Penn Medicine & CHOP Neuroscience Research Group

2026

OCD compulsion tracking wearable for diagnositics

Redesigning for accuracy, independent donning, comfort, and discretion — so patients wear it long enough to collect meaningful data.

Role

Product Designer

Timeline

8 weeks

Team

Yixin Hao, Rives Matson, Arik Singh, Yunbin Wang, Ashni Zaverchand

Platform

Wearable + Web app

Role

Product Designer

Timeline

8 weeks

Team

Yixin Hao, Rives Matson, Arik Singh, Yunbin Wang, Ashni Zaverchand

Platform

Wearable + Web app

Overview

The need to capture invisible data to assist in a proper diagnosis

OCD manifests across recognizable subgroups (checking, contamination, intrusive thoughts, ordering/symmetry), but psychiatric assessments only capture a snapshot, not daily life. Penn Medicine & CHOP's wearable EMG sleeve could continuously monitor physiological signals tied to compulsive behavior — but at 2 people and 15 minutes to put it on, it was failing the one requirement that made the data possible at all: that someone would actually wear it.

The System

Sensors: EMG electrodes, heart rate sensors

Wireless protocol: Bluetooth

Battery life: 60+ hours

Form factor: Forearm compression sleeve

Sleeve Materials: Athletic Moisture-wicking nylon, mesh, magnets, ASA 3D print, Rubber resin print, Lithium battery, microcontroller, PMS silicone

Situation

The difficulties with diagnosing OCD

50.5%

Of OCD in the United States goes misdiagnosed as other conditions

OCD is a very difficult mental condition to diagnose formally. Many OCD patients go undiagnosed, which has left many with improper mental health care that actually worsens their condition. Because OCD compulsions are scenario-dependent, it puts a lot of pressure on the patient to describe their OCD in a 15-minute consultation.

Its crazy because I have known I had OCD but was not properly diagnosed until recently by a professional. It does make me upset that doctors kind of threw medication at me when I was in highschool. With mental health, depression, anxiety you are just playing with brain chemistry which really impacted me for years.

Interview participant B

Individual with OCD

If I thought someone close to me was going to die, I couldn’t relax until I replaced it. As a kid, I remember turning the light on and off 75 times. I felt like I was responsible.

Interview participant A

Individual with OCD

It takes so much time to describe what my OCD looks like. I live it and its a part of me, but when I have only 15 minutes with a doctor I get a lot of anxiety around really describing it and I fear I leave a lot out. But I mean you only have 15 minutes to describe something that you know makes no sense.

Interview participant B

Individual with OCD

Task

In response, our client developed a wearable EMG sleeve.

Researchers needed a way to capture continuous physiological data without patients abandoning the device. Our interdisciplinary team of product managers, engineers, and designers created a comfortable, non-stigmatizing sleeve that patients would wear for full-day data capture while meeting clinical data requirements

My role spanned user research, UX design for the companion app, and workshop design testing both the sleeve's engineering and the app's UI directly with OCD patients, and running sessions throughout the project to validate (or correct) the team's assumptions.

1. Sleeve is in multiple parts, stands out on person, and easily shifts 2. MCU needs to be protected, catches on clothing, and has a fragile mechanical connection 3. App does not help with alignment, no onboarding experience, and little hierarchy


Donning is too long and arduous to fit into someone's routine

With the current design, putting the sleeve on takes 2 people and 15 minutes.


The ecosystem as a whole for the typical day:


Design insights from researchers

Seamless integration needs to be tied to accuracy

The user must be able to replicate how they put on the sleeve and need to stay close to their phones to retain connection, to be adopted applying the sleeve consistently must be easy and quickly to check for adjustment

Prioritze comfort and low profile

A good percentage of people with OCD wear long garments to protect themselves from contamination, this sleeves needs to be able to be won under clothing and no attract attention from others

Capturing invisible data

Not all OCD compulsions manifest themselves in a physical, repeated action.

Action

A device no one wears collects no data.

System: Understanding OCD feedback loop and triggers

Ran scenario-based research sessions with OCD patients to understand what a compulsion cycle actually looks like moment-to-moment, and how patients had worked through one on their own before — grounding the redesign in lived experience rather than assumption.

Designed and personally ran a workshop testing patients' confidence, instruction clarity, and comfort with the language used around donning and connecting the sleeve — the direct source of the Velcro/button usability findings below.


App: User testing UI concepts with OCD patients

Designed and ran a co-design workshop asking OCD patients to design their own version of the app's dashboard, to learn directly what mattered to them and what didn't. It surfaced that percentage-based progress bars caused anxiety (“79% would drive me crazy, I'd keep trying to get to 100%”) and that clinical language like “triggered” needed softening to “experienced a compulsion.”

Alignment system: easy, at-home measuring for applying the sleeve consistently

An anatomy-based two-band alignment system — using the ulna head and humeral epicondyles as reference points — that a clinician sets once in-office, so patients can re-align independently at home without relying on skin marks.


Sleeve: testing one-handed applications while optimizing for placement

A fastening redesign driven directly by that usability testing: patients rated the first prototype's Velcro donning “not intuitive” and button-looping “a bit difficult,” so the buttons were enlarged and the Velcro was replaced with embedded magnets.

Low fidelity sleeve prototype

Connection

Adjustability

Feedback

Dual Velcro straps

High

Straps allowed for tightening around different widths of the arm; Velcro kept snagging on long sleeves

Large, single Velcro plane

Medium

Strap was easy to put on; needs to wrap around the full arm for consistent compression for electrodes on the skin; Velcro also snagged clothing

Mesh/ second material compression

Low

Easy to put on but conflicts with the technology

Peg and knots

Medium

Fast and easy to understand how to attach; the first connection was the hardest to make with one hand

Cord compression

Medium

A single pulling action works well for single-hand use, but another hand was needed to hold the sleeve in place because of the upward pull


Hardware: a seamless, modular design

A flush MCU integration — bending the ribbon cable 180° so the electronics housing sits against the sleeve instead of protruding — plus an interchangeable clinician casing with a built-in battery pocket.

1) Connection research into including ribbon cable connection for more reliable and consistent connection. 2) MCU design that would have the casing slide into the sleeve. 3) Second MCU casing design for casing to hold MCU with small screws or pins

Results

Our redesigned sleeve achieves the following

53% speed improvement

Donning time cut from 15 minutes with 2 people down to 7 minutes with 1 person

9.3/10

Comfort score when worn for 8 hours

8.2/10

ease-of-donning score on the final magnet-fastened version, up from 7.3/10 before the fix

Initial appointment with the clinician

The final design of the sleeve with the attached MCU and the alignment system

At-home sleeve application



App results

Reflection

You must ask the questions, "What does it mean to have this condition?" and, "How does that impact the design?"

Firstly, I want to thank our workshop participants and interviewees who shared their experiences with OCD. Their time and feedback helped shape this project immensely and provided us with the principles we needed to design with for every component of this sleeve. This project emphasized the importance of reaching for primary sources before researching with text or other secondary sources. Also, in reflection, moving forward, this project needs to be tested over several days and will need to be adjusted for people who live different lifestyles or climates.

What is happening with the sleeve today?

The sleeve is now being refined and deployed to patients enrolled in the research trial, and the app — which I led the design of — is being connected to the sleeve to test live readings. The project's lead PhD researcher worked closely with the team through design pitches and sign-off, and that partnership is continuing, with me leading the app's design as we advance.