SacHacks V Winner for: - Best Design (no code) - Best Startup Idea

MT4Minds

A mobile application to connect patients seeking music therapy with nearby, volunteer musicians.

🎵

The therapeutic power of music

For the SacHacks V 24-hour hackathon, I worked as the sole UX/UI designer alongside 2 developers. While my teammates were strangers to me before the event, we quickly found common ground in our shared love for music. With research, we learned that music therapy can be incredibly useful for patients recovering from different conditions.

Research shows that music therapy can reduce pain levels, promote relaxation, strengthen communication skills, and provide comfort during hard times. It benefits many populations, including: hospice patients, older adults with memory disorders, people with substance use conditions, individuals on the Autism Spectrum, and many more.

🎵

Barriers to acquiring music therapy

Music therapy is a low-risk intervention that can improve patients’ health with minimal adverse effects and at a lower price than medications. Despite these proven benefits, most insurance plans fail to cover music therapy.

Based on this knowledge, we crafted our problem statement to address these barriers to music therapy.

Due to time constraints, we created user flows that focused on the most important features of our app:

.2. Finding a 2..Patient to 2..Volunteer For

  1. AI-generated Music Therapy Plan

🎵

How might we create a seamless experience for both patients & musicians looking to connect?

  • To make patient profiles more navigable, I considered ways to divide patient information into sections

  • The app highlights the volunteer’s strongest match- a patient who has a music therapy plan that best aligns with the musician’s skill set

  • Weighing out different layouts/button styles for patients completing the wellness survey

  • Considering what wireframe ensures users with various motor and visual disabilities are able to interact with the app

🎵

Better music, better company, better minds: therapy on YOUR terms.

The design for our app includes a survey for the patient to fill out. This will provide some input on their mental health, their music likes/dislikes, and if they are willing to disclose, their health condition(s). Using this information, a machine-learning algorithm will create a personalized music therapy plan with song/genre/instrument recommendations. This app will also be used by volunteers who enjoy playing music- they can view different patients' music therapy plans through their profiles and decide who they want to play for according to their needs/tastes.

Patient Account Setup

  • Different sign-up process for patients and volunteers

  • Submit general information (location, diagnoses) & music-listening preferences

  • Complete a wellness survey to gauge patient’s stress levels & overall wellbeing

  • Based on these answers, patient receives an AI-generated Music Therapy Plan curated to their likes and needs

Find Your Music Buddy!

  • Volunteer musicians can view their most compatible match at the top of the screen

  • Most compatible match based on if your preferences/skills as a musician align with the patient’s needs

  • Easily scan through a patient’s profile to find their therapy plan, preferences & general info

  • Connect with patients you are interested in playing music for

🎵

🎵

What we learned

What’s next?

My team struggled to work with new tools and we set goals that were unattainable in 24 hours. We decided to prioritize the prototype and design for our submission as we were not able to fully build the app before the deadline. Through these roadblocks, we learned how to be adaptable and quickly navigate a shifted landscape for our project.

Next, we would like to bring our vision to life by coding a polished mobile app and implementing IBM Z to carry out the machine learning in our product. In the future, we'd also like to have the volunteers fill out feedback forms about the music they played and the response from the patient, which would then be fed back into the ML model so it could find patterns in the data to enhance the efficiency of the recommendations.