Smart Patient Monitoring System Using Cloud with Deep Learning

Abstract

The rapid advancement of digital healthcare technologies has significantly transformed the accessibility and delivery of medical services. With the widespread adoption of mobile apps, telemedicine, and online portals, patients can now receive healthcare remotely, a shift that became even more crucial during the COVID-19 pandemic. The proposed system builds upon this digital evolution by integrating real-time IoT-based health monitoring with AI-powered disease detection, specifically targeting lung cancer. Utilizing an Arduino Uno board, the system collects data from sensors tracking vital parameters such as heart rate, SpO2 (oxygen saturation), and body temperature. This data is transmitted to a cloud-based platform via a Flask-powered web application, enabling healthcare professionals and caregivers to monitor patients remotely, particularly those with chronic conditions or residing in remote areas. Additionally, the system incorporates deep learning models like ResNet-101 and Convolutional Neural Networks (CNNs) to analyze chest X-ray images for early signs of lung cancer, outperforming traditional manual methods in diagnostic accuracy. The combination of IoT monitoring and AI-enhanced disease detection provides a comprehensive healthcare solution, facilitating continuous patient monitoring, early diagnosis, and preventive care. This integration reduces the burden on healthcare professionals, automates routine diagnostics, and enhances patient access to health data, ultimately leading to more efficient, scalable, and accessible healthcare services

Introduction

1. Overview of Smart Patient Monitoring

Smart Patient Monitoring is a real-time health tracking system that leverages IoT devices, wearable sensors, cloud computing, and AI/ML technologies. It continuously collects patient data like heart rate, SpO?, glucose levels, and temperature, enabling healthcare providers to monitor patients remotely, detect anomalies early, and intervene quickly—improving care efficiency and patient outcomes while reducing hospital visits and healthcare costs.

2. Key Technologies Used

• IoT: Collects data from wearable and embedded sensors.

• Cloud Computing: Stores and shares patient data securely for remote access.

• AI & ML: Enables predictive analytics for early disease detection.

• Deep Learning (DL): Particularly CNNs and ResNet-101 for image-based diagnostics like lung cancer.

3. Literature Survey Insights

• [1] Deep learning models (CNN, RNN) integrated with IoT devices improve real-time anomaly detection.

• [2] The MAR-ERNN model enhances activity recognition using spatial and temporal data for healthcare.

• [3] Edge computing paired with CNNs reduces latency and energy use for real-time monitoring.

• [4] IoT sensors and protocols like MQTT allow for efficient and continuous health data transmission.

• [5] Integrating IoT and AI within smart cities supports proactive and personalized healthcare solutions.

4. Proposed System Description

The system combines IoT-based real-time health monitoring with AI-driven lung cancer diagnosis:

• Hardware: Arduino Uno with sensors for heart rate, SpO?, and body temperature.

• Software:

  • Flask web app for real-time data visualization.

  • AI models (ResNet-101, CNN) for analyzing chest X-rays to detect lung cancer.

  • Sensor data stored via IPFS for integrity and decentralized access.

5. Architecture Highlights

• Data from sensors (gas, temperature) is collected via Arduino Uno.

• Data is cleaned, normalized, and used for training a deep learning model with sigmoid activation.

• The model predicts lung cancer presence using X-ray images and categorizes results as:

  • Valid (cancer detected)

  • Invalid (no cancer)

  • Null (insufficient data)

6. Implementation Steps

• Sensor Integration: Real-time health metrics collected using Arduino and transmitted via serial communication.

• Web Interface: Live health data displayed on Flask-based dashboard.

• Image Dataset: Chest X-ray images (e.g., from Kaggle) used for training AI models.

• Pre-processing: Includes resizing, normalization, and data augmentation.

• Feature Extraction: Uses ResNet-101 to capture disease-related patterns from X-rays.

• Model Training: Trained on labeled image data; tested for generalization.

• Prediction: AI model outputs diagnosis shown in the web app.

7. Results and Performance Evaluation

• Real-time Monitoring: Effective for elderly and chronic patients.

• AI Diagnostics: ResNet-101 and CNN models showed high accuracy in detecting lung cancer.

• Benefits:

  • Faster diagnosis

  • Reduced manual interpretation

  • Enhanced rural/urban healthcare access

8. Performance Metrics

• Accuracy: High classification performance in detecting lung cancer.

• Loss: Cross-entropy loss used; lower loss indicates better model training.

• Precision: Ensures minimal false positives.

• Recall: Maximizes detection of true positive (sick) cases—critical for life-threatening diseases.

9. Arduino & Sensor Details

• Libraries: Wire.h, MAX30100_PulseOximeter.h, DHT.h

• Sensors used: MAX30100 (heart rate & SpO?), DHT11 (temperature)

• System can be simplified to temperature-only monitoring by excluding MAX30100.

Conclusion

In conclusion, the integration of real-time IoT-based health monitoring with AI-powered diagnostic tools marks a transformative advancement in digital healthcare. The proposed system combines an Arduino Uno with vital sensors to continuously measure heart rate, SpO2, and body temperature, providing remote and real-time health data access via a Flask-based web platform. This setup proves particularly beneficial for chronic patients, the elderly, and individuals in remote regions, reducing the necessity of frequent hospital visits while ensuring constant supervision of health parameters. Furthermore, the inclusion of deep learning algorithms such as ResNet-101 significantly enhances the system’s diagnostic accuracy. By analyzing chest X-ray images, the AI models can efficiently detect diseases like with high precision. This not only speeds up the diagnostic process but also minimizes human errors, offering a reliable and consistent method for disease detection. The synergy between IoT monitoring and deep learning-based diagnostics results in a holistic healthcare approach. It supports early intervention, promotes patient engagement, and eases the burden on healthcare providers. Overall, this smart system contributes to a more proactive, scalable, and patient-centric healthcare model, paving the way for accessible and intelligent medical care in both urban and rural settings. The use of advanced AI models, including ensemble deep learning or transformer-based architectures, can improve disease classification accuracy and handle more complex image data. Synchronizing with Electronic Health Records (EHR) ensures seamless patient management and continuity of care. Finally, strengthening data privacy and security with robust encryption and compliance with regulations like HIPAA will safeguard patient confidentiality and ensure data integrity.

References

[1] Abdelaziz, M. Elhoseny, A. S. Salama, and A. M. Riad, ‘‘A machine learning model for improving healthcare services on cloud computing environment,’’ Measurement, vol. 119, pp. 117–128, Apr. 2018. [2] F. Alam, R. Mehmood, I. Katib, and A. Albeshri, ‘‘Analysis of eight data mining algorithms for smarter Internet of Things (IoT),’’ Proc. Comput. Sci., vol. 98, pp. 437–442, Jan. 2016. [3] A. Ara and A. Ara, ‘‘Case study: Integrating IoT, streaming analytics and machine learning to improve intelligent diabetes management system,’’ in Proc. Int. Conf. Energy, Commun., Data Anal. Soft Comput. (ICECDS), Aug. 2017, pp. 3179–3182. [4] L. Atzori, I. A. Iera, and M. Giacomo, ‘‘The Internet of Things: A survey,’’ Comput. Netw., vol. 54, pp. 2787–2805, May 2010. [5] R. Aziz, C. K. Verma, and N. Srivastava, ‘‘Artificial neural network classification of high dimensional data with novel optimization approach of dimension reduction,’’ Ann. Data Sci., vol. 5, no. 4, pp. 615–635, Dec. 2018. [6] G. Bajaj and A. Motwani, ‘‘Improving reliability of mobile social cloud computing using machine learning in content addressable network,’’ in Social Networking and Computational Intelligence (Lecture Notes in Networks and Systems). Singapore: Springer, 2020, pp. 85–103, doi: 10.1007/978-981-15-2071-6_8. [7] C. Cecchinel, M. Jimenez, S. Mosser, and M. Riveill, ‘‘An architecture to support the collection of big data in the Internet of Things,’’ in Proc. IEEE World Congr. Services, Jun. 2014, pp. 442–449. [8] M. Chen, W. Li, Y. Hao, Y. Qian, and I. Humar, ‘‘Edge cognitive computing based smart healthcare system,’’ Future Gener. Comput. Syst., vol. 86, pp. 403–411, Sep. 2018. [9] N. S. Chok, ‘‘Pearson’s versus Spearman’s and Kendall’s correlation coefficients for continuous data,’’ Univ. Pittsburgh, Pittsburgh, PA, USA, Tech. Rep., 2010. [10] N. Tangri, L. Inker, and A. S. Levey, ‘‘A systematic review finds prediction models for chronic kidney disease were poorly reported and often developed using inappropriate methods,’’ J. Clin. Epidemiol., vol. 66, no. 6, p. 697, Jun. 2013. [11] S. K. Das and D. J. Cook, ‘‘Designing smart environments: A paradigm based on learning and prediction,’’ in Proc. Int. Conf. Pattern Recognit. Mach. Intell. Cham, Switzerland: Springer, 2005, pp. 80–90. [12] J. B. Echouffo-Tcheugui and A. P. Kengne, ‘‘Risk models to predict chronic kidney disease and its progression: A systematic review,’’ PLoS Med., vol. 9, no. 11, Nov. 2012, Art. no. e1001344. [13] A. Forkan, I. Khalil, and Z. Tari, ‘‘CoCaMAAL: A cloud-oriented contextaware middleware in ambient assisted living,’’ Future Gener. Comput. Syst., vol. 35, pp. 114–127, Jun. 2014. [14] A. R. M. Forkan, I. Khalil, A. Ibaida, and Z. Tari, ‘‘BDCaM: Big data for context-aware monitoring—A personalized knowledge discovery framework for assisted healthcare,’’ IEEE Trans. Cloud Comput., vol. 5, no. 4, pp. 628–641, Oct. 2017. [15] G. Fortino, R. Giannantonio, R. Gravina, P. Kuryloski, and R. Jafari, ‘‘Enabling effective programming and flexible management of efficient body sensor network applications,’’ IEEE Trans. Hum.-Mach. Syst., vol. 43, no. 1, pp. 115–133, Jan. 2013. [16] V. V. Garbhapu and S. Gopalan, ‘‘IoT based low cost single sensor node remote health monitoring system,’’ Proc. Comput. Sci., vol. 113, pp. 408–415, Jan. 2017. [17] S. González-Valenzuela, X. Liang, H. Cao, M. Chen, and V. C. Leung, Body Area Networks Autonomous Sensor Networks. Berlin, Germany: Springer, 2012, pp. 17–37. [18] P. Gope and T. Hwang, ‘‘BSN-care: A secure IoT-based modern healthcare system using body sensor network,’’ IEEE Sensors J., vol. 16, no. 5, pp. 1368–1376, Mar. 2016. [19] M. Hämäläinen and X. Li, ‘‘Recent advances in body area network technology and applications,’’ Int. J. Wireless Inf. Netw., vol. 24, no. 2, pp. 63–64, Jun. 2017. [20] M. K. Hassan, A. I. El Desouky, S. M. Elghamrawy, and A. M. Sarhan, ‘‘Intelligent hybrid remote patient-monitoring model with cloud-based framework for knowledge discovery,’’ Comput. Electr. Eng., vol. 70, pp. 1034–1048, Aug. 2018. [21] M. S. Hossain and G. Muhammad, ‘‘Cloud-assisted industrial Internet of Things (IIoT)—Enabled framework for health monitoring,’’ Comput. Netw., vol. 101, pp. 192–202, Jun. 2016. [22] Gupta, P., Chouhan, A. V., Wajeed, M. A., Tiwari, S., Bist, A. S., &Puri, S. C. “Prediction of health monitoring with deep learning using edge computing.” Measurement: Sensors, 25, Article 100604, 2023. [23] Philip, J., Gandhimathi, S. K., Chalichalamala, S., Karnam, B., Chandanapalli, S. B., &Chennupalli, S. “Smart health monitoring using deep learning and artificial intelligence.” Revue d\'IntelligenceArtificielle, 37(2), 451-464, 2023. [24] Islam, M.R., Kabir, M.M., Mridha, M.F., Alfarhood, S., Safran, M.; Che, D. “Deep Learning-Based IoT System for Remote Monitoring and Early Detection of Health Issues in Real-Time.” Sensors 2023, 23, 5204. [25] D. Jensen, Beginning Azure IoT Edge Computing: Extending the Cloud to the Intelligent Edge. Berkeley, CA, USA: Apress, 2019, doi: 10.1007/978- 1-4842-4536-1. .

Copyright

Copyright © 2025 Mr. T. Rajan Babu, Dr. R. G. Suresh Kumar, Ms. Hemalatha N, Ms. Janani R, Ms. Ramya C, Ms. Vuddagiri Abhinaya. 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.