Hesam Akbari

Research Interests

Explainable machine learning for analyzing brain dynamics and biodata, with applications in mental disorder detection and treatment prediction. My work combines nonlinear analysis, deep learning, and interpretable models to build reliable clinical decision-support systems.

Publications

Journal Articles:
- Depression Detection Based on Geometrical Features Extracted from SODP Shape and Binary PSO Traitement du Signal, 2021
- Depression Recognition Based on Phase Space Reconstruction and Geometrical Features Applied Acoustics, 2021
- Schizophrenia Recognition Based on Phase Space Dynamics and Graphical Features Biomedical Signal Processing and Control, 2021
- Classification of Normal and Depressed Signals Using Centered Correntropy Health Information Science and Systems, 2021
- Alcoholic Signal Recognition Based on Phase Space Dynamics and Geometrical Features, Chaos, Solitons & Fractals, 2022
- Seizure Detection Using Poincaré Plot and Graphical Features in DWT Domain Bratislava Medical Journal, 2023
- Identification of Neural Diseases Using a Computer-Aided Diagnosis Framework Computers in Biology and Medicine, 2021
- Detection of Focal and Non-Focal Signals Using Nonlinear Features in EWT Domain Physical and Engineering Sciences in Medicine, 2021
- Identification of Depression Signals in VMD Domain Health Information Science and Systems, 2022
- Dynamical Pattern of Successive Bits for Therapy Outcome Prediction Computers in Biology and Medicine, 2025
- Predicting Therapy Outcomes Using Amplitude Polar Maps Brain Sciences, 2025
- Deep Learning Framework for Therapy Prediction Using Time-Frequency Features Brain Sciences, 2026
- Time–Frequency Transformations and CNN Models for Parkinson’s Detection, BioMedInformatics, 2026

Conference Papers:
- Early Detection of Depression and Alcoholism Disorders Neural Information Processing, 2023
- Automatic Scheme for Depression Detection with Diagnostic Index Health Information Science Conference, 2021
- Fractional Fourier Transform Framework for Alcoholism Identification Health Information Science Conference, 2022
- Prediction of Therapy Outcomes Using Tuned Q-Factor Transform World Congress in Computer Science, 2024
- Decoding Brain Signals for Therapy Success Prediction IEEE BIBE, 2025
- Optimal Time-Domain Features for Therapy Prediction IEEE BIBE, 2025

Book Chapters:
- Automated Detection of Posterior Myocardial Infarction Using Dynamical Patterns
- Deep Learning for Cardiac Signal Analysis, 2026

Software & Systems:
- TOP-EEG: A Robust Software to Predict the Outcomes of Therapies for Depression Using EEG Signals in DGMD Domain
- UNT-AT: A Robust Software to Predict the Outcome of Depression Therapies Using EEG Signals

Other Publications:
- Fast Classification Using SODP and EWT
- Geometrical Method for Signal Discrimination
- Retinal Vessel Segmentation Using Gabor Filters
- Comparative Feature Analysis in EEMD Domain
- Hybrid Optimization Algorithms (Sine Cosine + GND)
- Empirical Wavelet Transform for Signal Analysis
- Detection of Seizure Activity Using Optimization Methods
- Parkinson’s Disease Detection Using Machine Learning
- Additional workshop papers and preprints