top of page

Healthbuddy

Healthbuddy is a conversational AI-based solution that caters to help users find accurate, simplified and verified information to any medical queries.

Type

Academic Project

My Role

UX/UI Designer responsible for Primary Research, Ideation, Wireframing, Visual Design

Duration

10 weeks

UX Methods

Primary research, Secondary research, field study, user persona, storyboarding, user flows, affinity mapping, wireframing, prototyping, brainstorming, visual design, and usability testing.

Collaboration

Collaborated with 3 other HCI students for this project.

Tools

Figma, Figjam, Miro

Before moving forward here is a quick overview of the project problem, target users, and the solution.

Problem

People are constantly overwhelmed with the overload, complexity, and unreliability of medical information and require a reliable source to help address this concern.

Users

For this project, we targeted two types of users, one who has experienced the healthcare system recently and the other who does not have experience dealing with the healthcare environment.

Solution

A conversational AI-based solution that caters to help users find accurate, simplified and verified information to any medical queries.

Research

Literature Review

Epathise

I went through several research papers on online resources like google Scholarly to get perspective on sub-fields in health literacy, current health literacy status, and challenges faced in this domain. Insights gathered to set a base for user interviews and observations.

User Interview

I conducted 2 semi-structured user interviews, one user who had recently experienced American healthcare and other who had no experience with the system. I observed 6 others led by my teammates. Interviews were conducted in person and on zoom. Insights were useful to map and group user problems.

Online Observation

As access to the actual field was out of scope, I conducted online observations where I visited online social platforms like, Quora, Reddit, Twitter, and medical forums, to understand the kind of doubts users have.

Literature Review

Group 18.png

eHealth literacy is defined as the ability to seek, find, understand, and appraise health information from electronic sources and apply the knowledge gained to addressing or solving a health problem.

Health literacy is the degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions.

Insights

From user interviews and online observation

Group 20.png

A treemap mentioning the frequency of the issue helped us understand the severity of the problem by size.

Define

User  persona

Group 22.png
Group 25.png

Problem Statement

People are constantly overwhelmed with the overload, complexity, and unreliability of medical information and require a reliable source to help address this concern.

Area of Focus

Overload of Information

Lack of Reliability

Complexity of Terminologies

Ideate

Ideation

For this project, we choose health literacy and went forward with brainstorming in the same direction. Each team member individually came up with around 15 ideas which we further discussed and narrowed down to three. Later after a details analysis, we finalized the concept of Healthbuddy.

As a team, we used tools like Miro and Figjam for whiteboarding and used methods like storyboarding to detail the ideas.

Moving ahead with healthbuddy, we eliminated insurance related scenarios and decided to focus on following scenarios:

Idea Refinement

1. General Medical Queries
2. Data verification

Solution

Healthbuddy is the application that hosts Max. It allows users to get clarification on doubts and concerns while dealing with health information.

Solution

A conversational AI-based solution that caters to help users find accurate, simplified, and verified information for any medical queries.

Introducing Max

Group 26.png
Prototype

Information
Architecture

User Flow

Prototype

During prototyping, we took Monica's persona into account and created two scenarios that covered key features.
 

1. General medical information: Monica intends to go on a hike but is concerned that pollen will aggravate her asthma.
2. Information Verification: In the same scenario, she wants to verify some information about the effects of exertion on asthma patients.

I used Figma to create a mid-fidelity prototype and for UI design.

General Medical

information

To provide personalized input, the system will primarily gather information from the patient portal. If access to the patient portal is denied, the bot will ask the necessary questions to provide precise input.

Information Verification 

On the back end, the system will compare the provided information to verified sources such as WHO to determine its authenticity, gather the necessary data, simplify it, and display it in an understandable language.

Patient Portal

Healthbuddy requires access to the user's patient portal in order to provide a personalized experience. The user has the option to deny access to the portal. In this case, the bot will ask the appropriate questions before responding.

My Topics

The app will display a curated list of articles based on the user's interaction with the bot. This will be a short snippet of information that is easy to understand, similar to flashcards. Users can share and save these cards for later use.

Test

Testing

Following the Minimum Viable Product, we decided to conduct some usability testing. We performed heuristic evaluations based on the 10 Nielsen Norman heuristics. We also conducted a think-aloud session with actual users on two task flows covered earlier.

What Worked?

  • Users like the concept of healthbuddy and a conversational approach to doubt-solving.

  • Users gave positive feedback on UI and found it easy to use.

  • Users were delighted to get simplified answers in an easy-to-understand language.


What did not work?

  • A lack of system visibility status was observed during Heuristic evaluation where the user was not informed about the number of steps required to complete the given task.

  • The application did not provide a system walkthrough and FAQ sections.

  • The users felt the need to edit previous chats with the bot, which the application did not provide.

My learning

I learned how to break down large issues into numerous smaller ones and address each one separately in order to address the more large problem.

I acquired skills for working together and resolving design disagreements without obstructing design workflow.

Thank you for visiting!

Let's connect!

  • LinkedIn
  • Instagram
bottom of page