Print PDF

Seyedalireza Khoshsirat

27 years old, single

Tehran, Iran

Research Interests

I'm interested in combination of machine learning and image processing methods, and their application in medical image analysis.

I have experience of working with Convolutional Neural Networks, Active Shape Models, and Level-set methods.



University of Delaware

Computer Science

Jul 2018 – Present

Allameh Tabataba'i University (Tehran)

Computer Science / Intelligent Systems, GPA: 3.7

Sep 2015 – Sep 2017

University of Applied Science and Technology (Tehran)

Computer / Software Engineering, GPA: 3.4

Jan 2012 – Feb 2014

Shahid Shamsipour Technical College (Tehran)

Computer / Software Engineering

Jan 2009 – Sep 2011

Master's Thesis

Combined Deep-Learning and Level-set Approach to Segmentation of the Left Ventricle in 3D Cardiac MRI

Supervisor: Farzad Eskandari

Advisor: Mohammadreza Asghari Oskoei

Examiner: Seyed Ali Katanforoush (Shahid Beheshti University, Tehran)


Designing Evidence Based Risk Assessment System For Cancer Screening As An Applicable Approach For The Estimating Of Treatment Roadmap

Elham Maserat, Reza Safdari, Hamid Asadzadeh Aghdaei, Alireza Khoshsirat, Mohammad Reza Zali

BMJ Open, The 5th International Society for Evidence-Based Healthcare Congress, Kish Island, Iran

February 2017 - Volume 7 - Issue 1

doi: 10.1136/bmjopen-2016-015415.43

Standardized Tests


Total: 101, Reading: 29, Listening: 26, Speaking: 22, Writing: 24


Quantitative Reasoning: 158 (69th percentile), Verbal Reasoning: 146 (31th percentile), Analytical Writing: 3 (18th percentile)

Professional Service

Software Programmer

Mehrsys (Tehran)

Working on different software projects using up-to-date frameworks and technologies including Java (Spring, Hibernate, JasperReports), NodeJS, AngularJS, Ionic, MongoDB, SQL Server, TypeScript, Wordpress.

Sep 2014 – Aug 2018
Software Programmer

Raydana (Tehran)

Working on an ERP system (Enterprise Resource Planning) using Java (Struts, Spring, Hibernate, JSP, JSF, GWT), Oracle, MySQL, and etc.

Jun 2011 – Sep 2014
Software Programmer

Faranam (Tehran)

Frontend and backend software development with ASP.NET MVC, WPF, Entity Framework, SQL Server and jQuery.

Dec 2010 – Jun 2011

Technical Skills

Machine Learning

Convolutional Neural Networks, AutoEncoders, Principal Component Analysis

Image Processing

TensorFlow, Caffe, OpenCV, Active Shape Models, Active Appearance Models, Level-set methods


Oracle, MongoDB, SQL Server, Sqlite, LevelDB, HDF5

Programming Languages

Java, Python, Matlab, NodeJS, C#, TypeScript, HTML, JavaScript, CSS

Programming Frameworks

JavaEE, Spring, Hibernate, Struts, express.js, mongoose.js, Microsoft .NET, Entity Framework

User Interface

AngularJS, Telerik, KendoUI, Bootstrap, Ionic, Apache tiles, JavaFX, ASP.NET, ASP.NET MVC, WPF


Wordpress, Apache Maven, Apache Tomcat, JasperReports, Version Control Systems, gulp.js, CentOS, Photoshop

Teaching assistantship

Image Processing

Allameh Tabataba'i University

Professor Mohammadreza Asghari Oskoei

Spring 2017

Logic Programming

Allameh Tabataba'i University

Professor Hossein Teymoori

Fall 2016


Mohammadreza Asghari Oskoei

Allameh Tabataba'i University

Professor of Computer Vision

Hossein Teymoori Faal

Allameh Tabataba'i University

Professor of Mathematics and Logic

Mohammad Zebarjad

Mehrsys Co.

Chief Executive Officer


Master's Thesis Abstract

Segmentation of the left ventricle (LV) in cardiac magnetic resonance images (MRI) is an essential step for calculation of clinical indices such as ventricular volume and ejection fraction. In this thesis, we first explain essential concepts, then review existing methods for segmentation of LV. We continue by implementing and evaluating a method which employs deep learning algorithms combined with a level-set method to fully automatically segment the LV in short-axis cardiac MRI datasets.
The method employs deep learning algorithms to learn the segmentation task from the ground truth data. Convolutional networks are employed to automatically detect the LV chamber in MRI dataset. Stacked autoencoders are utilized to infer the shape of the LV. The inferred shape is incorporated into a level-set method to improve the accuracy and robustness of the segmentation. We validated our method using 45 cardiac MRI datasets taken from the MICCAI 2009 LV segmentation challenge and compared the results to the state-of-the-art methods. Excellent agreement with the ground truth was achieved. We computed validation metrics such as the percentage of good contours, Dice metric, average perpendicular distance, and conformity as respectively 83%, 80%, 3.4mm and 70%.

Keywords: Deep-Learning, Level-set method, Left Ventricle, Cardiac MRI, Machine Learning