About

Murat Seçkin Ayhan was born in Ankara, Turkey, where he received his B.S. and M.S. degrees in Computer Engineering from Başkent University in June 2004 and February 2007, respectively. In January 2008, he joined the University of Louisiana at Lafayette and received his M.S. degree in Computer Science in May 2010. From August 2010 to July 2012, he was a Graduate Fellow of the Louisiana Optical Network Initiative (LONI) Institute as well as a Ph.D. student at the Center for Advanced Computer Studies in Lafayette, Louisiana. In May 2015, he received his Ph.D. degree in Computer Science. Upon completion of his work, he joined the Department of Computer Engineering at Işık University, Istanbul, Turkey. Between Fall 2015 and Spring 2017 semesters, he taught undergraduates several courses including Analysis of Algorithms, Database Systems,  and Data Mining. At the graduate level, he taught Machine Learning. On top of the departmental courses, he was entertained with programming courses for non-majors: Introduction to Programming and Object-Oriented Programming with Java. 

In August 2017, Murat Seçkin Ayhan joined the Neural Data Science for Vision Research Group at the Institute for Ophthalmic Research, University of Tübingen, Germany. His research mainly aims at promoting the health and well-being of individuals and hence communities via machine learning. He also  maintains interest in algorithmic aspects of the field. His research interests lie in the span of the followings:

  • Machine Learning, Pattern Recognition, Data Mining, Descriptive and Predictive Modeling
  • Medical Image Analysis and Interpretable Deep Learning
  • Ensemble Methods and Anomaly Detection

Research

Alzheimer’s disease is a major cause of dementia. Its diagnosis €™requires accurate biomarkers that are sensitive to disease stages. Neuroimaging techniques such as PET and MRI have been used as imaging biomarkers. However, it is virtually impossible to visually detect a slight decrease in regional cerebral blood flow or glucose metabolism in early stages of the disease. Moreover, visual inspection is susceptible to other factors like subjectivity and experience of the physician. On the other hand, voxel-based representations of neuroimagery can be used to perform both standardization and data-driven analysis. Computerized methods can also improve the speed of diagnosis with no compromise of accuracy and facilitate accurate diagnosis in cases where there is a lack of access to an experienced physician.

During my graduate studies, I focused on predictive modeling of Alzheimer’s disease using the variations of high-dimensional neuroimaging data. Simply, I regarded the induction of a classifier as the creation of a computational biomarker for disease staging. In this regard, I investigated clever utilizations of a wide range of machine learning algorithms in order to get the best of them, given the challenges of working with the scarce and high-dimensional neuroimaging data. I also strove for simplicity and comprehensibility in these predictive models, which is highly valuable for clinical decision-making.

Projects

  • Data-driven Prognosis of Alzheimer’s Disease (Software Prototype), Patient Early Health Collaboration Project, 03/01/08 – 12/31/09,  Funding from GE Healthcare, PIs: Vijay V. Raghavan and C. H. Chu, Laboratory for Internet Computing, CACS, UL Lafayette. Research Assistant: Murat Seçkin Ayhan
  • Towards creating a Large and Scalable In-Memory Database Management System, State of Louisiana, Governor’s Information Technology Initiative Project, 07/01/07 – 07/31/09, PI: Vijay V. Raghavan, Laboratory for Internet Computing, CACSUL LafayetteResearch Assistant: Murat Seçkin Ayhan
  • Evliya Çelebi: A Middleware for Geographical Information, TÜBİTAK-SOBAG, 105K040, 2005, PI: Hayri Sever, Project completed at the Department of Computer Engineering, Baskent University, Research Assistant: Murat Seçkin Ayhan

Publication


Disclaimer: All the material listed and presented here is to promote the dissemination of scholarly and technical work in a timely fashion. All rights as well as the copyright are retained by authors and organizers/publishers.


Journal Articles

  1. Leveraging uncertainty information from deep neural networks for disease detection, C. Leibig, V. Allken, M.S. Ayhan, P. Berens, and S. Wahl [bioRxiv] [accepted]
  2. Multiple Kernel Learning and Automatic Subspace Relevance Determination for High-dimensional Neuroimaging Data, M.S. Ayhan, V.V. Raghavan and Alzheimer’s Disease Neuroimaging Initiative (ADNI) [arXiv]
  3. Exploitation of 3D Stereotactic Surface Projection for Predictive Modeling of Alzheimer’s Disease, M.S. Ayhan, R.G. Benton, V.V. Raghavan and S. Choubey, International Journal of Data Mining and Bioinformatics, 2013, Vol.7, No.2 [.pdf]

Conference and Workshop Papers

  1. Evaluation of Autoencoders for Bases to Represent Neuroimaging Data, A. Gupta, M.S. Ayhan, A.S. Maida, Workshop on Machine Learning and Interpretation in NeuroImaging, The 27th Annual Conference on Neural Information Processing Systems
    December 2013, Lake Tahoe, NV, USA [.pdf]
  2. Composite Kernels for Automatic Relevance Determination in Computerized Diagnosis of Alzheimer’s Disease, M.S. Ayhan, R.G. Benton, V.V. Raghavan and S. Choubey, The 2013 International Conference on Brain and Health Informatics
    October 2013, Maebashi City, GUNMA, JAPAN [.pdf]
  3. Natural Image Bases to Represent Neuroimaging Data, A. Gupta, M.S. Ayhan, A.S. Maida, Deep Learning and Neuroscience Track, The 30th International Conference on Machine Learning
    June 2013, Atlanta, GA, USA [.pdf] [supplement]
  4. Towards Indefinite Gaussian Processes, M.S. Ayhan, C.H. Chu, The Modern Nonparametric Methods in Machine Learning Workshop The 26th Annual Conference on Neural Information Processing Systems
    December 2012, Lake Tahoe, NV, USA [.pdf]
  5. Utilization of Domain-Knowledge for Simplicity and Comprehensibility in Predictive Modeling of Alzheimer’s Disease, M.S. Ayhan, R.G. Benton, V.V. Raghavan and S. Choubey, International Workshop on Multiscale Biomedical Imaging Analysis
    The IEEE International Conference on Bioinformatics and Biomedicine 2012
    October, 2012, Philadelphia, PA, USA [.pdf]
  6. Exploitation of 3D Stereotactic Surface Projection for Automated Classification of Alzheimer’s Disease According to Dementia Levels, M.S. Ayhan, R.G. Benton, V.V. Raghavan and S. Choubey, The IEEE International Conference on Bioinformatics and Biomedicine 2010 [.pdf]
    December, 2010, Hong Kong, SAR, CHINA
  7. Determining Relevant Features Based on 3D Stereotactic Surface Projection to Detect Dementia Caused by Alzheimer’s Disease, M.S. Ayhan, R.G. Benton, V.V. Raghavan and S. Choubey
    The 7th Annual Biotechnology and Bioinformatics Symposium (extended abstract)
    October 2010, Lafayette, LA, USA
  8. Comparison of Spatial Indexing Methods, M.S Ayhan, H. Sever, H. Gurcay and S. Ak
    The National Conference on Geographical Information Systems (in Turkish)
    November 2007, Trabzon, TURKEY

Theses

  1. A Probabilistic Biomarker for Alzheimer’s Disease, Dissertation, University of Louisiana at Lafayette, Advisor: Vijay Raghavan, May 2015
  2. Comparison of Spatial Indexing Methods, Master’s thesis, Baskent University, Advisor: Hayri Sever, February 2007


Work in Progress

  1. Learning Distributed Representations from the Relevant Regions of Brain via Autoencoders, Title says it all 🙂
  2. Indefinite Gaussian Processes, Follow-up to the NIPS 2012 workshop paper. This time, working on a joint optimization procedure to accomplish the Gaussian Process learning and kernel transformation/spectrum modification simultaneously.

Teaching

Courses taught at Isik University, Istanbul between September 2015 and July 2017

Spring 2017

CSE490 – Senior Year Project (3 students)
CSE566 – Machine Learning, Graduate-level (11 students)
CSE222 – Database Systems, 2 sections (57 students in total (29+28))


Fall 2016

CSE490 – Senior Year Project (1 student)
CSE485 – Data Mining (22 students)
CSE312 – Analysis of Algorithms (39 students)
CSE203 – Object Oriented Programming with Java (53 students)


Spring 2016

CSE490 – Senior Year Project (3 students)
CSE566 – Machine Learning, Graduate-level (9 students)
CSE222 – Database Systems, 2 sections (60 students in total (27+33))


Fall 2015

CSE490 – Senior Year Project (1 student)
CSE203 – Object Oriented Programming with Java, 2 sections (73 students in total (41+32))
CSE101T – Introduction to Programming (73 students)

Contact

Institute for Ophthalmic Research
University of Tübingen

E-mail: murat-seckin[dot]ayhan[at]uni-tuebingen[dot]de
Physical Address: Elfriede-Aulhorn-Strasse 7, 72076 Tübingen, Germany