Berens Lab

Data Science for Vision Research

Home » Labs » Berens Lab » Members » Murat Seckin Ayhan

Murat Seckin Ayhan

Surname Ayhan
First name Murat Seckin
Present position and title Post-doctoral Research Associate, PhD in Computer Science

Business address

Werner Reichardt Centre for Integrative Neuroscience (CIN)
Institute for Ophthalmic Research
University of Tübingen
Otfried-Müller-Str. 25
D-72076 Tübingen,
Germany

Phone: +49 7071 29-89106
E-mail: murat-seckin.ayhan[at]uni-tuebingen.de
Website: msayhan.com

Academic Education

Year Degree University Field of study
2015 PhD University of Louisiana at Lafayette, USA Computer Science
2010 M.Sc. University of Louisiana at Lafayette, USA Computer Science
2007 M.Sc. Baskent University, Ankara, Turkey Computer Engineering
2004 B.Sc. Baskent University, Ankara, Turkey Computer Engineering

Professional Experience

Period Institution Position Discipline
2015 - 2017 Isik University, Istanbul Assistant Professor Computer Engineering
2008 - 2015 Center for Advanced Computer Studies, University of Louisiana at Lafayette, USA Louisiana Optical Network Initiative (LONI) Fellow and PhD Student Computer Science
2006 - 2007 Atilim University Teaching and Research Assistant Computer Engineering
2004 - 2006 Baskent University Teaching and Research Assistant Computer Engineering

Research Interests

Deep learning, a subfield of machine learning, has enabled powerful tools for medical image analysis and diagnosis. These are mainly described as deep neural networks (DNNs) and typically used for the identification of anatomy or pathology from medical images. Such tasks are subject to uncertainty due to the factors including the inherent imaging noise and artifacts, patient variability, unfulfilled modeling assumptions as well as the inter-annotator variability. Also, in a medical setting, an AI system ideally gives the rationale for its decisions. In other words, it can simultaneously answer what is predicted and why it is predicted. However, DNNs have been long criticized for being black-boxes since their decisions are based on the non-linear interactions of many artificial neurons, which makes them hard for decision makers to understand and trust. Fortunately, recent methods promise to demystify their inner-workings and enable explanations for the decisions.

My research lies in the span of the uncertainty and interpretability dimensions of DNNs. I aim to construct accurate and reliable diagnostic tools via DNNs in order to promote the health and well-being of individuals and hence communities.

Selected Publications

Google Scholar: https://scholar.google.com/citations?user=ZX9sqrkAAAAJ