Data science · Health · Environment

I follow questions about health and the environment wherever the data leads.

I’m Ahmed, a data scientist with a background in medical sciences, working between Abu Dhabi and Cairo. Most of my work sits where health and the environment meet, and I like problems that cross fields: how buildings affect the people inside them, which treatments are worth paying for, how care could start before the clinic.

Wind crosses a city block. Air turns from teal to orange as it passes a busy road, and carries the pollution downwind.

What I do

Rooted in health and environment, open to where the questions lead
01

Connect environment and health

How air, heat and the buildings we live in affect the people inside them, from exposure research to building simulation.

Medical sciences · AEROME · Preprint, 2026
02

Build with data

Machine learning, statistics and simulation, on anything from genomic data to building energy models.

Python · SQL · LightGBM · DesignBuilder
03

Make it useful

Turning findings into tools and clear recommendations that patients, clinicians and building developers can act on.

PreVisit · PassiveCheck · AEROME client work

Selected work

Modelling · Simulation · Product
Machine learning · Precision medicine Complete

Beyond Tissue of Origin: Quantifying the Added Predictive Value of Genomic Features for Anti-Cancer Drug Response

An end-to-end machine-learning study of 8 cancer drugs across ~700 cell lines, joining GDSC drug-response data (Wellcome Sanger) with DepMap genetics (Broad Institute). The question: does reading a cancer’s genes predict how well a drug will work, better than knowing which organ the cancer came from?

Result
Genes beat the organ-only baseline for 7 of 8 drugs, with correlations of 0.42–0.76 on a locked test set. The models also rediscovered known biomarkers unprompted: BRAF for dabrafenib, TP53 for Nutlin-3a, BCL2 for venetoclax.
Rigour
Locked 20% test set, no leakage, bootstrap 95% intervals, and SHAP explanations checked against known biology.
Built with
Python · LightGBM · SHAP
Built environment & health · AEROME Client work

Buildings that are healthier to live in and cheaper to run

Site climate analysis and DesignBuilder simulations of energy and thermal comfort for two delivered AEROME projects: Zawaya Mall in 6th of October City, Egypt, and a 275 m² villa in Dubai.

Result
Dubai villa: 67.9 kWh/m²/yr, −53% against the Al Sa’fat code baseline. Zawaya Mall: −42% cooling cost, −35% total energy, about 79,000 EGP a year saved.
For occupants
The villa stays at 28°C indoors on a 44°C day; peak temperatures in the mall’s hottest units fall by 41%.
My role
Climate and thermal simulation data, site analysis, code and market benchmarking, and the business case. Architecture by AEROME’s design team.
Tools
DesignBuilder · Autodesk Forma
Digital health · Product v0.1 demo

PreVisit: the patient’s history, taken before the visit

A bilingual Arabic–English pre-visit history that becomes a structured brief a doctor can scan before the consultation, for outpatient clinics in the UAE and Egypt. The evidence, UAE law and economics behind it are written up alongside the code.

Result
A working demo on synthetic data: one intake engine behind a WhatsApp-style chat and a secure web form, four questionnaires, layered consent, rule-based red-flag checks, and encrypted, audited storage with FHIR export.
Honest scope
Not a clinical product: no real patient data and no pilot yet. Every claim in the write-up is labelled verified, hypothesis or illustrative.
Built with
Python · FastAPI · Next.js · PostgreSQL

Also building

Smaller tools and product ideas

PassiveCheck

70% built

Drop a pin on your plot and see roughly how much energy three tiers of passive design could save on a typical building there, using NASA POWER climate data for Abu Dhabi, Dubai and Cairo.

Is a gene test before a common heart drug worth paying for?

In progress

A cost-effectiveness model of CYP2C19 testing before prescribing clopidogrel, compared with standard antiplatelet therapy. Preprint in preparation.

Tempus AI deep dive

In progress

A look inside Tempus AI, the precision-medicine company: how it turns clinical and genomic data into a business, and where that could go next.

Whoop & Oura analytics

Streamlit · OpenAI

Upload a Whoop or Oura export and ask questions about your sleep, recovery and training in plain English.

Habit dashboard

HTML · PWA

An always-on dashboard at home that scores the day out of nine across prayer, training, nutrition and study.

Writing

Papers and preprints
2026 · Preprint
How Buildings Affect Human Health: Evidence from Air Quality, Stress, and Epigenetics

A narrative review of how exposure in the built environment affects human health. Zenodo, DOI 10.5281/zenodo.22012930.

In progress
Cost-effectiveness of CYP2C19-guided antiplatelet therapy

A health-economic evaluation of genotyping before prescribing clopidogrel. Preprint in preparation.

About

Abu Dhabi · Cairo

Trained in medical sciences, I moved into data science to work on questions that cross fields: how buildings and air affect the people inside them, which treatments are worth paying for, and how care could start before a clinic visit. I like problems that don’t sit neatly in one discipline, and I’m happy to follow one somewhere new.

I co-founded AEROME, an environmental design studio in Cairo and Abu Dhabi, and I’m studying for an MSc in Health Data Science at the University of Birmingham Dubai.

Now
MSc Health Data Science, University of Birmingham Dubai
Now
Co-founder, AEROME
Learning
CFA Level I
Before
M42 (G42 × Mubadala), Abu Dhabi
Before
Data analyst, Masterpiece Engineering Services, Abu Dhabi
Study
PGCert Data Science & Business Analytics, UT Austin
Study
BSc Medical Sciences, Oxford Brookes University