Biomedical AI
Models for diagnosis, prognosis, biomarkers, imaging, omics, tabular clinical data, and decision support.
Consultancy & Advisory
Biomedical AI consultancy, ML pipeline auditing, and forensic methodology review
I help biomedical, pharma, medtech, and clinical research teams build, validate, audit, and repair AI pipelines for biomarkers, MLE-NITs, diagnostic tests, prognostic models, and regulatory-facing evidence - backed by more than 10 years of production AI experience in the biomedical domain.
Low-data, explainable, auditable AI
Democratizing biomedical AI
Compact, train-from-scratch, CPU-friendly models for real biomedical settings: small cohorts, sensitive data, limited labels, omics batch effects, and the need for clinical interpretation.
What I can help with
MLE-NITs
Machine-learning-enhanced non-invasive tests: compact, interpretable upgrades to clinical scores and biomarker panels.
XAI & feature selection
Biomarker prioritization, biological signal retrieval, and potential drug-target identification for pharma-relevant discovery.
ML pipeline forensics
Leakage, cohort effects, batch effects, confounding, overfitting, split design, and reproducibility checks.
Low-data model design
ZACH-ViT, hZACH-ViT, s-DNNs, and compact baselines that can be trained and audited without massive compute.
Regulatory-facing evidence
Technical support around validation, transparency, GMLP, FDA AI/ML SaMD concepts, and lifecycle risk thinking.
LLMs & multimodal AI
RAG, clinical decision-support prototypes, and multimodal pipelines combining text, imaging, omics, and structured data.
Biomedical ML audit & regulatory-facing evidence
Forensic review of biomedical ML pipelines
Independent technical review for biomarker discovery, diagnostic-test development, prognostic modeling, patient stratification, clinical decision support, and near-production biomedical AI.
Pipeline forensics
Preprocessing, feature engineering, splits, validation, thresholding, leakage, and reproducibility.
Evidence quality
Biomarker stability, confounding, calibration, uncertainty, subgroup behavior, and clinical plausibility.
Production validation
Distribution shift, monitoring assumptions, audit trails, documentation, and failure modes.
Regulatory-facing support
Scientific and technical advisory aligned with GMLP, FDA AI/ML SaMD concepts, transparency, and lifecycle risk management.
Related high-stakes AI expertise
Biometrics as a trustworthiness pillar
My ECG and face biometrics work reinforces the biomedical AI brand through robustness, privacy, adversarial testing, leakage control, and deployment-aware validation.
Domain experience
- Biomedical AI: liver disease, MASLD/MASH, fibrosis, steatosis, AML, periodontitis, drug interactions, adverse drug reactions.
- Medical imaging: ultrasound, B-mode imaging, lung ultrasound, liver ultrasound, breast-lesion risk prediction, segmentation/classification.
- Biometrics and security: ECG authentication, face recognition, privacy-preserving biometrics, adversarial robustness.
- Industrial data science: route planning, ETA prediction, TSP, geocoding, speed-profile imputation, computer vision, production ML.
Typical deliverables
- Technical audit reports and methodology review documents.
- Reproducible Python notebooks and ML pipelines.
- Model validation reports with metrics, confidence intervals, and bias/confounding analysis.
- Feature-selection and explainability reports.
- Prototype models, proof-of-concept systems, and production-readiness recommendations.
- Scientific manuscript support for ML methodology and reporting.
Why work with me?
More than 10 years of production AI experience in the biomedical domain, combined with academic AI research at Amsterdam UMC and biometric-security research at BioML/CODE.
Experience across Hologic, Diagnostic Echotomography, MGH/Harvard, Huawei, Plotwise, and independent biomedical AI consultancy.
Useful when a project needs rigorous validation, reproducible code, clear interpretation, and defensible technical documentation.
Consultancy focus areas
Biomedical AI
Clinical ML
Explainable AI
Feature Selection
Model Validation
AI Audits
ML Pipeline Forensics
MLE-NITs
Diagnostic Test Validation
Prognostic Modeling
Biomarker Discovery
Regulatory-Facing Evidence
FDA AI/ML SaMD
Good Machine Learning Practice
AMA Augmented Intelligence
Bias Assessment
Data-Centric AI
Synthetic Data
Medical Imaging
Omics
LLMs
RAG
Biometrics
Low-Data AI
Democratizing AI
Train-from-Scratch Models
Computer Vision
Routing & Optimization
Education & Seminars
Responsible AI seminars for physicians and biomedical teams
In addition to consultancy and advisory work, I provide AI education for clinicians, researchers, and pharma-supported medical audiences: responsible AI use, explainability, LLMs, multimodal AI, and preparation for the agentic AI era in medicine.
Use this form for consultancy, ML pipeline auditing, biomarker/diagnostic/prognostic model review, regulatory-facing evidence support, or AI seminars.
Request a consultancy or AI seminar discussion
Use this form for consultancy, forensic ML pipeline auditing, diagnostic/prognostic model review, biomarker-discovery methodology review, regulatory-facing evidence support, responsible AI education, pharma-supported physician seminars, invited talks, or research collaboration inquiries.
Available in English and Greek - online, onsite, or hybrid.
Replace the placeholder Formspree endpoint in the HTML with your own endpoint when you create it.
Contact
Discuss a project
For consultancy, advisory, research collaboration, model validation, or technical review, contact me by email.