Speaking
Speaking across clinical AI, nephrology, cardiology, and translational medicine.
Recent invited lectures sit beside earlier conference work, with awards and public materials kept close to the relevant talk.
Archive span
15 talks across 4 years
Recognized work
10 awarded talks
Public materials
1 talk with slides or flyer material
Archive
Speaking archive by year.
Start with the full record, then narrow by year. Awards and public materials stay visible without dominating the list.
2026
From Risk Scores to Agents: What "AI" Really Means in Nephrology
A clinician-facing lecture that moves from classical risk scores and machine learning toward deep learning, retrieval, and agentic systems in nephrology.
Invited lecture
Slides, Flyer
Frames AI as a continuum of clinical tools rather than a single novelty category.
Forecasting ICU Acute Kidney Injury with Actionable Lead Time Using Interpretable Machine Learning: Development and Multi-Center Validation
A multi-center validation presentation on interpretable machine learning for forecasting ICU acute kidney injury with clinically actionable lead time.
Conference presentation
No public materials
Moves renal prediction from retrospective risk labeling toward earlier, interpretable warning that can support bedside decisions.
Generative Echocardiography from ECG and Clinical Context: Uncertainty-Aware Synthesis of a Full 9-View Echo Study
A generative multimodal AI presentation on synthesizing a full nine-view echocardiography study from ECG and clinical context with uncertainty awareness.
Conference presentation
No public materials
Extends cardiac multimodal modeling from parameter prediction toward uncertainty-aware image synthesis and clinically legible review.
CheXGPT: A Multimodal Reasoning Language Model for Chest X-Ray Labelling and Report Generation
A mentee-presented multimodal reasoning model for chest X-ray labelling and report generation.
Mentee presentation
No public materials
Shows how supervised mentorship work can turn medical vision-language models into concrete radiology reporting tasks.
Chest X-Ray-Predicted Age as a Biomarker for Mortality: Development and Validation of a Deep Learning Model
A mentee-presented deep learning project validating chest X-ray-predicted age as an imaging biomarker for mortality.
Mentee presentation
No public materials
Uses radiographic aging as a clinically interpretable bridge between image-derived representation learning and mortality risk.
CardioAtlas: An Evidence-Grounded, Multimodal Agentic AI Co-Pilot for Cardiologists
A mentee-presented agentic AI co-pilot for cardiology, grounded in evidence and multimodal clinical context.
Mentee presentation
No public materials
Positions agentic AI as a clinical reasoning layer that must stay grounded in evidence, context, and cardiology workflow.
Multimodal Deep Learning Model to Differentiate Viral from Bacterial Pneumonia Using CXR and Early Clinical Data
A mentee-presented multimodal model combining chest X-ray and early clinical data to distinguish viral from bacterial pneumonia.
Mentee presentation
No public materials
Connects early clinical variables with imaging so pneumonia classification can be framed closer to the first decision point.
2025
Does Refined Iron Deficiency Classification Improve Risk Prediction in Congestive Heart Failure? A Retrospective Cohort Study
Retrospective cohort work testing whether refined iron-deficiency phenotyping improves risk prediction in congestive heart failure.
Conference presentation
No public materials
Shows how clinically interpretable phenotyping can materially sharpen downstream mortality risk framing.
The Impact of Chronic Anticoagulation on ICU Outcomes in Ischemic Stroke Patients: A Retrospective Cohort Analysis from MIMIC-IV
A retrospective MIMIC-IV analysis studying how chronic anticoagulation relates to ICU outcomes in ischemic stroke.
Conference presentation
No public materials
Applies critical-care cohort analysis to a treatment question with direct neurological and ICU relevance.
GPU-Accelerated Real-Time Prediction of Critical Complications in MI Patients Using Multi-Modal AI: A Novel Continuous Early Warning System
A hardware-aware multimodal early-warning concept for myocardial infarction patients, built around real-time prediction of critical complications.
Conference presentation
No public materials
Connects continuous prediction, multimodal inputs, and systems design rather than treating clinical AI as a static model artifact.
Multi-Modal AI for Predicting Echocardiographic Parameters: Transforming Cardiac Assessment with ECG and Vital Signs
A multimodal prediction talk on estimating echocardiographic parameters from ECG and vital-sign signals.
Conference presentation
No public materials
Pushes multimodal prediction toward clinically useful measurement support rather than isolated classification tasks.
Dynamic Prediction of Mortality in ACS Patients
A time-aware prediction project focused on mortality risk in acute coronary syndrome patients.
Conference presentation
No public materials
Brings dynamic risk prediction into acute cardiovascular decision-making rather than relying only on static baselines.
2024
2023
The Effect of NPIs on 17 Notifiable Infectious Diseases in Poland
An early epidemiology presentation on how non-pharmaceutical interventions reshaped incidence patterns across notifiable infectious diseases in Poland.
Conference presentation
No public materials
Earlier surveillance work that later matured into publication-ready population-health analysis.
Exophiala dermatitidis: A Case Report of a Rare Fungal Infection
A clinical case report on a rare fungal infection, presented within the early WIMC conference work.
Conference presentation
No public materials
Keeps the speaking record grounded in core clinical medicine alongside later systems and modeling work.
Contact
Available for invited talks, workshops, and clinical AI teaching.
If the audience is clinical, translational, or multidisciplinary and the topic overlaps with prediction, multimodal systems, nephrology, or applied AI infrastructure, email is the simplest route.
On this page