E. Trivizakis

Eleftherios TrivizakisPhD

Ελευθέριος Τριβιζάκης

I build deep learning and machine learning models that read medical images, combine them with genomic and clinical data, and help predict how cancer will behave.

Researcher in AI for medical image analysis and multi-modal machine learning in oncology, with a focus on trustworthy, explainable and fair AI. Adjunct Professor at the Department of Electrical & Computer Engineering, Hellenic Mediterranean University, Heraklion, Crete.

Imaging, transcriptomic and cytogenetic signals fused into one prognostic signature

Research

My work sits between radiology, oncology and computer science: turning CT, MRI and pathology images into quantitative signatures, fusing them with other omics, and making the resulting models transparent enough to be trusted in the clinic.

900+citations
16h-index
15journal articles
22conference papers
1book chapter

Multi-omics and radiogenomics

Linking imaging features with transcriptomic, cytogenetic and clinical data to predict tumour subtypes, mutations, survival and therapy response, in lung cancer, multiple myeloma and beyond.

Key papers: [7] [14] [20] [23] [27] [28]

Deep learning for medical imaging

Detection, segmentation, classification and denoising across CT, MRI, mammography and histopathology, from 3D CNNs for liver tumours to bone marrow segmentation and prostate lesion detection.

Key papers: [2] [6] [12] [21] [25] [31]

Trustworthy AI

Explainability, fairness, federated learning and privacy-preserving synthetic data, so that clinical AI models are interpretable, equitable and deployable across institutions.

Key papers: [33] [34] [36] [42] [45]

Foundation models and generative AI

Benchmarking CT foundation models for lung nodule and survival prognosis, generative models for synthetic MRI, and GenAI pipelines that produce 3D animated content.

Key papers: [16] [35] [40] [41] [38]

Publications

Full list, newest first. Numbers match the references used throughout this site. For live citation counts see Google Scholar.

2026
  1. 45
    Survey of Methods for Trustworthy AI (tentative title)Trivizakis E. et al. Under review.Under review
  2. 44
    Testbed for foundation models in lung cancer prognosisKoutoulakis E., Trivizakis E. et al. 15th FORTH Retreat, Heraklion, Greece.Retreat
  3. 43
    BrainICP Imaging Archive: A dataset for AI models to evaluate intracranial pressure from CT brain scansTheodoropoulos D., Trivizakis E. Under review.Under review
  4. 42
    DP-BGMM-OCE: A differentially private synthetic radiomics generator in prostate cancer detection via Bayesian Gaussian Mixture ModelsPezoulas V., Trivizakis E. et al. Under review.Under review
  5. 41
    Deep Multimodal Lung Cancer Time-To-Event Risk Modelling Using CT Foundation ModelsKoutoulakis E., Trivizakis E. et al. Under review.Under review
  6. 40
    MedSAM-Guided Multimodal Attention CNN for Lung Nodule Malignancy Characterization on Low-Dose CTKoutoulakis E., Trivizakis E. et al. EuSoMII 2026, 9–10 October 2026, Heraklion, Greece.Conference
  7. 39
    Minimizing False Positive Annotations with A Fully Automated Artificial Intelligence Detection based on Radiologist Reports and Sparsely Labeled Lung Cancer CTsTrivizakis E. et al. EuSoMII 2026, 9–10 October 2026, Heraklion, Greece.Conference
  8. 38
    SmartAnima: A GenAI Pipeline for the Production of 3D Animated Movies (tentative title)Trivizakis E. et al. Under review.Under review
  9. 37
    The Use of Public Large Language Models for the Analysis of Screenplays and Theatrical Scripts (tentative title)Trivizakis E. et al. Under review.Under review
  10. 36
    Graph-Refined Probabilistic Mitigation and Fair Reweighing Improve Fairness without Sacrificing Performance in Prostate MRI RadiomicsTrivizakis E. et al. Accepted, MICCAI 2026, Strasbourg, France.Conference
  11. 35
    Benchmarking of Lung Cancer Foundational Models for Pulmonary Nodule Suspiciousness PrognosisKoutoulakis E., Trivizakis E. et al. Under review.Under review
  12. 34
    Federated Learning on Magnetic Resonance Imaging: A Critical ReviewMarkodimitrakis* E., Trivizakis* E. et al. Artificial Intelligence Review, Springer Nature. *Equal contribution.Journal
  13. 33
    A critical review of explainable deep learning in lung cancer diagnosisKoutoulakis* E., Trivizakis* E. et al. Artificial Intelligence Review, Springer Nature. *Equal contribution.Journal
2025
  1. 32
    AI-Powered ICP Evaluation from CT Scans: A Deep Learning ApproachTheodoropoulos D., Trivizakis E. et al. EANS 2025 Annual Congress, 5–9 October 2025, Vienna.Conference
  2. 31
    Predicting ICP Levels: A Deep Learning Approach Using CT Brain ScansTheodoropoulos D., Trivizakis E. et al. Neurosurgery.Journal
  3. 30
    Explainable AI Radiomics in Prostate Cancer Aggressiveness Prediction using different quantitative Diffusion MRI modelsIoannidis G. S., Trivizakis E. et al. 47th IEEE EMBC, Copenhagen, Denmark.Conference
  4. 29
    The Effect of Soft Annotations Compared to Pixel-based Masks of Prostate Gland on RadiomicsTrivizakis E. et al. 47th IEEE EMBC, Copenhagen, Denmark.Conference
  5. 28
    Artificial Intelligence-powered Multi-Omics in OncologyTrivizakis E. et al. ELIXIR All-Hands Meeting, Athens, Greece.Conference
2024
  1. 27
    Radiocytogenetics in Multiple Myeloma: Predicting Cytogenetic Aberrations from WBCT Imaging FeaturesTrivizakis E. et al. 8th ELECS, December 2024, Bern, Switzerland.Conference
  2. 26
    Label-Free Machine Learning-based Segmentation of Whole-Body Bone Marrow Imaging in Multiple MyelomaKoutoulakis E., Trivizakis E. et al. 8th ELECS, December 2024, Bern, Switzerland.Conference
  3. 25
  4. 24
2023
  1. 23
  2. 22
    A Machine Learning Framework for Hair Type Categorization to Optimize the Hair Removal Algorithm in Dermatoscopy ImagesIoannidis G. S., Trivizakis E. et al. IEEE EMBS DSE Healthcare, Malta.Conference
  3. 21
    Fully Automated Detection and Segmentation Pipeline for the Bone Marrow of the Lytic Bone of Multiple Myeloma PatientsKoutoulakis E., Trivizakis E. et al. IEEE EMBS DSE Healthcare, Malta.Conference
  4. 20
  5. 19
  6. 18
    LoockMe: An Ever Evolving Artificial Intelligence Platform for Location Scouting in GreeceTrivizakis E. et al. EANN 2023, Springer Nature Switzerland, pp. 315–327.Conference
  7. 17
    Deep Learning FundamentalsTrivizakis E. et al. In Introduction to Artificial Intelligence, Springer, pp. 101–131.Book chapter
2022
  1. 16
    Enhancing Cancer Differentiation with Synthetic MRI Examinations via Generative Models: A Systematic ReviewDimitriadis A., Trivizakis E. et al. Insights into Imaging, 13(1), 188.Journal
  2. 15
2021
  1. 14
  2. 13
  3. 12
  4. 11
    Cerebral CT perfusion in acute stroke: The effect of lowering the tube load and sampling rate on the reproducibility of parametric mapsIoannidis G. S., Christensen S., Nikiforaki K., Trivizakis E. et al. Diagnostics, 11(6), 1–14.Journal
  5. 10
    Pathomics and Deep Learning Classification of a Heterogeneous Fluorescence Histology Image DatasetIoannidis G. S., Trivizakis E. et al. Applied Sciences, 11(9), 3796.Journal
2020
  1. 9
  2. 8
  3. 7
2019
  1. 6
  2. 5
    Explainable AI: An Intuitive Analysis of Deep Learning in Medical Imaging and BiosignalsTrivizakis E. et al. 12th FORTH Scientific Retreat, Patras, Greece.Retreat
  3. 4
    The Vision of Integrating Artificial Intelligence in Health-CareTrivizakis E. et al. 12th FORTH Scientific Retreat, Patras, Greece.Retreat
  4. 3
    A novel multi-kernel 1D convolutional neural network for stress recognition from ECGGiannakakis G., Trivizakis E. et al. 8th ACIIW 2019.Conference
2018
  1. 2
2017
  1. 1
    EvoRDF: A framework for exploring ontology evolutionKondylakis H., Trivizakis E. et al. ESWC 2017 Satellite Events, Portorož, Slovenia, Springer, pp. 104–108.Conference

Research projects

National and EU-funded projects I have worked on, most recent first. Bracketed numbers link to the related publications.

  1. May 2026 – Dec 2026

    PATH-PNOngoing

    Trustworthy AI lead

    • Lead the project’s Trustworthy AI initiative.
    • Identify, introduce and implement Trustworthy AI solutions in existing workflows [45].
  2. Apr 2025 – Apr 2026

    SmartAnima

    Technical coordinator

    • Designed the AI workflow for generative AI in educational 3D content generation [37].
    • Developed the GenAI production workflow [38].
  3. Sep 2022 – Jun 2025

    GenoMed4All

    EU Horizon 2020, haematological diseases

    • Developed an ensemble machine learning model for patient survival prognosis with radiocytogenetics [20].
    • Designed deep learning pipelines for bone marrow segmentation [21] [26].
    • Developed models predicting chromosomal aberrations from imaging features [27] [28].
  4. Apr 2024 – Apr 2025

    Radioval

    • Developed a machine learning pipeline for predicting therapy response.
    • Developed a deep learning model for tumour detection.
  5. Jan 2022 – Sep 2024

    ProCancer-I

    EU Horizon 2020, prostate cancer AI platform

    • Developed a deep learning denoising model for T2-weighted MR images [25].
    • Developed deep learning models for prostate gland and lesion detection [29].
    • Developed machine learning models for lesion differentiation [30].
    • Developed AI fairness methods for prostate lesion differentiation [36].
  6. Jan 2021 – Dec 2024

    THORAX

    Lung cancer imaging

    • Designed the data collection process.
    • Developed machine learning models for lesion progression prognosis [23].
    • Developed deep learning models for lesion detection [24].
    • Integrated explainable AI [33].
  7. Jul 2021 – Sep 2023

    LoockMe

    AI platform for location scouting in Greece

    • Co-developed an AI-powered content management system.
    • Designed and coordinated the data collection process.
    • Developed interpretable deep learning models for landmark and landscape detection [18] [19].
  8. May 2019 – Jun 2021

    ARCHERS

    Funded by the Stavros Niarchos Foundation

    • Developed machine learning multi-omic models for assessing lung cancer outcomes [14].
    • Related output: [3]–[14].

Teaching & work history

Teaching at the Department of Electrical & Computer Engineering, Hellenic Mediterranean University, and earlier software development work.

  1. Sep 2026 – Feb 2027

    Adjunct ProfessorCurrent

    Dept. of Electrical & Computer Engineering, Hellenic Mediterranean University

    • Digital Image Processing, BSc program
  2. Feb 2026 – Jun 2026

    Adjunct Professor

    Dept. of Electrical & Computer Engineering, Hellenic Mediterranean University

    • Signals and Systems, BSc program
    • Advances in Digital Imaging and Computer Vision, AI topic for the MSc program
  3. Oct 2017 – Jan 2018

    Teaching Assistant

    Dept. of Electrical & Computer Engineering, Hellenic Mediterranean University

    • C Programming II
  4. Feb 2017 – Sep 2017

    Teaching Assistant

    Dept. of Electrical & Computer Engineering, Hellenic Mediterranean University

    • C Programming I
  5. Jan 2010 – Dec 2016

    Freelance developer

    Full-stack and back-end development

    • PHP, SQL, JavaScript, Java SE, Docker

Education

  • PhD, Prognostic Multi-omics with Artificial IntelligenceMedical School, University of Crete
  • MSc, Deep Learning in Medical ImagingDept. of Computer Engineering, Hellenic Mediterranean University
  • BSc, Software EngineeringDept. of Applied Informatics, Hellenic Mediterranean University (Former Technological Educational Institute of Crete)

Skills

  • Deep learning & ML model design
  • Medical imaging: CT, MRI, pathology
  • Multi-omics ensembles
  • Prognostic AI
  • Radiomics
  • Data science in oncology
  • Python
  • Keras
  • OpenCV
  • Scientific writing
  • National & EU proposal writing
  • Project management
  • Interdisciplinary collaboration
  • Teaching & mentoring

Contact

Open to research collaborations, joint proposals and student supervision in medical AI. Reach me through any of the profiles below.