Step2Health

Inspiration

We set out to build a clean, unified health platform that consolidates essential wellness tools into a single, seamless workspace. Instead of jumping between disconnected websites to calculate BMIs, track caloric intake, and monitor nutritional habits, Step2Health brings these vital resources together under one intuitive dashboard.

What It Does

Step2Health simplifies everyday health tracking by offering a centralized toolkit:

Core Health Metrics: Calculate BMI, estimate daily caloric targets, and set personalized weight management goals.
Nutrition & Food Tracking: Monitor daily food logs, macro targets, and micro-nutrients in real time.
Camera-Based Pulse Checker: Estimate heart rates dynamically by processing short fingertip video clips.
Mental Wellness Support: Identify high-risk keywords during session check-ins and instantly provide direct access to Canada's 988 Suicide Crisis Helpline.

How We Built It

We engineered Step2Health using a modern Python stack centered around Streamlit:

Frontend & State Management: Streamlit handles interactive routing and state handling, using st.session_state to persist user data across multi-page workflows.
Data Storage: Optimized Python dictionaries serve as lightweight data stores for custom food and vitamin profile lookup.
Computer Vision & Signal Processing: Integrated OpenCV, NumPy, and SciPy to extract frame-by-frame color variations from incoming video files and isolate peak frequency signatures corresponding to cardiac pulses.
Version Control: Managed collaborative feature integration and workflow tracking via Git and GitHub.

Challenges We Ran Into

Feature Integration vs. Simplicity: Balancing broad functional coverage with a minimalist interface required careful UX design to avoid cluttering the central dashboard.
Signal Extraction from Noisy Video: Extracting reliable heart rate estimates meant filtering significant environmental noise out of video recordings. We isolated the red channel and tuned signal processing pipelines to reliably detect subtle blood-flow color shifts.
State Persistence: Mastered Streamlit's rendering lifecycles and state architecture to prevent user entry loss when transitioning between dashboard tabs.

Accomplishments We're Proud Of

Delivered a fully functional, end-to-end health workspace within time-constrained hackathon conditions.
Successfully built a custom computer vision pulse detection pipeline, stepping outside our comfort zone to tackle real-world signal processing.
Preserved a clean, accessible layout while integrating features spanning physical health, computer vision, and crisis intervention.

What We Learned

Hands-on experience architecting modular Streamlit apps with advanced state management.
Fundamentals of digital signal processing and video manipulation using OpenCV, NumPy, and SciPy.
Practical strategies for agile teamwork, scope management, and swift debugging under hackathon deadlines.

What's Next for Step2Health

Expanded Databases: Integrate comprehensive third-party nutrition APIs for scalable food tracking.
Intelligent Insights: Develop personalized health recommendations tailored to individual long-term trends.
Enhanced Computer Vision: Refine PPG filtering algorithms to improve pulse checker accuracy under varying light environments.
Proactive Planning: Introduce automated weekly meal planning and actionable goal-tracking analytics.