Skip to main content

Advertisement

SpringerLink
Log in

Designing Efficient and Lightweight Deep Learning Models for Healthcare Analysis

  • Published:
Neural Processing Letters Aims and scope Submit manuscript

Abstract

The application of classification and prediction algorithms in healthcare can facilitate the detection of specific vital signs, which can be used to treat or prevent disease. In this study, a new framework based on deep learning architectures is introduced for the human activity recognition. Our proposed framework uses biosensors, electrocardiography (ECG) sensors, inertial sensors, and small single-board computers to collect and analyse sensory information. ECG and inertial sensory information is converted to images, and novel preprocessing techniques are used effectively. We use convolutional neural networks (CNN) along with sensor fusion, random forest, and long short-term memory (LSTM) with gated recurrent unit (GRU). The evaluation of the proposed approaches is carried out by considering well-known models such as transfer learning with MobileNet, CNN+MobileNet combined, and support vector machine in terms of accuracy. Moreover, the effects of the Null class, which are commonly seen in popular health-related datasets, are also investigated. The results show that LSTM with GRU, RandomForest and CNN with sensor fusion provided the highest accuracies with 99%, 98% and 98%, respectively. Since edge computing using sensors with relatively limited processing power and capacities has recently become quite common, a comparison is also provided to show the efficiency and edge computability of the architectures proposed in this study. The number of parameters used, the size of the models, and the training times as well as the testing times are evaluated and compared. The random forest algorithm provides the best training and testing time results, while models of LSTM with GRU have the smallest model size.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14

Similar content being viewed by others

Notes

  1. PAMAP2 - The dataset that contains data from 18 different physical activities and nine subjects, wearing 3 IMUs and an HR-monitor.

References

  1. Pang Z (2013) Technologies and architectures of the internet-of-things (IoT) for health and well-being. PhD thesis, KTH Royal Institute of Technology

  2. Yang G, Xie L, Mäntysalo M, Zhou X, Pang Z, Da Xu L, Kao-Walter S, Chen Q, Zheng L-R (2014) A health-IoT platform based on the integration of intelligent packaging, unobtrusive bio-sensor, and intelligent medicine box. IEEE Trans Industr Inf 10(4):2180–2191

    Article  Google Scholar 

  3. Qadri YA, Nauman A, Zikria YB, Vasilakos AV, Kim SW (2020) The future of healthcare internet of things: a survey of emerging technologies. IEEE Commun Surv Tutor 22(2):1121–1167

    Article  Google Scholar 

  4. Jara AJ, Zamora-Izquierdo MA, Skarmeta AF (2013) Interconnection framework for MHEALTH and remote monitoring based on the internet of things. IEEE J Sel Areas Commun 31(9):47–65

    Article  Google Scholar 

  5. John O, Yadufashije D et al (2018) E-health biosensor platform for non-invasive health monitoring for the elderly in low resource setting. Int J Biomed Eng Sci (IJBES) 5(3/4):11035

    Google Scholar 

  6. Sahoo S, Dash M, Behera S, Sabut S (2020) Machine learning approach to detect cardiac arrhythmias in ECG signals: a survey. IRBM 41(4):185–194

    Article  Google Scholar 

  7. Zhao T, Zhang J, Wang Z, Alturki R (2021) An improved deep learning mechanism for EEG recognition in sports health informatics. Neural Comput Appl 11:1–13

    Google Scholar 

  8. Yildirim Ö (2018) A novel wavelet sequence based on deep bidirectional LSTM network model for ECG signal classification. Comput Biol Med 96:189–202

    Article  Google Scholar 

  9. Rajpurkar P, Hannun AY, Haghpanahi M, Bourn C, Ng AY (2017) Cardiologist-level arrhythmia detection with convolutional neural networks. arXiv preprint arXiv:1707.01836

  10. Salem M, Taheri S, Yuan J-S (2018) ECG arrhythmia classification using transfer learning from 2-dimensional deep CNN features. In: 2018 IEEE Biomedical Circuits and Systems Conference (BioCAS), pp 1–4. IEEE

  11. Oresti Banos AS Rafael Garcia: MHEALTH Dataset Data Set. https://archive.ics.uci.edu/ml/datasets/MHEALTH+Dataset

  12. Gupta N, Gupta SK, Pathak RK, Jain V, Rashidi P, Suri JS (2022) Human activity recognition in artificial intelligence framework: a narrative review. Artif Intell Rev 55(6):4755–4808

    Article  Google Scholar 

  13. Liu R, Ramli AA, Zhang H, Henricson E, Liu X (2022) An overview of human activity recognition using wearable sensors: healthcare and artificial intelligence. In: Internet of things–ICIOT 2021: 6th international conference, held as part of the services conference federation, SCF 2021, Virtual Event, December 10–14, 2021, pp 1–14. Springer

  14. Yin H, Jha NK (2017) A health decision support system for disease diagnosis based on wearable medical sensors and machine learning ensembles. IEEE Trans Multi-Scale Comput Syst 3(4):228–241

    Article  Google Scholar 

  15. Nguyen TT, Fernandez D, Nguyen QT, Bagheri E (2017) Location-aware human activity recognition. In: International conference on advanced data mining and applications, pp 821–835. Springer

  16. Mo L, Li F, Zhu Y, Huang A (2016) Human physical activity recognition based on computer vision with deep learning model. In: 2016 IEEE international instrumentation and measurement technology conference proceedings, pp 1–6. IEEE

  17. Knez T, Machidon O, Pejović V (2021) Self-adaptive approximate mobile deep learning. Electronics 10(23):2958

    Article  Google Scholar 

  18. Sharma M, Garg DK (2022) Human activity detection-based upon CNN with pruning and edge detection. In: Mohanty MN, Das S (eds) Advances in intelligent computing and communication. Springer, Singapore, pp 9–16

    Chapter  Google Scholar 

  19. Isin A, Ozdalili S (2017) Cardiac arrhythmia detection using deep learning. Procedia Comput Sci 120:268–275

    Article  Google Scholar 

  20. Uddin MZ (2019) A wearable sensor-based activity prediction system to facilitate edge computing in smart healthcare system. J Par Distrib Comput 123:46–53

    Article  Google Scholar 

  21. Masum AKM, Hossain ME, Humayra A, Islam S, Barua A, Alam GR (2019) A statistical and deep learning approach for human activity recognition. In: 2019 3rd international conference on trends in electronics and informatics (ICOEI), pp 1332–1337. IEEE

  22. Human activity recognition from wearable sensor data using self-attention. arXiv preprint arXiv:2003.09018 (2020)

  23. O’Halloran J, Curry E (2019) A comparison of deep learning models in human activity recognition and behavioural prediction on the MHEALTH dataset. AICS, Jaffna

    Google Scholar 

  24. Chen K, Yao L, Zhang D, Wang X, Chang X, Nie F (2019) A semisupervised recurrent convolutional attention model for human activity recognition. IEEE Trans Neural Netw Learn Syst 31(5):1747–1756

    Article  Google Scholar 

  25. Li M, Rozgić V, Thatte G, Lee S, Emken A, Annavaram M, Mitra U, Spruijt-Metz D, Narayanan S (2010) Multimodal physical activity recognition by fusing temporal and cepstral information. IEEE Trans Neural Syst Rehabil Eng 18(4):369–380

    Article  Google Scholar 

  26. Zebhi S (2022) Human activity recognition using wearable sensors based on image classification. IEEE Sens J 22(12):12117–12126

    Article  Google Scholar 

  27. Guo H, Chen L, Peng L, Chen G (2016) Wearable sensor based multimodal human activity recognition exploiting the diversity of classifier ensemble. In: Proceedings of the 2016 ACM international joint conference on pervasive and ubiquitous computing, pp 1112–1123

  28. Chen K, Yao L, Wang X, Zhang D, Gu T, Yu Z, Yang Z (2018) Interpretable parallel recurrent neural networks with convolutional attentions for multi-modality activity modeling. In: 2018 International Joint Conference on Neural Networks (IJCNN), pp. 1–8 . IEEE

  29. Sena J, Santos JB, Schwartz WR (2018) Multiscale dcnn ensemble applied to human activity recognition based on wearable sensors. In: 2018 26th European Signal Processing Conference (EUSIPCO), pp. 1202–1206. IEEE

  30. King AC, Whitt-Glover MC, Marquez DX, Buman MP, Napolitano MA, Jakicic J, Fulton JE, Tennant BL et al (2019) Physical activity promotion: highlights from the 2018 physical activity guidelines advisory committee systematic review. Med Sci Sports Exerc 51(6):1340–1353

    Article  Google Scholar 

  31. Piercy KL, Troiano RP, Ballard RM, Carlson SA, Fulton JE, Galuska DA, George SM, Olson RD (2018) The physical activity guidelines for Americans. JAMA 320(19):2020–2028

    Article  Google Scholar 

  32. Li F, Shirahama K, Nisar MA, Köping L, Grzegorzek M (2018) Comparison of feature learning methods for human activity recognition using wearable sensors. Sensors 18(2):679

    Article  Google Scholar 

  33. Nguyen LT, Zeng M, Tague P, Zhang J (2015) Recognizing new activities with limited training data. In: Proceedings of the 2015 ACM international symposium on wearable computers, pp 67–74

  34. Libelium: MySignals HW V2 - eHealth and Medical IoT Development Platform for Arduino. https://www.cooking-hacks.com/mysignals-hw-ehealth-medical-biometric-iot-platform-arduino-tutorial/index.html

  35. Krupaviciute A, Fayn J, McAdams E, Verdier C, Nugent C, Rubel P (2010) Personal sensor-system modalities evaluation for simplified electrocardiogram recording in self-care. In: 2010 Computing in cardiology, pp 541–544. IEEE

  36. Ozdemir MA, Guren O, Cura OK, Akan A, Onan A (2020) Abnormal ECG beat detection based on convolutional neural networks. In: 2020 Medical technologies congress (TIPTEKNO), pp 1–4. https://doi.org/10.1109/TIPTEKNO50054.2020.9299260

  37. Sandler M, Howard A, Zhu M, Zhmoginov A, Chen L-C (2018) Mobilenetv2: Inverted residuals and linear bottlenecks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 4510–4520

Download references

Funding

This work was funded by NU Faculty-development competitive research grants program, Nazarbayev University, Grant Number-110119FD4543. We would like to thank the researchers in Multimedia Lab at NU for their contributions.

Author information

Authors and Affiliations

  1. Department of Computer Science, School of Engineering and Digital Sciences, Nazarbayev University, Nur-Sultan, Kazakhstan

    Mereke Baltabay, Adnan Yazici & Mark Sterling

  2. Computer Engineering, Middle East Technical University, Northern Cyprus Campus, 99738, Mersin 10, Güzelyurt, Turkey

    Enver Ever

Authors
  1. Mereke Baltabay
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Adnan Yazici
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Mark Sterling
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Enver Ever
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Adnan Yazici.

Ethics declarations

Conflict of interest

The authors have no relevant financial or non-financial interests to disclose.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Baltabay, M., Yazici, A., Sterling, M. et al. Designing Efficient and Lightweight Deep Learning Models for Healthcare Analysis. Neural Process Lett 55, 6947–6977 (2023). https://doi.org/10.1007/s11063-023-11246-9

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11063-023-11246-9

Keywords

Search

Navigation