Download PDF

Contact Information

Address:

No. 18, Niloofar 6th St., Sarv St., Ta'avon Blvd., Shahreziba, Tehran, Iran

Email:

[email protected]

Phone:

+989384996069

Research Interests

  • Deep Learning and Convolutional Neural Network

  • Computer Vision and Medical Image Processing

  • Pattern Recognition

  • Big Data techniques on Machine Learning Problems

  • Machine LearningApplications 

Current Study

Automatic Scoring of Lung HRCT Patterns in Cystic Fibrosis using Convolutional Neural Network

In Machine Vision and Medical Image Processing Lab, K.N.Toosi University of Technology. 

Prof. H. Abbrishami Moghaddam

Education

Sep 2015Present

M.S., Artificial Intelligence

K.N.Toosi University of Technology, Tehran Iran

Grade point average: 18.00/20 (4.0/4.0) (Ranked 1st)

  • 1st Semester: 17.62/20 (4.0/4.0) (Ranked 3rd)
  • 2nd Semester: 17.83/20 (4.0/4.0) (Ranked 2nd)
  • 3rd Semester: 17.88/20 (4.0/4.0) (Ranked 1st)
  • 4th Semester: 19.0/20 (4.0/4.0) (Master Thesis)

Relevant Courseworks:

    •      Machine Learning (18.5/20) (GPA 4) (Ranked 3rd)
    •      Artificial Neural Network (17.45/20) (GPA 4) (Ranked 5th)
    •      Multi-agent Systems (18.4/20) (GPA 4) (Ranked 2nd)
    •      Big Data and Hadoop (17/20) (GPA 4) (Ranked 5th)
    •      Digital Image Processing (16.2/20) (GPA 4.0) (Ranked 3rd)
    •      Evolutionary Computing (17.65) (GPA 4) (Ranked 3rd)
    •      Reinforcement Learning (18.5/20) (GPA 4) (Ranked 1st)
    •      Cloud Computing (17/20) (GPA 4.0) (Ranked 6th)
Sep 2011Aug 2015

B.S., Computer Enginering

K.N.Toosi University of Technology, Tehran, Iran

Grade point average: 16.92/20 (4.0/4.0) (Ranked 5th)

Relevant Courseworks:

    •      Discrete Mathematics (17.25/20)
    •      Probability & Statistics (18.7/20)
    •      Numerical Optimization (17.75/20)
    •      Electrical Circuits Theory (16.5/20)
    •      Principle of Operating Systems (19/20)
    •      Computer System Architecture (19.5/20) 
    •      Principle of Artificial Intelligence (17/20)
    •      Algorithms and Data Structure(16.5/20)
    •      Compilers (19/20)
    •      Software Engineering (18/20)

 

Sep 2007Aug 2011

High School Diploma in Mathematics and Physics

Mofid High School

Grade point average: 19.37/20 (4.0/4.0)

Papers

Sep. 2017Present

Submitted Paper, IEEE Journal of Biomedical and Health Informatics

Kasra Nezamabadi, Zeinab Naseri Samaghcheh, Hamid Abrishami Moghaddam, Mohammadreza Modarresi, Neda Pak, Mehrzad Mahdizade. "Lung HRCT Pattern Classification for Cystic Fibrosis using Convolutional Neural Network". IEEE J-BHI. JBHI-00679-2017

Revision of Paper, IEEE Journal of Biomedical and Health Informatics

"A Computer-Aided System Based on Compressive Sensing for Interstitial Lung Pattern Analysis in Three Dimensional High Resolution Computerized Tomography", IEEE J-BHI

Research Experience

Sep. 2017Present

Research Assistant

Computer Vision and Medical Image Processing Lab, K.N.Toosi University. Cystic Fibrosis Center, Tehran University of Medical Science.

Automatic scoring of lung HRCT patterns in Cystic Fibrosis using Convolutional Neural Network. Classifying Bronchiectasis patterns into 4 different classes based to which the severity of Cystic Fibrosis disease is defined.

Under the supervision of Prof. Hamid Abrishami Moghaddam.

Mar. 2017Sep. 2017

Research Assistant

Computer Vision and Medical Image Processing Lab, K.N.Toosi University. Cystic Fibrosis Center, Tehran University of Medical Science.

Lung HRCT pattern classification for Cystic Fibrosis using Convolutional neural network. Building patch-based dataset of CF inspiratory lung HRCT scans and classifying patterns into 3 different classes were the most important phases in this project.

Under the supervision of Prof. Hamid Abrishami Moghaddam and advisory of Prof. Mohammadreza Modarresi (Tehran University of Medical Science)

Jul 2016Present

Paper Revision Coordinator

Medical Imaging Center, Imam Khomeini Hospital Complex, Tehran University of Medical Science

Gathering data of lung HRCT images for interstitial lung disease.
Managment of labeling images with the cooperation of several resident radiologists.
Revision of paper "A Computer-Aided System Based on Compressive Sensing for Interstitial Lung Pattern Analysis in Three Dimensional High Resolution Computerized Tomography", Journal of Biomedical and Health Informatics.

Under the supervision of Prof. Hamid Abrishami Moghaddam and advisory of Prof. Kahkoohi.

Teaching Experience

Fall 2017Present

Teaching Assistant of Big Data and Hadoop Framework (M.S. Course)

Prof. S. Hossein Khasteh, K.N.Toosi University
Summer 2017Present

IEEE Lead Instructor in Deep Learning Workshop

K.N.Toosi University of Technology, IEEE Iran Section

Deep Learning workshop was held with the cooperation of IEEE Iran section in K.N.Toosi University of technology. I was the leading instructor in this 10-session workshop.

The key outlines of this workshop were:

  • Principles of Deep Learning
  • Theoretical Principles of ConvNets
  • ConvNet advanced Ideas
  • Theano and TensorFlow Object Oriented Programming in Python
  • ConvNet Implementation with Theano and TensorFlow
Spring 2017Present

Teaching Assistant of Machine Vision

Prof. Behrooz Nasihatkon, K.N.Toosi University
Spring 2014Present

Teaching Assistant of Algorithms and Data Structures

Prof. Amin Nikanjam, K.N.Toosi University
Spring 2013Present

Teaching Assistant of Java Advanced Programming

Prof. Saeed Seddighian, K.N.Toosi University
Spring 2013Present

Teaching Assistant of Computer System Architectures

Prof. Behrooz Nasihatkon, K.N.Toosi University

Certificates

Sep. 2017Present

Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization

Coursera Course Certificate, deeplearning.ai

Instructor: Prof. Andrew Ng

License DWNGGZ3ZP9YL

Grade: 98/100

Sep. 2017Present

Neural Networks and Deep Learning

Coursera Course Certificate, deeplearning.ai

Instructor: Prof. Andrew Ng

License GRV64F9MT24T

Grade: 100/100

Jan. 2016Present

Machine Learning

Coursera Course Certificate. Stanford University

Instructor: Prof. Andrew Ng

License RJ7232TEBJ3M

Grade: 96/100

Mar. 2017

Financial Technology Festival (FintechFest) Certificate of Attendance

Fanap Co.

Honors

Full Scholarship for M.S., Artificial Intelligence

K.N.Toosi University

I was one of the three selected students for entering master degree in artificial intelligence.

Aug. 2017Present

IEEE Certificate of Appreciation

K.N.Toosi University, IEEE Iran Section

I was the lead instructor in the Deep Learning workshop held in K.N.Toosi University with the cooperation of IEEE Iran section on May 2017.

License IEEE-I.S.S.B.-K.N.T.U.T./2017-137

Skills

Programming Languages

  • Python Advanced Programming
  • Swift Advanced Programming (iOS App Development)
  • Java
  • OpenCV
  • .Net
  • C / C++

Profesional Implementations and Frameworks

  • Advanced TensorFlow Programming
  • Advanced Theano Programming
  • Convolutional Neural Network Implementation
  • Machine Learning Algorithms Implementation
  • Big Data and Hadoop
  • Cloud Computing in Machine Learning Applications
  • Evolutionary Algorithms Implementation
  • iOS App Developing
  • Database Management
  • Matlab

Others

  • Team Working
  • Self-Reliant with Strong Management
  • Personal Communication
  • Presentation Skills
  • Teaching Skills

English Proficiency

  • English (Fluent)
  • Persian (Native)
  • Arabic (Familiar)

Latest Projects

  • Convolutional Neural Network to classify Cystic Fibrosis lung HRCT patterns using, Python,TensorFlow  

   Achieving accuracy of ~95.3%  and sensitivity and specificity of ~95.3 and ~97.6 ,respectively.

  • Convolutional Neural Network to Automatic Scoring Cystic Fibrosis lung HRCT patterns using, Python,TensorFlow  

     Achieving accuracy of ~74.3%. (In Progress)

  • Convolutional Neural Network to Recognize Persian Handwritten Digits using Python, Theano

        Achieving 97% accuracy on a 20,000 sample dataset

Exporting network parameters into an iOS app and building a keyboard to recognize handwritten Persian digits.

 

  • Convolutional Neural Network to recognize ILD Disease patterns in lung HRCT images in 3D Voxel Patches In Progress

  • PayMa, Social Mobile Banking App at Dotin, Fanap Co.

         Link: http://payma.ir

 

  • Implementation of Neural Network to Recognize MNIST Handwritten Digits

 

  • Design and Implementation of Plagiarism Detector Software

Undergraduate Final Project

Design and implementation of plagiarism detector software to detect and report plagiarism probability of a given document using Google Search Engine in C#.

  • Hoortash | Social Network App in iOS
    • There are some features of this app:
    1. Instantly updates Feeds whenever a user posts, likes or comments on a post.
    2. Cached image system to improve app performance.
    3.  2-way authentication, one-step Login / Sign Up and Facebook Authentication
    4.  Ability of leaving comment and like a post.
    5.  Ability of editing user profile
    6. ...

  • Recommender System based on Movie Ratings

From MovieLens 100k Dataset from GroupLens Research. This dataset consists of ratings on a scale of 1 to 5. The dataset has 943 users, and 1682 movies.

  • Anomaly Detection Application to Detect Anomalous Behavior in Server Computers

The features measure the throughput (mb/s) and latency (ms) of response of each server. Dataset consists of 307 examples.

  • Spam Filter using Support Vector Machine (SVM)

Implementation of a spam filter with Support Vector Machine (SVM) to classify spam (y=1) and non-spam(y=0) emails. The dataset is based on a a subset of the SpamAssassin Public Corpus, available at: http://spamassassin.apache.org/publiccorpus/

  • Image compression using K-Means Algorithm

References