★  Featured today 1 / 15

Explore this system
Decision Support Systems · AI & Data Science

AI that makes people better decision-makers — not AI that decides for them.

I build systems that sit next to human judgment, not in front of it: transparent models, honest uncertainty, and interfaces that explain themselves. At the center is a mission — augmenting the people who care for patients — with the same principle proven across workforce, credit, aerospace, and customer-retention decisions.

See the work Read the philosophy
Philosophy

On Human–AI Relationship

Every project on this site is built on the same working belief: the goal of applied AI is not to replace a person's judgment but to make that judgment sharper, faster, and better informed. A model that quietly makes the call is a liability. A model that shows its reasoning, flags what it doesn't know, and hands a clear decision back to a human — that's a tool worth trusting.

01

Augment, don't replace

Systems are designed as decision support — surfacing patterns, risks, and probabilities for a human to weigh, rather than issuing verdicts a person is expected to accept on faith.

02

Transparency over accuracy theater

A model's headline metric is never the whole story. Every project here documents what the data can't tell you — missingness, top-coding, sampling bias — alongside what the model gets right.

03

Built for the person who has to act

A dashboard nobody reads and a model nobody trusts are the same failure. Every system is designed around the actual decision-maker: what they need to see, in language they already use.

The Pento‑Helix

Underneath those three principles sits a broader frame I work from: the Pento‑Helix. Pento, for five — a structure meant to stay portable, flexible, and adaptable across contexts, not fixed to one project. Helix, because it behaves like DNA: one constant structure that expresses itself differently everywhere it's deployed, the way RNA carries out what DNA encodes.

01Human–AI collaboration and synchronization
02Human–AI harmony
03Each side strengthening the other's skills and expertise, while understanding the other's limitations — mutual respect
04Shared social responsibility for what gets built
05Progressive, together — the constant thread across every project, and what this site's mark is built around: a dot held steady at the center — the Tao — with a spiral of motion, a Milky Way, turning around it.

The first four pillars above are how the Pento‑Helix takes shape in the Human–AI Augment Systems work collected on this site. Other projects will express their own version of the same four; the fifth never changes.

Selected Work

Nine decision-support systems, grouped by the human problem they solve

Every system here rests on one belief: AI should make people better decision-makers, not replace them. That belief has a home — the Nurse Augmenting System, human-centered patient care and the core of the AEGIS programme. The rest prove the same principle holds across domains: two are deployed as live, working apps, and all carry honest, full write-ups of what each model can and can't tell you.

NEW SERIES · NOW LIVE

Diagnostic Augmenting Systems

A new series of screening tools grounded in the physiological kinetics of the disease itself — each returns an interpretable three-class risk stratification, surfaces its own uncertainty as a confidence score, and flags the at-risk patient while a clinician stays in command. First release: DRAS, diabetes risk from a 2-hour OGTT.

🌍 Plans across 24 countries worldwide Kinetics-grounded Human-in-command
Enter the Diagnostic Augmenting Systems
Diagnostic Stations · Clinical AI

Clinical Diagnostic Stations

Validated AI systems for the eye, the breast, the neurological system, and the skin — spanning voice and motor assessment for Parkinson’s Disease, false-colour MRI classification for Alzheimer’s, and zone-adaptive dermoscopy triage for skin disease. Each reads its own clinical signal, each reports an honest score, each with a defer-to-clinician confidence zone and a public learning station. A clinician always stays in command.

Healthcare · Ocular · Now Live

AEGIS Ocular Station — Eye-Disease Screening

One pass over a retinal fundus photo screens for diabetic retinopathy, AMD, glaucoma and hypertensive retinopathy — the ODIR baseline raised by four dedicated-data specialists. A public “Spot the Disease” learning game and screening aid.

0.90–0.99
ROC-AUC across
four specialists
4
dedicated-data
specialists (D·A·G·H)
1 pass
read by every
specialist at once
Open the Ocular Station
Healthcare · Oncology · Now Live

AEGIS Breast — The Unified Breast Station

One engine routes each breast modality — mammogram, histology, ultrasound, fine-needle-aspirate — to its own specialist, each with an honest score and a defer-to-human grey zone. A public “Spot the Finding” clinical game.

0.9955
top-specialist
ROC-AUC (WDBC)
Multi
modality router
(imaging + tabular)
Grey-zone
defer-to-clinician
on the rim
Open the Breast Station
Healthcare · Neurology · Now Live

AEGIS Parkinson — Voice & Drawing Assessment System

A validated clinical AI station that listens to a person’s voice and reads their drawing tests to assess Parkinson’s Disease — with a live public quiz, a companion textbook, and a structured laboratory sheet designed for biomedical engineering courses. A clinician always stays in command.

0.846
champion AUC
MLP Neural Net
11
engines in
consensus board
Book + Lab
textbook & lab sheet
institutional ready
Open the Parkinson Station
Healthcare · Neurology · MRI · Now Live

AEGIS NeuroScan — Alzheimer’s MRI Classification

A validated 4-engine SVM ensemble classifies Alzheimer’s Disease severity from brain MRI — with pixel-level false-colour tissue mapping, NDI scoring on a 0–100 scale, and six annotated demo cases spanning early preservation to advanced atrophy. A clinician always stays in command.

0.8511
champion AUC
4-engine ensemble
False-colour
MRI visualiser
pixel-level rendering
Grey-zone
defer-to-clinician
NDI 42.5–50.0
Open the NeuroScan Station
Healthcare · Dermatology · Now Live

AEGIS SkinScan — Skin Disease Triage Classifier

A zone-adaptive 3-engine ensemble classifies dermoscopic images across 7 skin disease categories — with an independent melanoma safety flag, confidence-zone triage, and 7 annotated demo cases from the HAM10000 dataset. A dermatologist always stays in command.

0.9131
HIGH-zone AUC
3-engine ensemble
7 classes
MEL · BCC · AK · VASC
NV · BKL · DF
Mel-flag
independent melanoma
safety alert
Open the SkinScan Station
Luminous data-lattice human heart with an ECG waveform flowing into a neural network
Domain 01 · Biomedical Intelligence

Biomedical Intelligence

Predicting a patient’s risk so a clinician can act early — AI that augments the nurse, never replaces her.

Healthcare · Core Project

Nurse Augmenting System for Biomedical Intelligence and Personalized Care Augmentation

Eight independently trained models — spanning linear, tree, kernel, boosting, and ensemble methods — converged on the same five predictors of 30-day heart-failure readmission. A hand-engineered clinical score (HCAS), not raw accuracy, was the point.

8 / 8
models agree on
the top 5 predictors
5
predictors in the
HCAS clinical score
30-day
readmission
window
View case study
Healthcare · Core Project

Medication Adherence & Healthcare Claims Analysis

A star-schema claims model and a five-model adherence benchmark on 24,084 diabetes and hypertension patients — surfacing that non-adherent patients cost 2.5× more, and including an honest report of the one model, out of five, that quietly failed.

81.98%
champion accuracy
(Gradient Boosting)
2.5×
claims cost of a
non-adherent patient
24,084
patients
modeled
View case study
A clinician at a dark command desk with a glowing gold K-wave ECG trace on the wall of cardiac monitors behind her
AEGIS Flagship · Wearable

Wearable

One wearable, three tiers — Lifestyle for everyone, Health Continuous Monitoring for the metabolic picture, and Clinical for the hospital desktop.

Lifestyle

AEGIS Protector — Your Personal 25/8 Body Guard

The everyday wearable: it learns your normal, catches the earliest electrical signs of a heart attack — the K-wave — before you feel a thing, and gets help moving. One device from the wrist to the ward, with a live interactive demo.

learns you
personal baseline,
always adapting
K-wave
early cardiac
warning
home→ward
one device,
everywhere
View case study
Clinical

AEGIS Clinical — Advanced Cardiac Algorithm & NDI (Non-Dimensional Index)

The clinical sibling that unlocks on the same device and integrates into the hospital desktop — one clinician watching every patient on a population danger scale. A self-learning K-wave engine validated on two independent, physician-labelled datasets.

~98%
accurate on the early
ischaemia signature
66,951
ECGs, two
independent datasets
96%
from a single lead —
a watch is enough
View case study
Health Monitoring

AEGIS Health — The Non-Invasive Glucose Watch

The continuous-monitoring tier: a non-invasive glucose reading from an infrared sensor — no lancet, no strip — plus SpO₂ and a paired 4-lead Bluetooth ECG. One adaptive engine learns your body, and a transparent-glass base station shows all four technologies at once. Two live 3D demos.

no prick
glucose from
light, not blood
index A
dimensionless
diabetic score
learns you
on-device, four
adaptive axes
View case study
Emotional

AEGIS Emotional — The Affective Wearable

The psychophysiological tier: a daily electrode band reads stress, arousal, cognitive load and recovery from skin conductance, HR/HRV, temperature and motion. One adaptive grey-zone engine learns your calm and defers to a human when unsure — shown on a live human-face Sentinel.

~0.92
AUC, honest subject-
independent (LOSO)
Ψ index
nondimensional
arousal score
defers
to a human on
the grey rim
View case study
Cutaway turbofan engine glowing from green through amber to red along its core
Domain 02 · Pre-Emptive Maintenance

Pre-Emptive Maintenance

Read the life that’s left in a machine, and service it before it fails.

Earth at night threaded with converging satellite orbital paths and a near-miss warning halo
Domain 03 · Orbital & Space Safety

Orbital & Space Safety

An advance-warning signal for satellite-collision risk — a small voice in a very large ocean.

Branching decision tree of light beside a balanced scale, one path highlighted in gold
Domain 04 · Decision Intelligence

Decision Intelligence

Weighing risk to make a fair, transparent call — in lending, retention, and livelihoods.

Finance

AI-Powered Loan Default Risk Decision Support System

A 12-model benchmark on 29.9 million LendingClub records — where an obvious leak was caught and removed, and a suspiciously perfect score that remained was investigated and explained rather than just reported.

View case study
Marketing

AI-Powered Marketing Decision Support System for Customer Retention

Nine models were benchmarked on IBM's Telco Churn dataset — and the highest-accuracy one wasn't picked. The production model was chosen for matching a real business threshold instead.

View case study
📬 Get new-app alerts ✉ Contact us