Health Informatics · Data Science

Nooruldeen
Taffar

Clinician turned data scientist building medical foundation models for cardiology — multimodal AI for diagnostic medicine.

MBChB MSc Health Informatics PGDip Data Science
View academic work GitHub
Edinburgh, UK
Nooruldeen Taffar

From the bedside to the model.

A physician who moved into computational medicine to build diagnostic tools that scale. I value vision-focused teamwork and global contribution to diagnostic medicine.

01
2012 – 2018 · Iraq

MBChB — College of Medicine

Clinical rotations across internal medicine, emergency, surgery, paediatrics, obstetrics and psychiatry.

02
2018 – 2022 · Iraq

Practising physician

Emergency and outpatient care across three hospitals; led clinical audits and service improvements.

03
2022 – 2024 · Swansea

MSc Health Informatics

Systematic reviews, clinical coding, telemedicine, secondary health data and gamification of healthcare.

04
2024 – now · Edinburgh

PGDip Data Science for Health

R / Python, statistics, dashboards (OECD, NHS Scotland), Shiny for Python, Apache Spark, SQL and NLP.

Multimodal AI for automated cardiovascular diagnosis.

The thesis: diagnostic systems that fuse the signals using multimodal methods to auto-diagnose CVDs and run in parallel with conventional clinical workflows.

ECG signal
waveform
EHR records
structured
Demographics
context
Foundation model
multimodal · multilayer
CVD diagnosis
decision support
Active · ongoing

PROBAST+AI scoring tool

A structured instrument for critically appraising medical foundation models — risk of bias and applicability for clinical ML.

Planned · 2027

PhD — Medical Informatics

Multimodal ML for automated cardiovascular diagnosis, fusing ECG, EHR and demographic data.

Fifteen projects across health informatics & data science.

MSc Mar 2022

Health Classifications & Records — ICD-10 & EHR

ICD-10 implementation and EHR systems' impact on healthcare delivery and patient experience.

ICD-10 EHR Clinical coding
View on GitHub
MSc Apr 2022

Using Secondary Health Data

Population-level analysis of secondary health data utilization in public-health research.

Population health Secondary data
View on GitHub
MSc May 2022

Mobile Referral App (Iraq)

A mobile prototype to modernize medical referrals across Iraqi healthcare.

Mobile Prototype Iraq
View on GitHub
MSc Jun 2022

Consumer Health Websites

Trustworthiness vs readability of health information using established standards and metrics.

Readability Trust Literacy
View on GitHub
MSc Jun 2022

Knowledge Management — Hospital Intranet

Design and implementation of a hospital intranet for improved clinical workflow.

Knowledge mgmt Intranet
View on GitHub
MSc Jul 2022

Wearables for CVD Patients

Phenomenological study design exploring cardiovascular patients' experiences with wearables (Morriston).

Wearables Cardiology Qualitative
View on GitHub
MSc Sep 2022

Dissertation — Gamification Ethics

Systematic PRISMA review of ethical challenges in healthcare gamification.

PRISMA Ethics Review
View on GitHub
PGDip May 2025

Scottish Injuries Analysis

Injury patterns across Scottish health boards in R and Python, with JSON data processing.

R Python JSON
View on GitHub
PGDip May 2025

Datathon — Teleconsultation (Portugal)

OECD healthcare data analysis to inform teleconsultation policy recommendations.

OECD R Policy
View on GitHub
PGDip May 2025

COVID-19 in Scotland

Pattern analysis of COVID-19 in Scotland using Python and interactive visualizations.

Python Plotly Viz
View on GitHub
PGDip May 2025

OECD Healthcare Dashboard

Interactive Shiny-for-Python dashboard analyzing OECD healthcare utilization.

Shiny Python Dashboard
View on GitHub
PGDip May 2025

mHealth Adherence & NHANES

Statistical analysis of mHealth impact on medication adherence and NHANES exploration.

Statistics NHANES R
View on GitHub
PGDip Jun 2025

Diabetic 30-Day Readmission

Gradient-boosted-trees classifier on 100,000+ hospital encounters using Apache Spark.

Apache Spark GBT 100k+ records
View on GitHub
PGDip Jun 2025

NHS Scotland Prescribing (2015–2023)

SQL and MariaDB analysis of nine years of GP prescribing data with ERD design.

SQL MariaDB ERD
View on GitHub
Independent 2026

GenECG-Diagnostic

Two-stage YOLO + Vision Transformer pipeline that localises ECG leads and predicts SNOMED-CT diagnoses from raw 12-lead images.

Computer Vision PyTorch Healthcare AI
View on GitHub

Education, experience & toolkit.

Education
2024 – ongoing

PGDip Data Science for Health & Social Care

University of Edinburgh

8/10 modules: big data analytics, databases, applied software dev, statistics, research foundations.

2022 – 2024

MSc Health Informatics

Swansea University

Gamification, knowledge management, systems & tech, secondary health data, classifications & records.

2012 – 2018

MBChB

College of Medicine, Iraq

Rotations across internal medicine, emergency, surgery, paediatrics, obstetrics and psychiatry.

Experience
2023 · Remote

Medical Writing Intern — Klarity Health

Clinically accurate, patient-friendly content; iterative editorial work with medical reviewers.

2018 – 2022 · Iraq

Physician — Emergency & Outpatient

Rotator / chief / emergency roles across hospitals and clinics; led audits and service improvements.

Technical toolkit
Languages & query
R / tidyverse Python / pandas SQL · MariaDB
Big data & ML
Apache Spark · sparklyr GBT classification EHR pipelines NLP
Visualization
ggplot2 Plotly Shiny for Python Tableau
Methods & workflow
PROBAST+AI Reproducible workflows Git / GitHub Research design
Memberships BCS · RSS · Iraqi Medical Association
Languages Arabic (native) · English (fluent)
Based in Baghdad, Iraq

Open to collaborations.

Particularly in medical informatics and AI for cardiology. Let's talk.