Javad Sheikh

Project Researcher at

University of Turku

About me

As a doctoral student with a deep interest in machine learning and deep learning, I specialize in developing innovative solutions for complex problems, focusing on computer vision and time series analysis. With hands-on experience in building and deploying ML and DL models, I aim to bridge the gap between cutting-edge research and real-world applications, contributing to fields that benefit from data-driven insights. I have also contributed to major EU-funded projects, including AI-ARC and EIS, collaborating with interdisciplinary teams to advance AI applications in the maritime and mining industries.

Download my Curriculum vitae.

Interests

Deep Learning
Machine Learning
Data Science
Computer Vision

Education

D.Sc. in Information and Communication Technology, 2023-present

University of Turku
M.Sc. in Artificial Intelligence & Robotics, 2018-2021

Amirkabir University of Technology
B.Sc. in Biomedical Engineering, 2013-2018

Amirkabir University of Technology

Skills

PyTorch

TensorFlow

Python

Experience

Project Researcher

University of Turku

Jul 2022 – Present Turku, Finland

Responsibilities :

Research
Collecting and Creating datasets
Train Deep learning model
Publishing papers

Freelance AI Engineer

Ponisha

Apr 2019 – Mar 2022 Tehran, Iran

Retinopathy detection app development:

Preprocessed retina images
Retrained DenseNet-169
Developed the AI powered mobile application

Artificial Intelligence App Developer

Sibbank

Sep 2018 – Sep 2019 Tehran, Iran

Responsibilities include:

Research
Train models
Develop application

Certifications

Data Scientist with Python Track

DataCamp Jan 2022

Gained the career-building Python skills need to succeed as a data scientist

See certificate

Deep Learning Summer School 2018

Tehran ACM Chapter Aug 2018

Deployed different Deep learning architectures:

Convolutional neural network
Recurrent neural network
Generative adversarial networks
…

Android Pack

Tehran Institute of Technology Jul 2017 – Oct 2017

Learned how to develop android application.

fMRI Workshop

General Office of Scientific Association and Olympiad Affairs May 2017

SCJP Java Programming Essentials

Tehran Institute of Technology Feb 2017 – May 2017

Learned Java Programming Essentials.

Featured Publications

Addressing imbalanced data for machine learning based mineral prospectivity mapping

This study addresses imbalanced data challenges in Mineral Prospectivity Mapping (MPM) using geophysical data from Lapland, Finland. It applies data-level imbalanced handling techniques and algorithm-level threshold adjustments to improve model predictions of rare mineral deposits. The performance of four ML models—MLP, RF, DT, and LR—was evaluated. The MLP model achieved the best accuracy (97.13%) on balanced data using synthetic oversampling.

Fahimeh Farahnakian, Javad Sheikh, Luca Zelioli, Dipak Nidhi, Iiro Seppä, Rami Ilo, Paavo Nevalainen, Jukka Heikkonen

Addressing imbalanced data for machine learning based mineral prospectivity mapping

A comparative study of state-of-the-art deep learning architectures for rice grain classification

This study explores deep-learning models for automated rice grain classification, which is essential for quality control, cost optimization, and supply chain efficiency. Testing models like ResNet, VGG, EfficientNet, and MobileNet on a dataset of 75,000 images across five rice categories, results showed that EfficientNet achieved the highest accuracy (99.67%), while MobileNet excelled in speed. The study concludes that deep-learning models offer scalable, efficient handling of large, complex data, outperforming traditional machine-learning methods.

Farshad Farahnakian, Javad Sheikh, Fahimeh Farahnakian, Jukka Heikkonen

A comparative study of state-of-the-art deep learning architectures for rice grain classification

Sea Ice Concentration Estimation Via Fusion of Sentinel-1 and AMSR2 based on Encoder-Decoder Architecture

This paper presents a new method using a Convolutional Neural Network (CNN) to estimate sea ice concentration in the Baltic Sea by combining Sentinel-1 and AMSR2 data. The CNN architecture retains different resolution inputs without losing valuable information. By using a focal loss function and skip connection, the proposed model achieves a high R2 score of 90.6%.

Javad Sheikh, Fahimeh Farahnakian, Farshad Farahnakian, Jukka Heikkonen

Sea Ice Concentration Estimation Via Fusion of Sentinel-1 and AMSR2 based on Encoder-Decoder Architecture

Short and Long Term Vessel Movement Prediction for Maritime Traffic

This study introduces short-term and long-term vessel trajectory prediction methods to enhance maritime traffic management. Tested with Baltic Sea AIS data, the short-term method shows high accuracy, while the long-term method optimizes speed and memory, promising improvements in efficiency and safety.

Farshad Farahnakian, Fahimeh Farahnakian, Javad Sheikh, Paavo Nevalainen, Jukka Heikkonen

Short and Long Term Vessel Movement Prediction for Maritime Traffic

Ice-Water Segmentation Using Deep Convolutional Neural Network-Based Fusion Approach

Presenting a novel CNN model merging SAR and AMSR2 imaging for precise maritime ice-water detection. This fusion enables automatic sea ice chart generation, outperforming uni-modal SAR architecture with 94.60% pixel-wise accuracy and an F1-score of 94.99%.

Javad Sheikh, Fahimeh Farahnakian, Farshad Farahnakian, Jukka Heikkonen

Ice-Water Segmentation Using Deep Convolutional Neural Network-Based Fusion Approach

A Comprehensive Study of Clustering-Based Techniques for Detecting Abnormal Vessel Behavior

The paper explores different clustering methods using AIS data to detect two dangerous abnormal behaviors in maritime vessels like dark ships (vessels turning off AIS for illegal activities) and spiral vessel movements. It evaluates K-means, DBSCAN, AP, and GMM techniques using three months of AIS data from the Baltic Sea. Results show that K-means is effective in identifying these threatening events, contributing to improving maritime safety.

Farshad Farahnakian, Florent Nicolas, Fahimeh Farahnakian, Paavo Nevalainen, Javad Sheikh, Jukka Heikkonen, Csaba Raduly-Baka

A Comprehensive Study of Clustering-Based Techniques for Detecting Abnormal Vessel Behavior

Design and implementation of brain surgery bipolar electrocoagulation simulator using haptic technology

In recent years, one of the most widely used tools in the training and evaluation of physicians and surgeons is surgical simulators. Introducing virtual reality and adding haptic technology to the field has made it feasible to develop ever-better surgical simulators. In this regard, we set out to design a brain surgery bipolar electrocoagulation simulator using haptic technology.

Reza Karimzadeh, Javad Sheikh, Hamed Azarnoush, Hossein Arabi

Design and implementation of brain surgery bipolar electrocoagulation simulator using haptic technology

Recent Publications

Quickly discover relevant content by filtering publications.

Javad Sheikh (2021). Anomaly detection on time series using deep neural networks.

Fahimeh Nazarimehr, Javad Sheikh, Mohammad Mahdi Ahmadi, Viet–Thanh Phamb, Sajad Jafari (2017). Fuzzy predictive controller for chaotic flows based on continuous signals. Chaos, Solitons & Fractals.

Cite DOI

Contact

Javad.Sheikh@utu.fi
+358 465 698 472
Turun yliopisto, Vesilinnantie 3, 20014, Vesilinnantie 3, Turku, 20500
Skype Me

Javad Sheikh

Project Researcher at

University of Turku

About me

Skills

Experience

Certifications

Featured Publications

Recent Publications

Popular Topics

Contact