Colorimetric Analysis of Sweat for Sodium Monitoring with Integrated User Interface

---

Project Summary & Clinical Need

This work presents a novel, strip-based, multimodal biosensing platform for the non-invasive, quantitative detection of sodium (Na⁺) in human sweat. The system is designed for point-of-care usability, bridging laboratory-grade precision with portability to enable the early detection of electrolyte imbalance (Hyponatremia/Hypernatremia) and hyperuricemia risk.

The primary innovation is the Dynamic Environmental Integration—combining chemical colorimetry with real-time temperature and humidity sensing to mitigate the largest source of error: sweat sample evaporation.

Kubelka–Munk Theory

Unlike transparent liquid samples which use the Beer–Lambert Law (absorption), this project utilizes a model for opaque, reflective surfaces (test strips).

• Model: Kubelka–Munk (K–M) Theory

• Principle: The Kubelka–Munk theory relates the measured reflectance (R) of the test strip to the absorption coefficient (K) and the scattering coefficient (S).

• Kubelka–Munk Function:

  F(R) = K/S = (1 − R)² / (2R)

• Relationship to Sodium Concentration:

  Under constant scattering conditions, the Kubelka–Munk function is proportional to the concentration of the absorbing species. Therefore:

  Sodium Concentration ∝ K/S

The ESP32 calculates this K/S value from the measured reflectance to estimate sodium concentration.

Multi-Modal System Architecture & Signal Chain

The platform integrates optical, electronic, and environmental sensing components to create a unified measurement pipeline.

1. Optical Sensing Module (The Core)

Component	Function	Detail
Sensing Platform	Colorimetric Strip & Sweat Sample	Produces a color change proportional to Na⁺ concentration.
LED-Photodiode Module	Light Illumination & Reflection Measurement	Precisely aligned LED/Photodiode geometry measures reflected light intensity.
TIA Front End	Analog Signal Conversion	Converts the low-level electrical current from the Photodiode into a proportional, stable voltage signal.
External ADC	Digitization	Converts the analog voltage signal into a digital (binary) magnitude for the processor.

2. Processing & Communication

Component	Function	Detail
Processing Unit	ESP32 Microcontroller	Handles real-time data acquisition, signal preprocessing, K/S calculation, and classification against physiological ranges.
Wireless Comm.	Wi-Fi/BLE	ESP32 transmits final sodium concentration and environmental data to the Web Dashboard.

3. Environmental Integration (Robustness)

Sensor	Data Logged	Purpose
Temperature/Humidity	Analog equivalent of ambient conditions.	Error Flagging: Logs conditions to explain anomalous high readings (due to evaporation). Correction: Provides data for a future advanced correction model.

---

Proposed Real-World Methodology

The final implementation and user-procedure are designed to mitigate real-world errors:

1. Preparation & Collection: User cleans skin patch and collects sweat via absorbent pads/direct strip contact.
2. Calibration (Two-Point): The device performs Dark Calibration (0% Reflectance) with the LED off (to zero out electronic noise) and White Calibration (100% Reflectance) using a clean strip (to standardize LED brightness).
3. Timed Reaction: The device enforces a precise, pre-programmed incubation time (e.g., 45 seconds) after sample application before taking the reading, controlling for the Evaporation Challenge.
4. Measurement: The appropriate color LED (likely Blue) is turned on, and the photodiode measures the reflected voltage ($V_{\text{sample}}$).
5. Final Calculation: The ESP32 calculates the true reflectance ($R$) using the calibrated voltages and plugs this into the K-M equation to determine the final Sodium Concentration.

Real-World Challenges and Robustness

The design explicitly addresses the most common sources of inaccuracy in portable biosensors:

Error Category	Challenge Addressed	Solution/Design Feature
Sample/User Errors	Evaporation: Rapid water loss concentrates Na⁺.	Timed Reaction programmed into the firmware to ensure consistent incubation.
Optical Errors	Ambient Light Leakage: External light skews readings.	3D-Printed Custom Enclosure designed to be light-tight, preventing external light interference (as shown in 3D-Simulation).
Electronic Errors	LED Brightness Drift: LED intensity changes with temperature/age.	Two-Point Calibration (Dark/White) performed frequently to normalize the photodiode reading ($R$ is a ratio of voltages).
Calibration Errors	Non-Linearity: The K/S vs. Concentration graph curves at extremes.	The system defines a precise "Linear Range" and advises users/systems to only use the linear regression within that range for maximum accuracy.
Chemical Errors	Chemical Interference: Other ions (K⁺, Cl⁻) in sweat.	Rely on the High Selectivity of the sodium ionophore reagent on the test strip.

Web-Based User Interface ("Wireless Dashboard")

The ESP32 hosts a lightweight web server and REST API for comprehensive user feedback.

• Data Visualization: Displays real-time sensor data (Colorimetric AU, Temperature, Humidity) and the final calculated Sodium (Na⁺) concentration.
• Control Panel: Includes controls for Start Chart, Stop Chart, Calibrate, and Clean Cycle.
• System Log: Integrated log panel displays real-time status messages, connection events, and hardware errors from the ESP32, aiding diagnostics.

---

Team Members & Faculty Guide

Role	Name	Roll Number
Team Member	RAGHAVAN	22L255
Team Member	SARRANADHITHIYAA G	22L262
Team Member	SHRIRAM R S	22L268
Team Member	SHYAAMALAN P	22L270
Faculty Guide	Mrs. DEEPIKA J	Assistant Professor, ECE

Affiliation: Electronics and Communication Engineering, PSG College of Technology