Christopher Duym

Back to projects

Senior project · 7-person team · Sep–Dec 2024

Pointing Span Test Web App

A remote cognitive assessment platform for clinicians to administer tests and analyze patient results.

My role: Owned database setup, data-aggregation pipelines and user-facing reports.

  • Python
  • Django
  • SQLite
  • Bootstrap CSS

Developed a Django web application that enables clinicians to generate secure links for patients to complete a digitized pointing-span cognitive assessment remotely.

Designed an accessible test interface that presents stimuli audibly and visually, captures responses through an on-screen keyboard, and records latency metrics.

Built a clinician dashboard for creating tests, monitoring completion status, and reviewing individual test analytics.

Created a data aggregation pipeline that summarizes results and compares individual performance with collected averages to support pattern discovery.

How it works

The path of a single test, from the clinician creating it to the results report.

  1. Clinician creates a test

    A signed-in clinician enters the patient's ID and age and gets a unique test link to send to the patient.

  2. Patient takes the test remotely

    Digit and mixed letter-and-digit sequences are presented visually and read aloud; the patient answers on an on-screen keypad and the latency of every keypress is recorded in milliseconds.

  3. Database

    Built by me

    Stimuli, tests, and responses sit alongside the two statistics tables I added. Every migration seeds the database with the stimuli and 120 synthetic completed tests spread evenly across six age groups, so averages are meaningful before real patients exist.

    • stimulus
    • test
    • response
    • individual_statistics
    • aggregated_data

  4. Aggregation pipeline

    Built by me

    Turns a completed test into accuracy and average latency for the digit and mixed tasks, rolls those up into averages for each age group, and ranks the patient as a percentile against peers of the same age.

  5. Clinician dashboard and reports

    A doctor-only dashboard charts the results, lists open and completed tests, and exports them as CSV or JSON. I added the controls that aggregate a test by patient ID and recompute percentiles and age-group averages.

Data flows top to bottom. Highlighted steps are the parts I built.
  • Source private