Enhancing Multi-class Diabetic Retinopathy Detection Using Tuned Hyper-parameters and Modified Deep Transfer Learning
Investigates image enhancement, transfer learning and hyperparameter tuning for retinal fundus classification.
Paper / DOIAcademic curriculum vitae
Postgraduate Researcher (PhD) in Bioengineering
Durham University
Durham, United Kingdom · farhaad.abedinzade@gmail.com
farhadabedinzadeh.github.io · ORCID: 0000-0002-0021-2009
I am a postgraduate researcher (PhD) in Bioengineering at Durham University working on interpretable AI for early prediction of childhood myopia using longitudinal multimodal retinal imaging and biometric data. My research interests include OCT/OCTA analysis, retinal biomarkers, ocular biometry, longitudinal modelling and multimodal learning.
My previous research spans biomedical image and signal processing, computational neuroscience and machine learning, including EEG connectivity, EMG, gait, voice and MRI analysis in neurodevelopmental and neurodegenerative disorders.
Research: Interpretable AI for Early Prediction of Childhood Myopia Using Longitudinal Multimodal Retinal Imaging and Biometric Data.
Islamic Azad University, Mashhad Branch, Iran · GPA: 17.40/20
Thesis: Detection of ADHD Using Phase-Based Brain Connectivity and Graph Theory.
Sadjad University of Technology, Mashhad, Iran
Thesis: Brain Tumor Detection Using Image Processing and Artificial Neural Networks.
Preparation of practical notebooks and learning materials for students with limited Python experience.
Workshop held by the National Brain Mapping Laboratory in Mashhad, Iran.
Islamic Azad University, Mashhad Branch. Designed lectures, supervised practical sessions and guided projects in signal analysis, wavelets, pattern recognition and machine learning.
Research with Seyyed Abed Hosseini, Islamic Azad University, Mashhad Branch. Work on brain connectivity and multimodal data in neurological disorders.
Python and MATLAB applications, data analysis, research methodology and support for undergraduate and master’s research projects.
Razavi HighTech Industries, Mashhad, Iran.
Completed at Durham University.
Python, MATLAB, PyTorch, TensorFlow/Keras, scikit-learn and Git.
Image and signal preprocessing; EEG, EMG, ECG, gait and MRI analysis; connectivity, graph theory and time-frequency methods.
OCT/OCTA exploration, EyePy workflows, retinal segmentation, thickness maps, en face projections and biometric integration.
MNE, EEGLAB, Brainstorm, FieldTrip, BrainNet Viewer, BRAPH, HERMES, SPM, LaTeX, Mendeley, EndNote and SPSS.
Investigates image enhancement, transfer learning and hyperparameter tuning for retinal fundus classification.
Paper / DOICombines linear and phase-based EEG connectivity with an attention-based convolutional network.
Paper / DOIUses interlimb wavelet coherence and transfer learning to study neurodegenerative gait patterns.
Paper / DOICombines EEG scalograms with convolutional and recurrent models for autism classification.
Paper / DOIComplete publication list: farhadabedinzadeh.github.io/publications/.
Journal peer reviewer:
Persian: native · English: advanced (C1, IELTS Academic).