Clinical AI research leader for regulated medical-device programmes
I lead clinical AI programmes from protocol design and statistical analysis plans through validation, stakeholder reporting, and FDA/CE MDR evidence across biosignals, imaging, and regulated deployment.
Based in France · collaborating across Europe and North America
Start here
Selected work
Four programmes showing end-to-end clinical delivery, regulatory strategy, and research leadership, from protocol design through validation and stakeholder reporting.
Head of Research · DESKi
Set research direction and validation strategy for cardiac ultrasound AI under FDA and EU device regulation.
Machine Learning Lead · Sensome
Owned three clinical trial programmes end to end, from protocol design and statistical analysis plans to stakeholder reporting.
Principal ML Engineer · Sama Therapeutics
Led biosignal strategy and feature extraction for a transprognostic treatment-response programme validated on unseen clinical cohorts.
PhD researcher · NeuroSpin / Sorbonne University
PhD research on how humans and language models process language, informing translational clinical AI strategy.
Selected outputs
Five outputs across clinical trial delivery, lung oncology, psychiatric validation, cognitive neuroscience, and computational linguistics.
Peer-reviewed journal
n=22; combined MEG and EEG during sentence reading. Only hierarchical structure was decodable from brain signals; transition and congruity stayed at chance. A two-layer LSTM on the same sentences showed all three effects decodable.
Conference paper
Rapid serial visual presentation agreement task (n=34): humans and language models err more with proximal attractors; linear response-time effect of distance; GPT-Neo-1.3B and grammar-corrected T5 compared to humans.
Clinical abstract
In vivo clinical validation (n = 17); 100% primary endpoint success for lesion impedance data. Clot identification in peripheral arterial disease (tissue rich in red blood cells).
Also presented at Paris Vascular Insights (PVI) 2024, Paris
Clinical abstract
Machine learning (ML) tissue classification during bronchoscopic lung biopsy; 26 patients across Australia and France.
Conference poster
Major depressive disorder, ADHD, OCD, and PTSD; external validation on unseen clinical cohorts (TRIPOD Type 4); ranked first in the TDBRAIN international competition.
About
Ph.D. in cognitive neuroscience · Sorbonne University / NeuroSpin · supervised by Prof. Stanislas Dehaene
PhD-trained research leader. I have directed teams and owned end-to-end delivery across three full-cycle clinical trial programmes (SEPARATE, E-SEPARATE, INSPECT), from clinical protocol design and statistical analysis plans through analysis, Scientific Advisory Board reporting, and regulatory-facing evidence. Current focus: cardiac ultrasound AI at DESKi under FDA PCCP and CE MDR.
Education
Sorbonne University / NeuroSpin (Unicog Lab, CEA Paris-Saclay)
Thesis: On the dissociation of structural and linear operations in sentence processing. Supervisor: Prof. Stanislas Dehaene.
Donders Institute, Radboud University
Thesis: Reconstructing the perceptual organization of sound from neural responses.
Aristotle University of Thessaloniki
Specialisation: Analysis and applications of EEG: Theory & Practice.
Collaboration
I work with clinical investigators, regulatory specialists, and ML engineers on in vivo catheter lab trials, ex vivo histology validation studies, psychiatric clinical trial cohorts, and cardiac imaging R&D. Recent delivery spanned Belgium, Australia, and France (peripheral arterial and first in human lung studies); psychiatric biomarker work drew on pooled international trial cohorts.
Based in Paris. Career across Greece, the Netherlands, and France; comfortable with distributed teams, including remote collaboration with US partners. Clinical abstract outputs include Paris Vascular Insights (PVI), JET OPEN (Osaka), and American Thoracic Society (ATS, Orlando).
Contact
For research leadership, regulated medical AI, collaborations, invited talks, or consulting, email with a subject line below (opens your mail client).
Or write directly: christonik@gmail.com