1 / 17
Domain Adaptation for Medical Image Segmentation Using Transformation-Invariant Self-Training
2 / 17
Domain Adaptation for Medical Image Segmentation Using Transformation-Invariant Self-Training
3 / 17
LensID: A CNN-RNN-Based Framework Towards Lens Irregularity Detection in Cataract Surgery Videos
4 / 14
LensID: A CNN-RNN-Based Framework Towards Lens Irregularity Detection in Cataract Surgery Videos
5 / 14
DeepPyramid: Enabling Pyramid View and Deformable Pyramid Reception for Semantic Segmentation in cataract Surgery Videos
6 / 17
DeepPyramid: Enabling Pyramid View and Deformable Pyramid Reception for Semantic Segmentation in cataract Surgery Videos
7 / 17
ReCal-Net: Joint Region-Channel-Wise Calibrated Network for Semantic Segmentation in Cataract Surgery Videos
8 / 17
ReCal-Net: Joint Region-Channel-Wise Calibrated Network for Semantic Segmentation in Cataract Surgery Videos
9 / 17
ReCal-Net: Joint Region-Channel-Wise Calibrated Network for Semantic Segmentation in Cataract Surgery Videos
10 / 14
ReCal-Net: Joint Region-Channel-Wise Calibrated Network for Semantic Segmentation in Cataract Surgery Videos
11 / 17
ReCal-Net: Joint Region-Channel-Wise Calibrated Network for Semantic Segmentation in Cataract Surgery Videos
12 / 17
Relevance-Based Compression of Cataract Surgery Videos Using Convolutional Neural Networks
13 / 17
Relevance-Based Compression of Cataract Surgery Videos Using Convolutional Neural Networks
14 / 17
Relevance-Based Compression of Cataract Surgery Videos Using Convolutional Neural Networks
15 / 17
Relevance Detection in Cataract Surgery Videos using Spatio-Temporal Action Localization
16 / 17
Deblurring Cataract Surgery Videos Using a Multi-Scale Deconvolutional Neural Network
17 / 17
Deblurring Cataract Surgery Videos Using a Multi-Scale Deconvolutional Neural Network

About

I am a Research Scientist in the AI for Medical Devices team at Haag-Streit, where I develop cutting-edge deep learning solutions for surgical skill assessment, virtual reality guidance, and automated video analysis in ophthalmology. My work bridges AI research and real-world clinical applications, focusing on enhancing surgical precision, improving diagnostic accuracy, and optimizing workflow efficiency in medical imaging and surgical video analysis.

Prior to joining Haag-Streit, I conducted postdoctoral research at the University of Bern’s AI for Medical Imaging lab, where I led projects in domain adaptation, semi-supervised learning, and phase recognition in surgical videos. My PhD research at Klagenfurt University focused on advanced deep learning models for surgical video analysis, including phase and action recognition, real-time anomaly detection, and relevance-based video compression. My work has contributed to AI-driven surgical content retrieval systems currently in use at major hospitals, supporting both clinical decision-making and surgical training.

News

Interview: Deep Learning can contribute to improving surgical outcomes.
Distinguished Reviewer: Certificate of Distinction as an IEEE TMI Distinguished Reviewer.

Fields of Expertise & Research

Domain

  • Computer Vision & Deep Learning
  • Medical Image & Video Analysis
  • AI for Medical Devices & Healthcare
  • Image & Video Compression
  • Information Theory

Core AI Techniques

  • Supervised, Self-Supervised, and Semi-Supervised Learning
  • Domain Adaptation & Generalization
  • Generative Modeling & Diffusion Models
  • Large Language Models (LLMs) & Multimodal Learning
  • Retrieval-Augmented Generation (RAG) & LangChain

Applications in Healthcare AI

  • Semantic Segmentation & Object Detection – Surgical scene understanding, pathology detection
  • Action Recognition & Irregularity Detection – Surgical skill assessment, anomaly detection
  • Virtual Reality & AI-assisted Surgery – Real-time guidance, augmented intelligence for medical professionals
  • Edge & Cloud AI for Medical Devices – Optimizing AI models for real-time hospital deployment

Education

Ph.D. in Computer Science

03.2019 - 09.2021

Department of Information Technology, Alpen-Adria-Universität Klagenfurt, Austria

Dissertation Title: "Deep-Learning-Assisted Analysis of Cataract Surgery Videos"

Supervisors:
Assoc. Prof. Dr. Klaus Schoeffmann
Prof. Dr. Christian Timmerrer

Examiners:
Prof. Dr. Henning Müller, HES-SO Valais and University of Geneva, Switzerland
Prof. Dr. Raphael Sznitman, University of Bern, Switzerland

Grade: 1 - excellent

Visiting Scholar

08.2021 - 09.2021

HES-SO, University of Applied Science Western Switzerland

M.Sc. in Electrial Engineering

09.2013 - 10.2016

Department of Electrical Engineering, Ferdowsi University of Mashhad, Iran

Thesis Title: "Undetectable Video Steganography Based on Statistical Differences of Motion Vectors, Before and After Embedding"

Supervisor:
Prof. Morteza Khademi

Examiners:
Ass. Prof. Dr. Seyed Alireza Seyedin, Ferdowsi University of Mashhad, Iran
Prof. Dr. Raphael Sznitman, Ferdowsi University of Mashhad, Iran

Employment

Research Scientist

08.2024 - Present

Haag-Streit

  • Developing AI-powered solutions for surgical skill assessment, including both subjective and automated evaluation in wetlab cataract surgery.
  • Leading research on real-time intraoperative AI for surgical video analysis, optimizing procedural workflows and improving decision support.
  • Implementing virtual reality guidance systems for cataract surgery, enhancing precision and intraoperative visualization.
  • Developing AI-driven quality assessment models for slit-lamp videos, enabling automated relevance detection and active learning.
  • Bridging AI research with clinical applications to ensure model robustness, validation, and seamless real-world integration.

Postdoctoral Researcher

02.2022 - 07.2024

ARTORG Center for Biomedical Engineering Research, Department of Medicine, University of Bern

  • Conducted research on semi-supervised learning and domain adaptation for medical image analysis.
  • Focused on semantic segmentation, phase recognition in surgical videos, and AI-driven workflow analysis.
  • Supervised MSc and PhD students in deep learning applications for medical imaging.

Postdoctoral Researcher

10.2019 - 12.2021

Department of Information Technology, Alpen-Adria-Universität Klagenfurt, Austria

  • Developed real-time intraoperative irregularity detection systems for surgical safety enhancement.
  • Designed deep learning models for postoperative complication prediction, leveraging intraoperative data and patient-specific metrics.

Research Assistant

03.2019 - 09.2021

Department of Information Technology, Alpen-Adria-Universität Klagenfurt, Austria

  • Developed deep learning architectures for surgical phase recognition, object localization, and instance tracking.
  • Contributed to relevance-based compression techniques for efficient storage and retrieval of surgical videos, now deployed in major eye hospitals in Austria.
  • Focused on surgical scene segmentation, overcoming challenges like occlusion, motion blur, and transparency.

Teaching

Biomedical Engineering Laboratories

02.2024 - 05.2024

University of Bern

Master's Course, Biomedical Engineering

Biomedical Engineering Laboratories

02.2023 - 05.2023

University of Bern

Master's Course, Biomedical Engineering

Biomedical Engineering Laboratories

02.2022 - 05.2022

University of Bern

Master's Course, Biomedical Engineering

Web Technologies

10.2019 - 01.2020

University of Klagenfurt

Bachelor's Course, Computer Science

Image and Video Processing with Matlab

06.2016 - 09.2016

Ferdowsi University of Mashhsad

Workshop, Electrical Engineering

Image Processing with Matlab

06.2015 - 09.2015

Ferdowsi University of Mashhsad

Workshop, Electrical Engineering

Fields and Waves Electromagnetics

09.2014 - 01.2015

Eqbal Institute of Higher Education

Bachelor's Course, Electrical Engineering-Electronics and Electrical Engineering-Robotics

Technical Skills

Programming Languages

  • Python (Expert)
  • C++ (Intermediate)
  • C (Intermediate)
  • Matlab (Intermediate)
  • Web Development (HTML, CSS, JavaScript, NodeJS - Advanced)

AI & Deep Learning Frameworks

  • PyTorch (Expert)
  • TensorFlow (Intermediate)
  • Keras (Advanced)

Data Processing & GPU Acceleration

  • Image & Video Processing (OpenCV, FFmpeg - Intermediate)
  • Medical Image Analysis (MONAI - Intermediate)
  • GPU Acceleration (CUDA, cuDNN - Intermediate)

Model Optimization & Inference

  • ONNX (Model Interoperability - Intermediate)
  • TensorRT (High-Performance Inference - Intermediate)
  • Triton Inference Server (Scalable AI Deployment - Intermediate)

AI Deployment & Cloud Services

  • Cloud AI (AWS, Azure, Google Cloud - Intermediate)
  • Model Compression (Quantization, Pruning, Knowledge Distillation - Advanced)
  • Edge AI (Optimized AI for Low-Power Devices - Intermediate)
  • Federated Learning (FLARE - Intermediate)

Selected Publications

  • All
  • PhD Dissertation
  • Conference Papers
  • Journal Papers
Cataract-1K: Cataract Surgery Dataset for Scene Segmentation, Phase Recognition, and Irregularity Detection
[Nature Scientific Data 2024]
Predicting Postoperative Intraocular Lens Dislocation in Cataract Surgery via Deep Learning
[IEEE Access 2024]
DeepPyramid+: medical image segmentation using Pyramid View Fusion and Deformable Pyramid Reception
[IJCARS 2024]
Domain Adaptation for Medical Image Segmentation Using Transformation-Invariant Self-Training
[MICCAI 2023]
DeepPyramid: Enabling Pyramid View and Deformable Pyramid Reception for Semantic Segmentation in cataract Surgery Videos
[MICCAI 2022]
LensID: A CNN-RNN-Based Framework Towards Lens Irregularity Detection in Cataract Surgery Videos
[MICCAI 2021]
ReCal-Net: Joint Region-Channel-Wise Calibrated Network for Semantic Segmentation in Cataract Surgery Videos
[ICONIP 2021]
Relevance-Based Compression of Cataract Surgery Videos Using Convolutional Neural Networks
[ACM MM 2020]
Deblurring Cataract Surgery Videos Using a Multi-Scale Deconvolutional Neural Network
[ISBI 2020]
Relevance Detection in Cataract Surgery Videos using Spatio-Temporal Action Localization
[ICPR 2020]
Enabling Relevance-Based Exploration of Cataract Videos
[ICMR 2020]
Blind MV-based Video Steganalysis Based on joint Inter-frame and Intra-frame Statistics
[MTAP 2020]
Undetectable video steganography by considering spatio-temporal steganalytic features in the embedding cost function
[MTAP 2020]
Deep-Learning-Assisted Analysis of Cataract Surgery Videos
[MTAP 2020]

Academic Services

Journal Reviewer

  • IEEE Transactions on Medical Imaging (TMI)
  • Medical Image Analysis
  • IEEE Transactions on Multimedia
  • Multimedia Tools and Applications (MTAP)
  • Multimedia Systems
  • Journal of Clinical Medicine
  • Expert Systems with Applications
  • Biocybernetics and Biomedical Engineering

Program Committee Member

  • MICCAI 2024 – International Conference on Medical Image Computing and Computer-Assisted Interventions
  • MICCAI 2023
  • ACM ICMR 2021 – International Conference on Multimedia Retrieval
  • MMM 2021 – 27th International Conference on Multimedia Modeling
  • LSC'21 – Fourth Lifelog Search Challenge at ACM ICMR
  • CBMI 2021 – 18th Conference on Content-Based Multimedia Indexing
  • ACM MM Asia 2020
  • ACM MM 2020 – ACM Multimedia Conference
  • ICMR 2020 – International Conference on Multimedia Retrieval

Organization Committee

  • Best Poster/Demo Committee, ACM International Conference on Multimedia Retrieval (ICMR 2020)

Contact

Location:

ARTORG Center for Biomedical Engineering Research, Murtenstrasse 50, 3008 Bern, Switzerland

Call:

+41 31 632 75 91

Loading
Your message has been sent. Thank you!