approach

for combining multiple real object images to synthesize realistic

noise-free images. In addition, a dataset of 60 synthetic cilia crosssection

TEM images is presented that can serve as noise-free

references to facilitate quantitative and qualitative evaluation of

denoising techniques targeted towards structural ultramicroscopy

applications. The dataset additionally defines masks for features

of interest for ultrastructural analysis of cilia. Such masks enable

performance evaluation focusing on the ability of the denoising

techniques to restore/preserve fine details of specific interest. The

paper ultimately presents a study comparing four commonly used

denoising techniques at four characteristic noise levels observed

at different exposure settings of TEM. The results demonstrate

the usefulness of the proposed dataset in the selection of a suitable

denoising method.

National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:uu:diva-381798 (URN)
Available from: 2019-04-13 Created: 2019-04-13 Last updated: 2019-04-17
7. Super-resolution Reconstruction of Transmission Electron Microscopy Images using Deep Learning
Open this publication in new window or tab >>Super-resolution Reconstruction of Transmission Electron Microscopy Images using Deep Learning
2019 (English)In: 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), IEEE, 2019, p. 548-551Conference paper, Published paper (Refereed)
Abstract [en]

Deep learning techniques have shown promising outcomes in single image super-resolution (SR) reconstruction from noisy and blurry low resolution data. The SR reconstruction can cater the fundamental. limitations of transmission electron microscopy (TEM) imaging to potentially attain a balance among the trade-offs like imaging-speed, spatial/temporal resolution, and dose/exposure-time, which is often difficult to achieve simultaneously otherwise. In this work, we present a convolutional neural network (CNN) model, utilizing both local and global skip connections, aiming for 4 x SR reconstruction of TEM images. We used exact image pairs of a calibration grid to generate our training and independent testing datasets. The results are compared and discussed using models trained on synthetic (downsampled) and real data from the calibration grid. We also compare the variants of the proposed network with well-known classical interpolations techniques. Finally, we investigate the domain adaptation capacity of the CNN-based model by testing it on TEM images of a cilia sample, having different image characteristics as compared to the calibration-grid.

Place, publisher, year, edition, pages
IEEE, 2019
Series
Biomedical Imaging, IEEE International Symposium on, E-ISSN 1945-7928
National Category
Computer Vision and Robotics (Autonomous Systems)
Identifiers
urn:nbn:se:uu:diva-381797 (URN)10.1109/ISBI.2019.8759153 (DOI)000485040000121 ()978-1-5386-3641-1 (ISBN)
Conference
16th IEEE International Symposium on Biomedical Imaging (ISBI), APR 08-11, 2019, Venice, ITALY
Funder
Swedish Foundation for Strategic Research , SB16-0046Vinnova, 2016-02329
Available from: 2019-04-13 Created: 2019-04-13 Last updated: 2019-10-23Bibliographically approved

Open Access in DiVA

fulltext(2096 kB)77 downloads
File information
File name FULLTEXT01.pdfFile size 2096 kBChecksum SHA-512
704dd49f9a97ec4dffff7cce8e8eaedae2e4e9f9d16597398acceccdf8f5442a3791ffc56f9c79639e4695f2d31bd6fa4dbf73ec7011bb2507325fea366d20cc
Type fulltextMimetype application/pdf
Buy this publication >>

Search in DiVA

By author/editor
Suveer, Amit
By organisation
Computerized Image Analysis and Human-Computer InteractionDivision of Visual Information and Interaction
Other Electrical Engineering, Electronic Engineering, Information Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 77 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 1036 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf