Desktop UI Design

KMI Gait Analysis Interface

Turning complex gait data into clear clinical insights

Duration

10 weeks

Team

Liu Wei, Ph.D (research mentor)

Tools

Figma

My Role

UI/UX Designer - conceptualization, user research & analysis, lo-fi & hifi prototyping

Challenge

Gait assessment in clinics is often limited to visual observation and checklists, which lack consistency. To address this, KMI—backed by an NIH-funded Alzheimer’s and fall-prevention initiative—developed a motion-capture system that generates detailed gait metrics. Yet the system’s raw numerical outputs were too complex for clinicians to interpret quickly, reducing its effectiveness for diagnosing fall risk and muscle weakness.

oPportunity

Design an interface that translates raw gait data into clear, actionable insights, supporting faster, more accurate clinical decision-making and improving fall-prevention outcomes.

Desktop UI Design

KMI Gait Analysis Interface

Turning complex gait data into clear clinical insights

Duration

10 weeks

Team

Liu Wei, Ph.D (research mentor)

Tools

Figma

My Role

UI/UX Designer - conceptualization, user research & analysis, lo-fi & hifi prototyping

Challenge

Gait assessment in clinics is often limited to visual observation and checklists, which lack consistency. To address this, KMI—backed by an NIH-funded Alzheimer’s and fall-prevention initiative—developed a motion-capture system that generates detailed gait metrics. Yet the system’s raw numerical outputs were too complex for clinicians to interpret quickly, reducing its effectiveness for diagnosing fall risk and muscle weakness.

oPportunity

Design an interface that translates raw gait data into clear, actionable insights, supporting faster, more accurate clinical decision-making and improving fall-prevention outcomes.

Desktop UI Design

KMI Gait Analysis Interface

Turning complex gait data into clear clinical insights

Duration

10 weeks

Team

Liu Wei, Ph.D (research mentor)

Tools

Figma

My Role

UI/UX Designer - conceptualization, user research & analysis, lo-fi & hifi prototyping

Challenge

Gait assessment in clinics is often limited to visual observation and checklists, which lack consistency. To address this, KMI—backed by an NIH-funded Alzheimer’s and fall-prevention initiative—developed a motion-capture system that generates detailed gait metrics. Yet the system’s raw numerical outputs were too complex for clinicians to interpret quickly, reducing its effectiveness for diagnosing fall risk and muscle weakness.

oPportunity

Design an interface that translates raw gait data into clear, actionable insights, supporting faster, more accurate clinical decision-making and improving fall-prevention outcomes.

Overview

The final dashboard supported the entire clinical workflow, from deviations to treatment:


Menu Sidebar:

Quick access to past reports for patient progression.


Gait Deviation Table:

Color-coded overview of gait issues by significance.


Diagnosis page:

Breakdown of deviations with most likely causes.


Treatment page:

Recommendations aligned with each diagnosis.


Range of Motion (ROM) + stickmen diagrams:

Gait phases visualized for easy reference.


3D muscle model:

Simplified and positioned alongside tables to provide visual anatomical context.

Stage 1: Research

To explore this opportunity further, I conducted research on current diagnostic processes for gait abnormalities.

Inefficient & inconsistent diagnostic process

Traditionally, medical professionals rely on observational gait analyses, which uses checklist forms that are limited in accuracy and vulnerable to observer bias.



Current tools often suffer from excessive complexity or lack user-friendly interfaces, leading to time-consuming workflows and potential diagnostic errors.

JAKC's Observational Gait Analysis

Other Observational Gait Analysis Form

Raw data

Audience

Based on my literature research and guidance from my mentor, I narrowed down two types of users:


Experienced Healthcare Professionals

  • Motivated to understand comprehensive patient deviations quickly to faster diagnose gait problems and track improvement.

  • Interested in concise and easy-to-read data with emphasis on the most useful features for diagnosis, visual comparison of various data forms, and easy toggle of patient reports of different dates.


Less Experienced Users:

  • Motivated to use the application to generate accurate diagnosis and treatment plans

  • Interested in more detailed diagnoses, suggestions for treatment, informational diagrams, and links to more resources.

I needed to make a design that emphasized with both user groups, to maximize the functionality of this interface.

Key takeaway

No interface to bridge raw technical data with the way clinicians make real diagnostic decisions.

Problem Statement

How might we design an interface that transforms complex gait data into clear, actionable insights, so clinicians can efficiently move from observation to diagnosis and treatment when assessing fall risk and muscle weakness?

Stage 2: Design

User Flow

Comparison Between Original and Proposed User Flows

Seeing the problems of the current clinical diagnostic process, I created this flow comparison with feedback from my mentor. I narrowed down the user flow to the following:


  1. Deviations: Overview of problems using color-coded indicators.

  2. Diagnosis: Breakdown of deviations with likely causes.

  3. Treatment: Recommendations for each major deviation.


This user flow mirrored clinicians’ real decision-making process and workflow, making the system intuitive and reducing the learning curve. Because of the dense user flow, I prioritized which features should appear simultaneously to reduce navigation steps and support clinicians in making quick, informed comparisons.

Initial prototype

Draft 1

The first prototype focused primarily on deviations mapping, with a simplified layout to explore how motion-capture data could be visualized. At this stage, the goal was simply to represent gait deviations in an accessible way, without yet addressing diagnosis or treatment.


This iteration featured:

  • Patient ROF Comparison Line Graph with toggles for joint and side of body, comparable to normal ROM

  • Gait Deviations chart with measurements of concern, color coded by severity.

  • 3D model for easy visualizing of relevant joints.

Feedback & Painpoints

Through mentor feedback and early reviews, several issues surfaced that highlighted the need to go beyond visualization and support the full diagnostic workflow.


  • The Gait Deviations Chart became cluttered when displaying multiple deviations, making interpretation difficult.

  • Causes and treatment information were not sufficiently emphasized; hover popups in the deviation table did not provide enough detail.

  • The 3D skeleton model was too complex and distracting compared to the rest of the interface.

Gait Deviation Graph

Causes & Treatment

3D Skeleton

Later iterations

Based on this feedback, I expanded and refined the interface:

  • Diagnosis and Treatment Pages → Added as dedicated sections to give clinicians thorough, structured reports. This addition stemmed from mentor guidance that the system should align with clinical decision-making, not just visualization.

  • Simplified Deviation Table → Labels were minimized and reorganized to reduce clutter while maintaining clarity.

  • 3D Model Refinement → Simplified the anatomy view and relocated it beside the table, enabling easier comparison between muscle visuals and gait data.

Gait Deviation Graph

Causes & Treatment

3D Skeleton

These iterations moved the dashboard from a data-only tool toward a platform that delivers actionable insights, supporting clinicians in both identifying problems and recommending treatments.

Stage 4: Future Directions

Next steps

  • Conduct usability testing with clinicians using real patient data.

  • Integrate prototype with KMI’s motion-capture hardware for live validation.

Current Limitations

  • Prototype built on hypothetical datasets, not yet validated clinically.

  • Some advanced features (predictive analytics, patient-facing views) remain outside current scope.

Potential SOlutions

  • Add filtering/customization by patient demographics and conditions.

  • Expand treatment modules with adaptive recommendations.

  • Explore AI-driven fall-risk forecasting.

Check Out My Other Work!

Because one project is never enough 👀

Tag 1

Feature title.

This is a feature description spanning a couple of lines.

Tag 1

Feature title.

This is a feature description spanning a couple of lines.

Tag 1

Feature title.

This is a feature description spanning a couple of lines.

Tag 1

Feature title.

This is a feature description spanning a couple of lines.

Tag 1

Feature title.

This is a feature description spanning a couple of lines.

Tag 1

Feature title.

This is a feature description spanning a couple of lines.

Tag 1

Feature title.

This is a feature description spanning a couple of lines.

Tag 1

Feature title.

This is a feature description spanning a couple of lines.

Tag 1

Feature title.

This is a feature description spanning a couple of lines.