IoT-Based Wearable System for Early Detection of Parkinson\'s Disease Symptoms Using ESP32, MPU6050, and MAX30102

Abstract

Parkinson\'s disease is a progressive neurological disorder whose earliest motor symptom — a resting hand tremor — is frequently missed during short clinical appointments. This paper presents a low-cost, IoT-based wearable monitoring device designed to continuously detect and log physiological signals that may indicate early neurological abnormality. The system integrates an MPU6050 inertial measurement unit for three-axis motion and tremor analysis, and a MAX30102 optical pulse oximeter for heart rate (BPM) and blood oxygen saturation (SpO2) measurement. An ESP32 microcontroller performs on-device signal processing using a variance-threshold algorithm that identifies repetitive oscillation patterns in the clinically significant 3–6 Hz Parkinson\'s tremor frequency range. Multi-parameter decision logic combines motion and cardiac data before raising an alert, reducing false positives. Sensor data is transmitted wirelessly over Wi-Fi to a cloud dashboard accessible by doctors and caregivers. Local feedback is delivered via an OLED display and audible buzzer. Testing confirmed correct alert behaviour across all operating states, with approximately 98% Wi-Fi packet delivery rate and 8–10 hours of battery life per charge. The device is intended as a screening and monitoring tool — not a clinical diagnostic instrument — and all flagged results must be reviewed by a qualified medical professional.

Introduction

The text discusses a low-cost IoT-based wearable system for early detection and monitoring of Parkinson’s disease (PD), focusing mainly on identifying resting hand tremors — one of the earliest and most important symptoms of PD. Since tremors are often missed during short clinical visits, the proposed solution uses a continuously worn wrist device to monitor patients in real time and support early diagnosis and intervention.

The system is built using affordable hardware components including an ESP32 microcontroller, MPU6050 motion sensor, and MAX30102 heart rate and SpO? sensor. Motion and physiological data are continuously collected and processed using a lightweight variance-threshold algorithm to detect tremor patterns and abnormal cardiac conditions. When abnormal activity is detected, alerts are provided locally through an OLED display and buzzer, while data is transmitted via Wi-Fi to a cloud dashboard for remote access by doctors and caregivers.

The architecture consists of five layers:

1. Sensing Layer – collects motion and cardiac data.
2. Processing Layer – analyzes sensor data using the ESP32.
3. Local Output Layer – provides alerts through display and buzzer.
4. Communication Layer – transmits data to the cloud using Wi-Fi.
5. Remote Access Layer – enables remote monitoring through a cloud dashboard.

The methodology includes:

• Continuous motion monitoring at 100 Hz.
• Calculation of motion intensity from 3-axis accelerometer data.
• Tremor detection using rolling variance analysis over time.
• Heart rate and SpO? monitoring using photoplethysmography (PPG).
• Classification of normal and abnormal physiological conditions based on predefined thresholds.

The literature survey highlights previous research showing that wearable accelerometers, smartphones, and machine learning techniques can effectively detect Parkinson’s tremors. However, many existing systems are expensive or impractical for continuous daily use. The proposed system addresses these limitations by combining motion and cardiac monitoring in a low-cost, cloud-connected wearable suitable for continuous home-based monitoring.

Conclusion

This paper presented a low-cost, IoT-based wearable system for continuous monitoring and early screening of Parkinson\'s disease symptoms. The proposed device integrates an MPU6050 inertial measurement unit and a MAX30102 optical sensor with an ESP32 microcontroller to capture motion, heart rate, and SpO2 data continuously from the patient\'s wrist. A variance-threshold algorithm detects tremor patterns in the 3–6 Hz Parkinson\'s frequency range, and a two-factor classification matrix reduces false positive alerts by combining evidence from both sensor streams. The system provides immediate local feedback through an OLED display and audible buzzer, and wirelessly transmits timestamped data to a cloud dashboard for remote doctor and caregiver access. Testing confirmed correct alert behaviour across all three operating states, with approximately 98% Wi-Fi packet delivery rate, sub-5-second end-to-end latency, and 8–10 hours of battery life per charge — at a total hardware cost of approximately ?2,000–?3,000. The device is intended as a screening and monitoring tool, not a clinical diagnostic instrument. It fills a critical gap by providing objective, continuous physiological data that complements the short observation window of standard clinical appointments. With further development — including FFT-based frequency analysis, machine learning classification, and clinical validation — this platform has strong potential for meaningful real-world impact in neurological health monitoring.

References

[1] S. Patel, K. Lorincz, R. Hughes, N. Huggins, J. Growdon, D. Standaert, M. Akay, J. Dy, M. Welsh, and P. Bonato, \"Monitoring motor fluctuations in patients with Parkinson\'s disease using wearable sensors,\" IEEE Trans. Inf. Technol. Biomed., vol. 13, no. 6, pp. 864–873, Nov. 2009. [2] M. Bächlin, M. Plotnik, D. Roggen, I. Maidan, J. M. Hausdorff, N. Giladi, and G. Tröster, \"Wearable assistant for Parkinson\'s disease patients with the freezing of gait symptom,\" IEEE Trans. Inf. Technol. Biomed., vol. 14, no. 2, pp. 436–446, Mar. 2010. [3] L. Peng, Z. G. Li, and M. Ling, \"Smartphone-based tremor assessment for Parkinson\'s disease using accelerometers,\" in Proc. IEEE Int. Conf. Bioinformatics and Biomedicine (BIBM), Washington, DC, USA, 2015, pp. 673–676. [4] S. Aich, P. M. Pradhan, J. Park, N. Sethi, V. S. S. Vathsa, and H. C. Kim, \"A validation study of freezing of gait detection and machine-learning-based FoG prediction using estimated gait characteristics,\" Sensors, vol. 18, no. 10, p. 3287, 2018. [5] L. Lonini, N. Bhatt, C. Garibotti, L. Dai, J. Shawen, B. Simuni, T. Simuni, and A. Jayaraman, \"Wearable sensors for Parkinson\'s disease: Which data are worth collecting for training symptom detection models,\" NPJ Digit. Med., vol. 1, no. 1, p. 64, 2018. [6] Espressif Systems, \"ESP32 Technical Reference Manual, Version 5.1,\" Espressif Systems, Shanghai, China, 2023. [Online]. Available: https://www.espressif.com/sites/default/files/documentation/esp32_technical_reference_manual_en.pdf [7] InvenSense Inc., \"MPU-6000 and MPU-6050 Product Specification Revision 3.4,\" TDK InvenSense, San Jose, CA, USA, 2013. [8] Maxim Integrated (Analog Devices), \"MAX30102 High-Sensitivity Pulse Oximeter and Heart-Rate Sensor Datasheet,\" Analog Devices Inc., Wilmington, MA, USA, 2018.

Copyright

Copyright © 2026 Amay Shukla, Adarsh Singh, Aditya Singh. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.