200+
Physiotherapists Consulted
8
Hospitals & Clinics Visited
6 Days
Clinical Trials at Nanavati Hospital, Mumbai
4
4 Core Screens Redesign
My Role & Ownership
I owned the UX design layer across the Rehab Buddy companion app — from field research, current-state workflow analysis and UX audit through information architecture, wireframes, design iterations, high-fidelity screens, usability validation and engineering collaboration. The work required designing for a system where the software, physiotherapist, patient and robotic device all interact.
The Product — IMLO & Rehab Buddy
ForHealth is a healthcare technology startup focused on revolutionising physiotherapy and rehabilitation. Their flagship product — the IMLO (Intelligent Multi-Limb Orthosis) — is a robotic assistive device that works alongside physiotherapists to deliver upper limb rehabilitation exercises to patients recovering from stroke, musculoskeletal injury, and neurological conditions.
The companion mobile application bridges the hardware with the clinical workflow — allowing physiotherapists to configure sessions, monitor patient progress in real time, and adjust treatment parameters during live sessions. With the IMLO handling repetitive tasks, therapists can focus their expertise where it matters most: direct patient interaction.
Redesign existing features
Address usability failures from trials and user feedback.
Design New Features
Exercise Library, upper limb flow, session management.
Usability Testing
Internal CA Program + clinical trials.
Engineering Collaboration
Implement designs within the product roadmap.
What I Owned
Discovery & Research
-
Contextual inquiry and structured interviews across 8 hospitals/clinics
-
Current-state workflow and IA analysis
-
Heuristic evaluation of the existing product
Design & Systems
-
Information architecture and user flows
-
Wireframes, design variations and high-fidelity prototypes
-
Reusable UI patterns for future development
Validation
-
Continuous Assessment sessions before clinical exposure
-
Clinical validation at Nanavati Hospital
-
Feedback prioritisation and iteration
Cross-functional Delivery
-
Worked with clinical/domain stakeholders and product/engineering teams
-
Balanced clinical needs, usability and technical constraints
-
Prepared designs for implementation
Why this project is challenging
this is not a conventional mobile-app redesign. A design decision can affect how quickly a physiotherapist configures therapy, how clearly a patient understands instructions, and how safely the device is operated.
Device and Proof of Concept
Problem Statement
From initial to redefined
Initial Problem Statement
What the app was failing at — four system-level failures
Complex Exercise Configuration
Cumbersome and unintuitive — leading to increased setup time and errors.
Disorganised Exercise Library
Lacks organisation and easy navigation — difficult to find and select exercises quickly.
Inefficient Session Planning
Not user-friendly — difficulties creating and modifying session plans efficiently.
Unclear Session Status Monitoring
Cluttered and unclear — hard to track and adjust exercise parameters in real-time.
Redefined Problem Statement
After research — reframed around the clinical user's real need
"Physiotherapists face significant challenges in managing diverse patient needs with limited time and resources. The current lack of intuitive and flexible rehabilitation tools hampers the personalisation and efficiency of upper limb therapy sessions. There is a need to design a comprehensive library of exercises, along with a seamless and user-friendly flow for upper limb exercises, session plans, and session status pages — to enhance the physiotherapist's ability to deliver personalised care efficiently and improve patient outcomes."
The shift: from "what the app fails at" to "what the physiotherapist actually needs." This reframing drove every design decision that followed.
Solution Statement
What the redesign set out to deliver
Comprehensive Exercise Library
Categorised by condition and movement type, with a Favourites tab and inline parameter adjustment. Solving disorganisation and complex configuration in one screen.
Seamless Exercise Flow
Single modular exercise card replacing the four-card system. Includes the Teach and Learn device method for intuitive motion input by the physiotherapists.
Clear Session Status Monitoring
Two-tab session status page: real-time force graphs for the physiotherapist and clear instructions for the patient. No longer cluttered or ambiguous.
Safety-First Device Checks
Two-tab session status page: real-time force graphs for the physiotherapist and clear instructions for the patient. No longer cluttered or ambiguous.
Research — Eight Hospitals, Six Methods
Hospitals & Clinics
Jahangir Hospital
Ruby Hall Clinic
Nobel Hospital
Maeers Physiotherpay
Vishwaraj Hospital
Dr. Pradhan's Clinic
GB Neuro Rehab
Nanavati Hospital
Research Methods
Contextual Enquiry
Close observation and structured interviews across all 8 hospitals — watching physiotherapists work, not just interviewing them.
UX Audit
Heuristic evaluation of all 16 existing screens against Nielsen's 10 Usability Heuristics. Every violation documented with clinical impact.
User Personas
Two personas grounded in field research — Dr. Sneha Gupta and Dr. Priya Sharma — representing the primary and secondary user types.
Competitive Analysis
Benchmarked against global rehabilitation device companies — identifying gaps in VR motivation, EMG feedback, and setup efficiency.
IA Analysis
Mapped existing and proposed information architectures. Revealed that four screens were required for a task that should take one.
Design Variations
4 wireframe variations per key screen — evaluated against heuristics and personas before converging on a final direction.
Key Insights — What Changed My Design Direction
From Finding to Design Decision
EFIICIENCY
Exercise setup carried too much interaction cost
The existing four-card configuration created unnecessary steps. This led to a single modular exercise card with inline editing and direct library access.
RECOGNITION
Experts still needed faster exercise discovery
A flat list forced recall. Categorisation, visual recognition, filters and Favourites were introduced to support expert-user scanning.
VISIBILITY
Session monitoring needed meaningful live feedback
Imprecise dials did not give therapists enough clinical visibility. The redesigned status experience prioritised live force graphs and controls.
CONTROL
The flow needed to support experienced clinicians
Linear navigation created friction. The proposed IA and shortcuts were designed to reduce unnecessary movement through the system.
SAFETY
Safety had to be a first-class interaction
Device position verification, error handling and a fail-safe stop became explicit parts of the flow rather than hidden system behaviour.
PATIENT
The patient experience mattered too
The app also had to communicate clearly to neurological patients. Instructions and session-end feedback were treated as usability and motivation problems.
User Personas - Designing for clinicians
Dr. Neha Gupta - 40
Physiotherapist
Goals
Personalised rehab plans
Reduce manual workload
Real Time patient tracking
Pain Points
Personalised rehab plans
Reduce manual workload
Real Time patient tracking
Needs
Real-time performance data
Easy session plan creation
Exercise adherence monitoring
Tech Comfort
Proficient with EHR tools
Open to new rehab tech
Dr. Priya Sharma - 35
Physiotherapist
Goals
Personalised rehab plans
Reduce manual workload
Real Time patient tracking
Pain Points
Personalised rehab plans
Reduce manual workload
Real Time patient tracking
Needs
Real-time performance data
Flexible treatment plan software
Adherence monitoring in workflow
Tech Comfort
Open to new rehab devices
Seeks efficiency - enhancing tools
Key Insights
In a clinical product, generic user archetypes are insufficient — the user is a trained medical professional with domain knowledge and patient accountability
UX Audit — Heuristic Evaluation
Key Heuristic Violations Found
01 - 4 cards to configure 1 exercise
Violates Flexibility & Efficiency — a task that should take 30 seconds was taking 3–4 minutes. Multiplied across a full day, this lost hours of clinical time weekly.
02 - No real-time session feedback
Violates Visibility of System Status — no live data during a session. Imprecise dials left physiotherapists without the clinical intelligence they needed mid-session.
03 - Recognition forced into recall
Violates Recognition Over Recall — flat, unfiltered exercise list required physiotherapists to remember exercise names rather than recognise them visually.
04- Linear navigation with no shortcuts
Violates User Control and Freedom — no way to jump between sections. Every task required navigating from the start, regardless of experience level.
05 - Inconsistent UI patterns
Violates Consistency and Standards — buttons, icons, and terminology changed meaning across screens, breaking the expectation of a stable, learnable system.
06 - No patient motivation at session end
No positive reinforcement on session completion — a clinical miss, since patient motivation and compliance directly affect rehabilitation outcomes.
Existing App Screens (Annotated)
Drag and Drop Exercise Configuration
Drag and Drop Exercise Configuration
Drag and Drop Exercise Configuration
Output
Each violation documented with screen, heuristic, and clinical impact
Design Iterations
Process
4 variations each → heuristic evaluation → team feedback → selected direction
Physio Dashboard
Variation 1
Simple vertical list. Fast to scan but lacked device status and clinical overview.
Variation 2
Card-based with device status prominent. Better hierarchy but underused screen space.
Variation 3
Device position preview + session history. Richer but cognitively heavy for a home screen.
Variation 4 - Selected
Balanced dashboard: patient list + device status + improvement score at a glance. Full history one level deeper.
Exercise Configuration
Variation 1
Step-by-step wizard. Clear but too many taps — did not solve the 4-card problem.
Variation 2
Drag-and-drop layout. Efficient but learning curve inappropriate for clinical use.
Variation 3
Three-section layout. Space-efficient but visually complex at a glance.
Variation 4 - Selected
Single modular card with inline editing, copy function, and direct library access. Reduced configuration to one interaction point.
Exercise Library
Variation 1
Simple list with intensity levels. Fast to scan but no visual exercise recognition.
Variation 2
Image-based library with search and filter. More intuitive but loading speed was a concern.
Variation 3
High-density grid. Compact but overwhelming for a large exercise set.
Variation 4 - Selected
Categorised view + filter panel + Favourites tab + custom exercise creation. Balanced discovery with expert-user efficiency.
Final Designs — All Screens
Exercise Library, Session Plan, Device Safety Checks, Session Status
Core App Screens
Redesigned Core Screens
Usability Testing
Two testing methods — an internal simulation programme and formal clinical trials at Nanavati Hospital.
Clinical Trials
Physiotherapists & Neurological Patients
6-day formal trial at Nanavati Hospital with physiotherapists, an MSK specialist, and neurological patients.
MSK specialist tested device reliability and clinical applicability
Senior physiotherapist used it as both clinician and patient
Neurological patients used the app independently to test clarity
Qualitative + quantitative data collected from clinicians
Continuous Assessment
Internal Session Simulation
Structured internal programme — team members role-played physiotherapist and patient to test the full app and device workflow before clinical exposure.
Scenario-based testing of common physiotherapy use cases
Error handling — incorrect inputs and unexpected situations tested
All team members involved for diverse coverage
Feedback prioritised by severity, fed back into design cycle
Realistic conditions mirroring actual clinic environments
Also — Internal Software Testing
Engineering ran unit, integration, system, and regression tests in parallel to ensure no regressions were introduced by design changes.
Strengths, gaps, opportunities.
What the Redesign Delivered
01 - Session setup dramatically faster
Four-card system reduced to one modular card. Verified through direct clinical observation at Nanavati.
02 - Real-time clinical intelligence
Session status page transformed from imprecise dials to live force graphs and in-session controls.
03 - Patient instruction clarity validated
Neurological patients used the app independently, the most demanding usability benchmark in the project.
04- Clinical trust through safety
Redesigned safety check screens increased physiotherapist confidence in device setup — critical for adoption.
05 - Scalable design system
Component library now serves as the foundation for all future app development.
Strengths
Comprehensive, extensible exercise library
Highly customisable session plans
Real-time patient performance tracking
Safety-first device architecture
Weakness
Requires onboarding for new users
Dependent on device connectivity
Resource constraints limited v1 scope
Opportunities
Broader health monitoring integration
Remote physiotherapy & telehealth
AI-powered treatment personalisation
Threats
Pressure from established global competitors
Device malfunctions eroding clinical trust
Regulatory and data privacy requirements
What This Project Demonstrates
Why this case study matters beyond the screens
01 — Complex workflows
Designing for a multi-actor system
The experience connects a physiotherapist, patient, mobile interface and robotic rehabilitation device. I had to reason about the full workflow, not isolated screens.
02 — Evidence-led decisions
Research changed the problem
The work moved from fixing visible app failures to understanding the clinical user's real needs — then translating those findings into IA and interaction decisions.
03 — Product thinking
Balancing usability with feasibility
Design decisions considered clinical usefulness, learnability, device behaviour and implementation constraints rather than optimising only for visual polish.
04 — Validation
Testing in the environment that matters
The work moved from fixing visible app failures to understanding the clinical user's real needs — then translating those findings into IA and interaction decisions.
My takeaway
This project strengthened my ability to take an ambiguous, domain-heavy problem, structure the research, make explicit design trade-offs, validate them with users, and translate the outcome into a system the product and engineering teams can build.