Skip to content
In-house buildVerified Live

Multi-disease risk screening model

MAUK R&D · applied machine learning

Risk scores for ten conditions from 25 routine measurements, trained on 281,000 records and private enough to run on the patient's own device.

A screening tool that takes 25 routine health measurements and returns a risk score for ten conditions at once, each with its own confidence figure. It runs entirely in the browser, so a prediction takes about a millisecond and no health data is ever sent anywhere. It is a research and teaching build, not a medical device.

What it is
AI system
Industry
Healthcare
Status
Live
Year
2026

A research and teaching build. It is not a medical device and gives no medical advice.

Disease risk screening
Risk report screen: an overall high-risk verdict with confidence, a chart of the underlying model probabilities, and cards for type 2 diabetes, hepatic steatosis and metabolic syndrome with confidence scores
The risk report, with a card for each condition.

The problem

What needed solving

Screening models are usually judged on one headline accuracy number, which hides the thing that matters in health: missing a sick patient is far worse than sending a healthy one for a second look. The model also had to be honest about its own limits, and light enough to run without a server.

The solution

What we built

A four-step health questionnaire that ends in a full risk report, backed by a model tuned to catch at-risk patients and a diagnostics panel that shows exactly how it performs.

Core functionality

  • Ten conditions scored at once, each with a confidence figure
  • 25 inputs across demographics, vital signs, lab results and lifestyle
  • Five ready-made patient profiles for quick testing
  • Built-in BMI calculator and reference ranges on every field
  • Each condition names the measurements that drove its score
  • A diagnostics panel showing accuracy and exactly where the model gets it wrong

Screens

More screenshots

The built-in diagnostics: where the model is right, and where it is wrong.
Step one, pre-filled from one of the built-in patient profiles.
Step two collects the lab values, each with its reference range.
Step three covers lifestyle and family history.
Each condition names the measurements that drove its score.
Phone layout.

Captured from the live site, a demo workspace, or the product running on sample data. No customer records, passwords or private admin data are shown.

Business impact

What it achieved

  • 86.88%

    accuracy on patient records the model never saw

    Against 87.56% on its training data, a gap of just 0.69 points.

    Verified
  • 95.40%

    of at-risk cases correctly flagged

    233 missed cases against 4,834 correctly detected in the 8,000-record test set.

    Verified
  • 281k

    patient records in the source dataset

    Verified
  • ~1 ms

    per prediction, on the device

    With no server call.

    Verified

What sets it apart

Why it works

  1. 01

    Built to miss fewer sick patients

    The model deliberately accepts some extra false alarms so that far fewer at-risk patients slip through, and shows that trade-off in plain numbers.

  2. 02

    No health data leaves the device

    Predictions run in the browser itself, privately and instantly, with no server involved.

  3. 03

    Clinical rules on top

    Established thresholds override the model where medicine is clear, such as an HbA1c result in the diabetic range.

  4. 04

    Honest about its weak spot

    Heart disease is rare in the data and its scores are less precise. The project says so instead of hiding it behind overall accuracy.

For your business

What we can build for you

We build systems like Disease risk screening through our Machine Learning & Computer Vision service. For your business, that can mean:

  • Prediction & scoring models

    Models that score risk, predict outcomes or classify messages.

  • Computer vision

    Detection and tracking in images and video, tested in hard conditions.

  • Explainable results

    Every prediction comes with the reasons behind it.

  • Honest evaluation

    Accuracy reported on unseen data, broken down by the cases that matter.

Capabilities

Related projects

Similar projects

  • Live
    Client projectVerified

    Fort Healthcare Solutions website and operations platform

    Medical billing & revenue cycle company · Virginia, USA

    A new website for a U.S. medical billing company, and the operations platform built to run its sales, client work, people and pay.

    13service pages on the new website

    Custom software · Chatbots · Automation

    View project
  • Live
    MAUK productVerified

    IPC360 infection prevention assessment platform

    MAUK product · hospital infection prevention & control

    Hospitals assess their infection prevention programme, see the gaps and track every improvement to a report leadership can read.

    Custom software · Data

    View project
  • Demo buildVerified

    Apnadoc doctor booking and online consultation platform

    Demo build · patient and doctor platform for Pakistan

    Patients find a doctor, see real open slots and book a clinic visit, video call or chat. Doctors run every clinic they work at from one profile.

    3ways to see a doctor, each with its own fee

    Custom software

    View project

Discuss a similar project

Building something like Disease risk screening?

We'll tell you what we'd reuse, what we'd change and what it would take.

Chat on WhatsApp