Moving Healthcare AI Support Systems for Visually Detectable Diseases to Constrained Devices
Source abstract
Abstract
Image classification usually requires connectivity and access to the cloud, which is often limited in many parts of the world, including hard-to-reach rural areas. Tiny machine learning (tinyML) aims to solve this problem by hosting artificial intelligence (AI) assistants on constrained devices, eliminating connectivity issues by processing data within the device itself, without Internet or cloud access. This study explores the use of tinyML to provide healthcare support with low-spec devices in low-connectivity environments, focusing on the diagnosis of skin diseases and the ethical use of AI assistants in a healthcare setting. To investigate this, images of skin lesions were used to train a model for classifying visually detectable diseases (VDDs). The model weights were then offloaded to a Raspberry Pi with a webcam attached, to be used for the classification of skin lesions without Internet access. It was found that the developed prototype achieved a test accuracy of 78% when trained on the HAM10000 dataset, and a test accuracy of 85% when trained on the ISIC 2020 Challenge dataset.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: COVID-19 diagnosis using AI · Retinal Imaging and Analysis · AI in cancer detection
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Mathematical and Computer Sciences, Heriot-Watt University, Edinburgh EH14 4AS, UK
School of Computing, Engineering & the Built Environment, Edinburgh Napier University, Edinburgh EH11 4BN, UK
Division of Computer Science and Informatics, London South Bank University, London SE1 0AA, UK
Computational BioMedicine Laboratory (CBML), Institute of Computer Science (ICS) Foundation for Research and Technology Hellas (FORTH), GR-700 13 Heraklion, Greece
Funding
Edinburgh Napier University Starter
References (39)
Show source-provided references
1. United Nations (2023, June 20). ITU: 2.9 Billion People Still Offline. Available online: https://www.un.org/en/delegate/itu-29-billion-people-still-offline.
2. Statista (2024, October 23). Countries with the Highest Offline Populations 2024. Available online: https://www.statista.com/statistics/1155552/countries-highest-number-lacking-internet/.
3. (2024, November 14). 25 Countries with Lowest Number of Smartphones per Capita—Insider Monkey. Available online: https://www.insidermonkey.com/blog/25-countries-with-lowest-number-of-smartphones-per-capita-1191140/.
4. Fabila, J., Campello, V.M., Martín-Isla, C., Obungoloch, J., Leo, K., Ronald, A., and Lekadir, K. (2024). Democratizing AI in Africa: FL for Low-Resource Edge Devices. arXiv.
5. Wu, M., Zhou, J., Peng, Y., Wang, S., and Zhang, Y. (2024, January 9–10). Deep Learning for Image Classification: A Review. Proceedings of the 2023 International Conference on Medical Imaging and Computer-Aided Diagnosis, Cambridge, UK.10.1007/978-981-97-1335-6_31 ↗
6. Phase Space Graph Convolutional Network for Chaotic Time Series Learning10.1109/TII.2024.3363089 ↗
7. Self-supervised learning for medical image classification: A systematic review and implementation guidelines10.1038/s41746-023-00811-0 ↗
8. User allocation-aware edge cloud placement in mobile edge computing10.1002/spe.2685 ↗
9. Edge Cloud Offloading Algorithms10.1145/3214306 ↗
10. Convergence of Edge Computing and Deep Learning: A Comprehensive Survey10.1109/COMST.2020.2970550 ↗
11. Towards the Decentralised Cloud10.1145/3243929 ↗
12. (2023, June 20). Fog Computing and the Internet of Things: Extend the Cloud to Where the Things Are What You Will Learn. Available online: https://www.researchgate.net/profile/Mohamed-Mourad-Lafifi/post/Is_there_any_simulation_tool_for_fog_computing/attachment/59d638c079197b8077995f4c/AS%3A398883160117248%401472112564706/download/Fog+Computing+and+the+Internet+of+Things++Extend+the+Cloud+to+Where+the+Things+Are.pdf.
13. A 16-nm SoC for Noise-Robust Speech and NLP Edge AI Inference with Bayesian Sound Source Separation and Attention-Based DNNs10.1109/JSSC.2022.3179303 ↗
14. Goulas, A., Malamas, N., and Symeonidis, A.L. (2022). A Methodology for Enabling NLP Capabilities on Edge and Low-Resource Devices. Natural Language Processing and Information Systems, Proceedings of the International Conference on Applications of Natural Language to Information Systems, Valencia, Spain, 15–17 June 2022, Springer. 13286 LNCS.10.1007/978-3-031-08473-7_18 ↗
15. A Comprehensive Survey on TinyML10.1109/ACCESS.2023.3294111 ↗
16. Miori, L., Sanin, J., and Helmer, S. (2017, January 10–12). A Platform for Edge Computing Based on Raspberry Pi Clusters. Proceedings of the 31st British International Conference on Databases, BICOD, London, UK. 10365 LNCS.10.1007/978-3-319-60795-5_16 ↗
17. Javed, H., Toosi, A.N., and Aslanpour, M.S. (2022). Serverless Platforms on the Edge: A Performance Analysis. New Frontiers in Cloud Computing and Internet of Things, Springer.10.1007/978-3-031-05528-7_6 ↗
18. Keep Your Data Locally: Federated-Learning-Based Data Privacy Preservation in Edge Computing10.1109/MNET.011.2000215 ↗
19. (2024, October 25). Gapminder Tools. Available online: https://www.gapminder.org/tools/?from=world#$ui$chart$trails:false;;&model$markers$bubble$encoding$size$data$space@=geo&=time;;&scale$domain:null&type:null&zoomed:null;;&y$data$concept=cell_phones_per_100_people&source=sg&space@=geo&=time;;&scale$domain:null&zoomed:null&type:null;;&x$data$concept=non_net_users_prc&space@=geo&=time;;&scale$domain:null&zoomed:null&type:null;;&frame$value=2019;;;;;&chart-type=bubbles&url=v2.
20. Janbi, N., Mehmood, R., Katib, I., Albeshri, A., Corchado, J.M., and Yigitcanlar, T. (2022). Imtidad: A Reference Architecture and a Case Study on Developing Distributed AI Services for Skin Disease Diagnosis over Cloud, Fog and Edge. Sensors, 22.10.3390/s22051854 ↗
21. (2023, June 20). Chatbot for Healthcare—IBM Watson Assistant. Available online: https://www.ibm.com/products/watson-assistant/healthcare.
22. Yvanoff-Frenchin, C., Ramos, V., Belabed, T., and Valderrama, C. (2020). Edge Computing Robot Interface for Automatic Elderly Mental Health Care Based on Voice. Electronics, 9.10.3390/electronics9030419 ↗
23. Review of the application of the most current sophisticated image processing methods for the skin cancer diagnostics purposes10.1007/s00403-024-02828-1 ↗
24. Gordon, L.G., Elliott, T.M., Wright, C.Y., Deghaye, N., and Visser, W. (2016). Modelling the healthcare costs of skin cancer in South Africa. BMC Health Serv. Res., 16.10.1186/s12913-016-1364-z ↗
25. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions10.1038/sdata.2018.161 ↗
26. Nugroho, A.A., Slamet, I., and Sugiyanto, S. (2019, January 20). Skins Cancer Identification System of Haml0000 Skin Cancer Dataset Using Convolutional Neural Network. Proceedings of the International Conference on Science and Applied Science (ICSAS), Surakarta, Indonesia.10.1063/1.5141652 ↗
27. Aburaed, N., Panthakkan, A., Al-Saad, M., Amin, S.A., and Mansoor, W. (2020, January 23–25). Deep Convolutional Neural Network (DCNN) for Skin Cancer Classification. Proceedings of the 27th IEEE International Conference on Electronics, Circuits and Systems (ICECS), Glasgow, UK.10.1109/ICECS49266.2020.9294814 ↗
28. Nyatte, S., Perabi, S., Abessolo, G., Ndjakomo Essiane, S., and Ele, P. (2022, January 5–6). Enhancing the Diagnosis of Skin Neglected Tropical Diseases by Artificial Neural Networks Using Evolutionary Algorithms: Implementation on Raspberry Pi. Proceedings of the 3rd International Conference on Electronics, Biomedical Engineering, and Health Informatics, ICEBEHI, Surabaya, Indonesia.10.1007/978-981-99-0248-4_32 ↗
29. Yuan, C., Zhao, D., and Agaian, S.S. (2024). UCM-Net: A lightweight and efficient solution for skin lesion segmentation using MLP and CNN. Biomed. Signal Process. Control, 96.10.1016/j.bspc.2024.106573 ↗
30. Codella, N.C.F., Gutman, D., Celebi, M.E., Helba, B., Marchetti, M.A., Dusza, S.W., Kalloo, A., Liopyris, K., Mishra, N., and Kittler, H. (2018, January 4–7). Skin Lesion Analysis Toward Melanoma Detection. Proceedings of the IEEE 15th International Symposium on Biomedical Imaging (ISBI), Washington, DC, USA.
31. Tai, C.-E.A., Janes, E., Czarnecki, C., and Wong, A. (2024). Double-Condensing Attention Condenser: Leveraging Attention in Deep Learning to Detect Skin Cancer from Skin Lesion Images. Sensors, 24.10.3390/s24227231 ↗
32. Jaworek-Korjakowska, J., Brodzicki, A., Cassidy, B., Kendrick, C., and Yap, M.H. (2021). Interpretability of a Deep Learning Based Approach for the Classification of Skin Lesions into Main Anatomic Body Sites. Cancers, 13.10.3390/cancers13236048 ↗
33. Tan, M., and Le, Q.V. (2019). EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. arXiv.
34. Chatbot for Health Care and Oncology Applications Using Artificial Intelligence and Machine Learning: Systematic Review10.2196/27850 ↗
35. Xu, F., Uszkoreit, H., Du, Y., Fan, W., Zhao, D., and Zhu, J. (2019). Explainable AI: A Brief Survey on History, Research Areas, Approaches and Challenges. Natural Language Processing and Chinese Computing, Proceedings of the 8th cCF International Conference, NLPCC 2019, Dunhuang, China, 9–14 October 2019, Springer. 11839 LNAI.10.1007/978-3-030-32236-6_51 ↗
36. Large language model AI chatbots require approval as medical devices10.1038/s41591-023-02412-6 ↗
37. Conversational agents in healthcare: A systematic review10.1093/jamia/ocy072 ↗
38. A patient-centric dataset of images and metadata for identifying melanomas using clinical context10.1038/s41597-021-00815-z ↗
39. Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., and Chen, L. (2018, January 18–22). MobileNetV2: Inverted Residuals and Linear Bottlenecks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00474 ↗