• Dr. M. Siyamalan

    Senior Lecturer Gr.II
    B.Sc. [Hons] (Jaffna), MSc(France), PhD(UK)
  • siyam@univ.jfn.ac.lk
  • +94212218194
  • LinkedIn
  • Google Scholar
  • Research Interests: Deep learning, Machine Learning, Computer Vision, Medical Image Analysis
  • Ph.D. (UK) [2015]
  • M.Sc. (France) [2011]
  • B.Sc. Special in Computer Science, First Class (Jaffna, Sri Lanka) [2006]
Postgraduate Supervision

MSc

  • Mr. V. E.Nirmalan, “Comparative Analysis of Different Features and Encoding Methods for Rice Image Classification”, MSc in Computer Science, PGIS, University of Peradeniya,

MPhil

  • Ms. R. Nirthika, “Convolutional Neural Network-based Semi-Supervised Learning for Medical Image Analysis”, Reading for MPhil in Computer Science, University of Jaffna, Nov. 2018 – Nov. 2020. [Supervisor: Dr. M. Siyamalan, Co-Supervisor: Dr. A. Ramanan]
In overall: Deep learning, Machine Learning, Computer Vision, Medical Image Analysis

Example projects:

Multiple-Instance Learning for Retinal Nerve Fiber Layer Visibility Classification

Introduction and motivation: The optic nerve transmits visual information from the retina to the brain. The expansion of the neural fibers in the optic nerve enters the retina at the optic disc. Its form the Retinal Nerve Fiber Layer (RNFL), the innermost retinal layer (Figure 1). The RNFL has been implicated in prediagnostic stages of glaucoma and recently considered as a potential biomarker for dementia, by assessing its thickness in optical confocal tomography (OCT) images. However, RNFL is not always visible in fundus camera (FC) images. Can it’s visibility in FC images a potential biomarker for dementia?
Objective: Automatically segment RNFL visible regions (if any) in the FC images using weakly supervised learning approaches.

Example results: Experiments with a RNFL dataset containing 884 images annotated by two ophthalmologists give a system-annotator agreement (kappa values) of 0:73 and 0:72 respectively, with an inter-annotator agreement of 0:73. Comparative tests with three public datasets (MESSIDOR and DR for diabetic retinopathy, UCSB for breast cancer) show that our novel MIL approach improves performance over the state-of-the-art results.



Related publications:

  • S. Manivannan et al., “Sub-Category Classifiers for Multiple-Instance Learning and its Application to Retinal Nerve Fiber Layer Visibility Classification”, IEEE Transactions on Medical Imaging, 2017.
  • S. Manivannan et al., “Sub-Category Classifiers for Multiple-Instance Learning and its Application to Retinal Nerve Fiber Layer Visibility Classification”, International Conference on Medical Image Computing and Computed Assisted Interventions, 2016.
Local structure prediction for gland segmentation

Introduction and motivation: Histological assessment of gland formation and morphology informs diagnosis, prognosis and treatment planning of patients. It is useful for grading of adenocarcinomas in colon, breast, and prostate. Such assessment is labour intensive, performed by highly trained pathologists, and often has limited reproducibility. The emergence of whole-slide imaging is increasing the volume of digital histology image data to be analysed, exacerbating the problem. Algorithms capable of reliably segmenting glandular structures automatically would accelerate analysis and provide reproducible, quantitative measures of gland morphology.

Objective: Automatically segment individual glands from colon histology images.

Results: Direct comparison with other published results indicates that our method is the top ranked.

Related publications:

  • S. Manivannan, W. Li, J. Zhang, E. Trucco, S. J. McKenna, “Structure Prediction for Gland Segmentation with Hand-Crafted and Deep Convolutional Features”, Transactions on Medical Imaging, 2017
  • S. Manivannan, W. Li, S. Akbar, J. Zhang, E. Trucco, S. J. McKenna, “Local structure prediction for gland segmentation”, in International Symposium on Biomedical Imaging, 2016
  • W. Li, S. Manivannan, J. Zhang, E. Trucco, S. J. McKenna, “Gland segmentation in colon histology images using hand-crafted features and convolutional neural networks”, International Symposium on Biomedical Imaging, 2016.
Undergraduate Teaching
  • CSC107S2 & CSC103G2: Multimedia Technologies
  • CSC236SC2 & CSC234GC2: Operating Systems
  • CSC316GE2: Introduction to Systems and Network Administration
  • CSC421SC3 & CSC421MC3: Systems Analysis, Design and Project Management
Previous Teaching Experience (Faculty of Engineering, University of Jaffna)
  • EC6050: Computer Architecture and Organization
  • EC9560: Data Mining
  • EC2010: Computer Programming
  • EC5060: Operating Systems
Invited Publications (2)
    2014
  • S.Manivannan, W.Li, S.Akbar, R.Wang, J.Zhang and S.J.McKenna*, HEp-2 specimen classification using multi-resolution local patterns and SVM, 1st Workshop on Pattern Recognition Techniques for Indirect Immunoflurescence Images, International Conference on Pattern Recognition, 2014, 41-44 [PDF]
  • S.Manivannan, W.Li, S.Akbar, R.Wang, J.Zhang and S.J.McKenna*, HEp-2 cell classification using multi-resolution local patterns and ensemble SVMs, 1st Workshop on Pattern Recognition Techniques for Indirect Immunoflurescence Images, International Conference on Pattern Recognition, 2014, 37-49 [PDF]
International Journals (3)
    2018
  • S. Manivannan, W. Li, J. Zhang, E. Trucco and S. McKenna, Structure Prediction for Gland Segmentation with Hand-Crafted and Deep Convolutional Features, IEEE Transactions on Medical Imaging, 2018, 37(1):210-221. [PDF]
  • 2017
  • S.Manivannan, C.Cobb, S.Burgess and E.Trucco, Sub-Category Classifiers for Multiple-Instance Learning and its Application to Retinal Nerve Fiber Layer Visibility Classification, IEEE Transactions on Medical Imaging, 2017, 36(5):1140–1150 [PDF]
  • 2016
  • S.Manivannan, W.Li, S.Akbar, R.Wang, J.Zhang and S.J.McKenna, An automated pattern recognition system for classifying HEp-2 cells and specimens, In Pattern Recognition, 2016, 51:12-26[PDF]
International Conference and Workshop Papers (14)
    2019
  • R. Zhang, S. Tan, S.Manivannan, R. Wang, H. Lin, J. Chen, and W. Zheng, : “Biomarker Localization by Combining CNN Classifier and Generative Adversarial Network”, In the 22nd International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Shenzhen, 13-17 October 2019. [Accepted]
  • 2018
  • E.N. Vijayaratnam, R.D. Nawarathna, and S.Manivannan, Comparative Analysis of Different Features and Encoding Methods for Rice Image Classification, In IEEE International Conference on Information and Automation for Sustainability (ICIAfS), 2018.[PDF]
  • 2017
  • A.E. Fetit, S.Manivannan, S.McGrory, L.Ballerini, A.Doney, T.J. MacGillivray, I.J.Deary, J.M. Wardlaw, F.Doubal, G.J. McKay, S.J. McKenna and E.Trucco, Retinal Biomarker Discovery for Dementia in an Elderly Diabetic Population, In Ophthalmic Medical Image Analysis – MICCAI, 2017, 150-158[PDF]
  • 2016
  • S.Manivannan, R.Wang and E.Trucco, Hierarchical mix-pooling and its applications to biomedical image classification, In IEEE 13th International Symposium on Biomedical Imaging, 2016, 541 – 544[PDF]
  • S.Manivannan, W.Li, S.Akbar, J.Zhang, E.Trucco and S.J.McKenna, Local structure prediction for gland segmentation, In IEEE 13th International Symposium on Biomedical Imaging, 2016, 799 – 802[PDF]
  • W.Li, S.Manivannan, S.Akbar, J.Zhang, E.Trucco and S.J.McKenna, Gland segmentation in colon histology images using hand-crafted features and convolutional neural networks, In IEEE 13th International Symposium on Biomedical Imaging, 2016, 1405-1408[PDF]
  • S.Manivannan and E.Trucco, Extended Multi-resolution Local Patterns – A Discriminative Feature Learning Approach for Colonoscopy Image Classification, In Computer Assisted Robotic Endoscopy – MICCAI, 2016, 48-58[PDF]
  • S.Manivannan, C.Cobb, S.Burgess and E.Trucco, Sub-Category Classifiers for Multiple-Instance Learning and its Application to Retinal Nerve Fiber Layer Visibility Classification, In Medical Image Computing and Computer-Assisted Intervention – MICCAI, 2016, 308-316 [PDF]
  • 2015
  • S.Manivannan and E.Trucco, Learning discriminative local features from image-level labelled data for colonoscopy image classification, IEEE 12th International Symposium on Biomedical Imaging, 2015, 420-423 [PDF]
  • S.Manivannan and E.Trucco, Weakly supervised multi-class feature learning for colonoscopy image classification, In SICSA Medical Image Analysis Workshop, 2015.
  • 2014
  • S.Manivannan, R.Wang and E.Trucco, Inter-cluster features for medical image classification, In Medical Image Computing and Computer-Assisted Intervention – MICCAI, 2014, 345-352[PDF]
  • G.A.Puerto-Souza, S.Manivannan, M.Trujillo, J.Hoyos, E.Trucco and G.Mariottini, Enhancing normal-abnormal classification accuracy in colonoscopy videos via temporal consistency, In Computer-Assisted Robotic Endoscopy, MICCAI, 2014, 129-139.[PDF]
  • S.Manivannan, R.Wang, M.Trujillo, J.Hoyos, E.Trucco and G.Mariottini, Video-specifc SVMs for colonoscopy image classification, In Computer-Assisted Robotic Endoscopy, MICCAI, 2014, 11-21[PDF]
  • S.Manivannan, H.Shen, W.Li, R.Annunziata, H.Hamad, R.Wang and J.Zhang, Brain tumor region segmentation using local co-occurrence features and conditional random fields, In Brain Tumour Digital Pathology Segmentation Challenge, MICCAI, 2014, 15-19[PDF]
  • 2013
  • S.Manivannan, R.Wang and E.Trucco, Extended Gaussian-Filtered Local Binary Patterns for Colonoscopy Image Classification, In 2013 IEEE International Conference on Computer Vision Workshops, 2013, 184-189 [PDF]
  • S.Manivannan, R.Wang, E.Trucco and A.Hood, Automatic normal-abnormal video frame classification for colonoscopy, In IEEE 10th International Symposium on Biomedical Imaging, 2013, 644-647[PDF]
International Contest winning
  • 2018: Member of the team which got the third place (out of 77 teams) at the MICCAI 2018 grand challenge “ISIC 2018: Skin Lesion Analysis Towards Melanoma Detection, Task 3 – Lesion Diagnosis”.
  • 2014: Member of the winning team for the challenge “Performance Evaluation of Indirect Immunofluorescence Image Analysis Systems” organized by ICPR, Stockholm, Sweden (over 18 teams participated worldwide).
  • 2014: Member of the team which got the runner-up prize for the “Brain Tumor Digital Pathology Segmentation Challenge” organized by MICCAI, Harvard Medical School, Boston, USA (first place won by Microsoft Research Asia).
  • 2015: Member of the winning team for the challenge “Early Barrett’s Cancer Detection” organized by MICCAI, Munich, Germany.
  • 2015: Member of the winning team for the challenge “Detection of Abnormalities in Gastroscopic Images” organized by MICCAI, Munich, Germany.
Scholarships
  • 2011-2015: PhD full International studentship funded by the European Funding Council under the IDEAS programme grant, “Colonic Disease Investigation by Robotic Hydro-colonoscopy (CODIR)”.
  • 2010-2011: MSc full International studentship awarded by French National Institute for Research in Computer Science and Control (INRIA).
  • 2001-2006: Mahapola Higher Education Scholarship awarded by the Government of Sri Lanka.
Academic Awards
  • 2017: Presidential Award for Scientific Research
  • 2016: Nominated for the young scientist award by the International conference on Medical Image Computing and Computer Assisted Interventions, 2016. One of the only 30 authors selected from an author pool of around 250 accepted papers, filtered in turn from about 1,000 submissions.
  • 2014: Runner-up prize for the outstanding doctoral student, School of Computing, University of Dundee, UK.
  • 2006: University Prize for having the best performance (highest average marks) at the fourth year examination in science, University of Jaffna.
  • 2006: Sabalingam Memorial Prize for Computer Science for having the highest average marks at the Special degree examination in Computer Science, University of Jaffna.
  • 2003: University Prize for having the best performance (highest average marks) at the second year examination in science, University of Jaffna.
  • 2003: Handy Perinbanayaham Memorial Prize for having the best performance in the first and second examinations in Science.
  • 2002: University Prize for having the best performance (highest average marks) at the first year examination in science, University of Jaffna.
  • 2002: Sabalingam Memorial Prize for having the highest average marks at the first year examination in Science, University of Jaffna.
Travel Awards
  • 2013: BMVA travel award, British Machine Vision Association, UK.
  • 2013: AISB travel award, The Society for the Artificial Intelligence and Simulation Behavior, UK.
  • 2013: Leng Trust Travel Award, College of Medicine, Dentistry and Nursing, University of Dundee.
Non-academic Awards and Recognitions
  • 2003–2005: University Chess Champion, University of Jaffna (3 consecutive years)
  • 2003–2005: University Colors Insignia for the best performance in Chess, University of Jaffna (3 consecutive years)
  • University of Jaffna Chess Team vice-captain in 2004 and captain in 2005
  • Represented the University of Jaffna Chess team in various International and National level tournaments including GACC World Inter-varsity chess championship held in Malaysia in 2003 and 2005
Other recognitions
  • Reviewer for various journals and conferences including Pattern Recognition Journal (reviewed over 12 papers) and IEEE Transactions on Medical Imaging (reviewed over 4 papers).
  • Outstanding contribution in reviewing by Pattern Recognition Journal– awarded for reviewers who are within the top 10th percentile of reviewers for the Journal, in terms of the number of manuscript reviews completed in 2015 and 2016.
May 2018 - Present Senior Lecturer - Department of Computer Science, Faculty of Science,University of Jaffna, Sri Lanka.
Responsibilities include conducting lectures for undergraduate students and supervising undergraduate and postgraduate student projects.
April 2019 - May 2019
March 2018 - April 2018
Visiting Researcher - Sohool of Data and Computer science, Sun Yat-Sen University, Guanzhou, China.
Focusing on recent research in Deep Learning.
Sept 2016 - April 2018 Senior Lecturer - Department of Computer Engineering, Faculty of Engineering,University of Jaffna, Sri Lanka.
Responsibilities include conducting lectures for undergraduate students and supervising undergraduate and postgraduate student projects.
April 2015 – August 2016 Post doctoral researcher - CVIP, School of Computing, University of Dundee, UK.
This work was mainly focused on identifying retinal bio-markers for vascular dementia. Main responsibility was to develop machine learning algorithms to automatically segment visible retinal nerve fiber layer regions (if any) in retinal color fundus images using weakly supervised approaches.
Nov. 2011 – April 2015 PhD researcher - CVIP, School of Computing, University of Dundee, UK.
This work was focussed on developing novel machine learning approaches for learning feature representations for medical image classification and lesion detection. It was also related to feature encoding approaches, multiple-instance learning, and deep-learning using convolutional neural networks.
April 2011 – August 2011 Research Intern - ASCLEPIOS Research Team, INRIA, France
Research and software development for fast segmentation of regions of interests in medical images
April 2008 – July 2010 Research and Development Software Engineer - Excel Technology Lanka (Pvt.) Ltd. Sri Lanka
Research and software development in the application of Image processing and 3D Computer graphics for laser related pre-processes.
Sept. 2007 – Feb. 2008 Assistant Lecturer - Department of Computer Science, University of Jaffna, Sri Lanka
Conducting Lectures, practical and tutorial classes for undergraduate students.
Feb. 2007 – Sept. 2007 Instructor - Department of Computer Science, University of Jaffna, Sri Lanka
Conducting practical and tutorial classes for undergraduate students.
Research Grants
  • 2017: University grant of worth Rs. 0.78 Million to purchase a high-performance computing system.
  • 2017: Received NVidia TitanXP GPU under the Small Scale GPU Grants Programs by NVidia Corporation, USA.
  • Member of the Career Guidance Cell of the Faculty of Science, 2019