Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I'm Theodoros Ioannidis, a MSc Artificial Intelligence and MEng Electrical and Computer Engineering graduate with a focus on applied computer vision, video-based inference, and deployable ML pipelines. I enjoy translating clinical and real-world problems into robust AI solutions, from non-intrusive sleep state classification in the NICU to anomaly detection in wireless capsule endoscopy videos. My work blends research with practical deployment, using PyTorch, OpenCV, YOLO, and Grad-CAM to make models trustworthy and actionable. I am passionate about embedded ML, model optimization, and field-deployable AI systems. I thrive on collaboration with medical professionals and engineers to translate clinical observations into structured models, and I enjoy building end-to-end pipelines and visualization tools that explain the decisions behind AI predictions.…I'm Theodoros Ioannidis, a MSc Artificial Intelligence and MEng Electrical and Computer Engineering graduate with a focus on applied computer vision, video-based inference, and deployable ML pipelines. I enjoy translating clinical and real-world problems into robust AI solutions, from non-intrusive sleep state classification in the NICU to anomaly detection in wireless capsule endoscopy videos. My work blends research with practical deployment, using PyTorch, OpenCV, YOLO, and Grad-CAM to make models trustworthy and actionable. I am passionate about embedded ML, model optimization, and field-deployable AI systems. I thrive on collaboration with medical professionals and engineers to translate clinical observations into structured models, and I enjoy building end-to-end pipelines and visualization tools that explain the decisions behind AI predictions.

Theodoros Ioannidis

AI Engineer, Data Scientist, App Developer, +2





I'm Theodoros Ioannidis, a MSc Artificial Intelligence and MEng Electrical and Computer Engineering graduate with a focus on applied computer vision, video-based inference, and deployable ML pipelines. I enjoy translating clinical and real-world problems into robust AI solutions, from non-intrusive sleep state classification in the NICU to anomaly detection in wireless capsule endoscopy videos. My work blends research with practical deployment, using PyTorch, OpenCV, YOLO, and Grad-CAM to make models trustworthy and actionable. I am passionate about embedded ML, model optimization, and field-deployable AI systems. I thrive on collaboration with medical professionals and engineers to translate clinical observations into structured models, and I enjoy building end-to-end pipelines and visualization tools that explain the decisions behind AI predictions.…I'm Theodoros Ioannidis, a MSc Artificial Intelligence and MEng Electrical and Computer Engineering graduate with a focus on applied computer vision, video-based inference, and deployable ML pipelines. I enjoy translating clinical and real-world problems into robust AI solutions, from non-intrusive sleep state classification in the NICU to anomaly detection in wireless capsule endoscopy videos. My work blends research with practical deployment, using PyTorch, OpenCV, YOLO, and Grad-CAM to make models trustworthy and actionable. I am passionate about embedded ML, model optimization, and field-deployable AI systems. I thrive on collaboration with medical professionals and engineers to translate clinical observations into structured models, and I enjoy building end-to-end pipelines and visualization tools that explain the decisions behind AI predictions.

Available to hire

I’m Theodoros Ioannidis, a MSc Artificial Intelligence and MEng Electrical and Computer Engineering graduate with a focus on applied computer vision, video-based inference, and deployable ML pipelines. I enjoy translating clinical and real-world problems into robust AI solutions, from non-intrusive sleep state classification in the NICU to anomaly detection in wireless capsule endoscopy videos. My work blends research with practical deployment, using PyTorch, OpenCV, YOLO, and Grad-CAM to make models trustworthy and actionable.

I am passionate about embedded ML, model optimization, and field-deployable AI systems. I thrive on collaboration with medical professionals and engineers to translate clinical observations into structured models, and I enjoy building end-to-end pipelines and visualization tools that explain the decisions behind AI predictions.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Language

English

Fluent

Greek, Modern

Fluent

German

Advanced

Work Experience

Chest X-ray Pathology Classification Developer at Independent Project

April 1, 2026 - Present

Built a web application for chest X-ray pathology classification using a pre-trained DenseNet model from Torch XRay Vision. Developed a structured FastAPI backend to handle image uploads, clinical preprocessing pipelines, inference, and prediction outputs. Integrated Grad-CAM visualizations to provide explainable AI insights by highlighting clinical regions.

Server-side Real-time Video Analysis Engineer at Independent / University project

April 1, 2026 - Present

Built a server-side real-time video analysis engine using OpenCV and MediaPipe to decode video frames and track human skeletal keypoints across frames. Implemented temporal fall-detection logic with in-memory frame tracking cache and strict type safety with Pydantic. Set up structured project configuration management via pyproject.toml.

Research Intern at AUTh Biomedical Technology Lab

April 1, 2026 - May 1, 2026

Participated in AUTh projects focusing on signal processing and biomedical technology; contributed to project development and literature-informed experiments.

Web Application Developer - Chest X-ray Pathology Classification at UMC Utrecht

April 1, 2026 - Present

Built a web application for chest X-ray pathology classification using a pretrained DenseNet model from TorchXRayVision. Implemented image upload, clinical preprocessing, inference, and output handling; integrated Grad-CAM visualizations for explainable AI; aligned prediction outputs with medical device workflow standards.

Student Researcher - Server-side Real-time Video Analysis Engine at UMC Utrecht

April 1, 2026 - May 1, 2026

Built a server-side real-time video analysis engine using OpenCV and MediaPipe to decode video frames and track human skeletal keypoints across frame sequences. Designed temporal fall-detection logic using vertical velocity, body aspect ratio, spatial displacement, and an in-memory frame-tracking cache. Implemented strict type safety constraints using MyPy, Pydantic data models, and structured project configuration management via pyproject.toml.

Student Researcher - Signal Processing & Biomedical Technology Lab at UMC Utrecht

April 1, 2026 - Present

Research on biomedical signal processing and biomedical technology projects; contributed to real-time video analysis and AI-enabled biomedical workflows.

MSc Thesis Researcher at Utrecht University / NICU Sleep State Research Project

July 1, 2025 - April 1, 2026

Conducted MSc thesis research on computer vision for non-intrusive sleep state classification of preterm infants in the NICU. Trained and evaluated Vision Transformer models in PyTorch Lightning for behavioral cue prediction and sleep state classification. Coordinated data collection and validation with medical professionals to align clinical observations with model development.

MSc Thesis Researcher - Computer Vision for Sleep State Classification in NICU at Utrecht University

July 1, 2025 - April 1, 2026

Conducted MSc thesis research on computer vision for non-invasive sleep state classification of preterm infants in the NICU.

MEng Thesis Researcher at Aristotle University of Thessaloniki

May 1, 2023 - November 1, 2023

Conducted MEng thesis research on anomaly detection and classification in wireless capsule endoscopy videos. Developed an end-to-end diagnostic object detection pipeline combining SimCLR with YOLOv8 for anomaly localization. Managed data processing pipelines and evaluated architectures using PyTorch and PyTorch Lightning with GPU acceleration.

MEng Thesis Researcher - Anomaly Detection in Wireless Capsule Endoscopy at Aristotle University of Thessaloniki

May 1, 2023 - November 1, 2023

Conducted MEng thesis research on anomaly detection in wireless capsule endoscopy. Developed an end-to-end diagnostic pipeline combining SiamCLR with YOLO v8 for anomaly localization; managed data processing pipelines; validated pipeline with GPU acceleration.