{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,11]],"date-time":"2026-04-11T22:45:16Z","timestamp":1775947516334,"version":"3.50.1"},"reference-count":150,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Engineering Applications of Artificial Intelligence"],"published-print":{"date-parts":[[2026,6]]},"DOI":"10.1016\/j.engappai.2026.114526","type":"journal-article","created":{"date-parts":[[2026,3,23]],"date-time":"2026-03-23T17:13:49Z","timestamp":1774286029000},"page":"114526","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["Edge-based artificial intelligence: Understanding the evolution of hardware and software and future trends"],"prefix":"10.1016","volume":"174","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7477-1591","authenticated-orcid":false,"given":"Tariq M.","family":"Khan","sequence":"first","affiliation":[]},{"given":"Qazi Emad","family":"Ul Haq","sequence":"additional","affiliation":[]},{"given":"Shahzaib","family":"Iqbal","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8560-0026","authenticated-orcid":false,"given":"Toufique Ahmed","family":"Soomro","sequence":"additional","affiliation":[]}],"member":"78","reference":[{"issue":"6","key":"10.1016\/j.engappai.2026.114526_b1","doi-asserted-by":"crossref","DOI":"10.3390\/mi13060851","article-title":"TinyML: Enabling of inference deep learning models on ultra-low-power IoT edge devices for AI applications","volume":"13","author":"Alajlan","year":"2022","journal-title":"Micromachines"},{"key":"10.1016\/j.engappai.2026.114526_b2","series-title":"2017 International Conference on Engineering and Technology","first-page":"1","article-title":"Understanding of a convolutional neural network","author":"Albawi","year":"2017"},{"issue":"11","key":"10.1016\/j.engappai.2026.114526_b3","doi-asserted-by":"crossref","DOI":"10.3390\/a15110419","article-title":"Recent developments in low-power AI accelerators: A survey","volume":"15","author":"\u00c5leskog","year":"2022","journal-title":"Algorithms"},{"key":"10.1016\/j.engappai.2026.114526_b4","series-title":"Edge artificial intelligence: A systematic review of evolution, taxonomic frameworks, and future horizons","author":"Ali","year":"2025"},{"key":"10.1016\/j.engappai.2026.114526_b5","series-title":"AWS IoT greengrass V2 developer guide","author":"Amazon Web Services","year":"2025"},{"key":"10.1016\/j.engappai.2026.114526_b6","series-title":"AMD Kria{TM} system-on-modules: Accelerating innovation at the edge","author":"AMD","year":"2025"},{"key":"10.1016\/j.engappai.2026.114526_b7","doi-asserted-by":"crossref","first-page":"80543","DOI":"10.1109\/ACCESS.2023.3300658","article-title":"Opportunities, applications, and challenges of edge-AI enabled video analytics in smart cities: A systematic review","volume":"11","author":"Badidi","year":"2023","journal-title":"IEEE Access"},{"key":"10.1016\/j.engappai.2026.114526_b8","series-title":"MicroNets: Neural network architectures for deploying tinyml applications on commodity microcontrollers","author":"Banbury","year":"2021"},{"key":"10.1016\/j.engappai.2026.114526_b9","series-title":"Edge AI: Reshaping the future of edge computing with artificial intelligence","first-page":"133","author":"Banjanovi\u0107-Mehmedovi\u0107","year":"2023"},{"issue":"1","key":"10.1016\/j.engappai.2026.114526_b10","first-page":"1","article-title":"Edge AI: Reshaping the future of edge computing with artificial intelligence","volume":"14","author":"Banjanovi\u0107-Mehmedovi\u0107","year":"2023","journal-title":"Int. J. Adv. Comput. Sci. Appl."},{"key":"10.1016\/j.engappai.2026.114526_b11","series-title":"Proceedings of the 32nd International Conference on Neural Information Processing Systems","first-page":"5151","article-title":"Scalable methods for 8-bit training of neural networks","author":"Banner","year":"2018"},{"key":"10.1016\/j.engappai.2026.114526_b12","series-title":"Artificial intelligence at the edge","author":"Bertino","year":"2020"},{"key":"10.1016\/j.engappai.2026.114526_b13","series-title":"Computer Vision \u2013 ECCV 2020","first-page":"213","article-title":"End-to-end object detection with transformers","author":"Carion","year":"2020"},{"key":"10.1016\/j.engappai.2026.114526_b14","series-title":"2018 5th International Conference on Information, Cybernetics, and Computational Social Systems","first-page":"117","article-title":"Hyperspectral image classification based on convolutional neural network","author":"Chao","year":"2018"},{"key":"10.1016\/j.engappai.2026.114526_b15","series-title":"2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"5260","article-title":"Mobile-former: Bridging MobileNet and transformer","author":"Chen","year":"2022"},{"key":"10.1016\/j.engappai.2026.114526_b16","doi-asserted-by":"crossref","unstructured":"Chen, Y., Dai, X., Liu, M., Chen, D., Yuan, L., Liu, Z., 2020. Dynamic Convolution: Attention Over Convolution Kernels. In: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition. CVPR.","DOI":"10.1109\/CVPR42600.2020.01104"},{"issue":"10","key":"10.1016\/j.engappai.2026.114526_b17","doi-asserted-by":"crossref","first-page":"6232","DOI":"10.1109\/TGRS.2016.2584107","article-title":"Deep feature extraction and classification of hyperspectral images based on convolutional neural networks","volume":"54","author":"Chen","year":"2016","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"issue":"8","key":"10.1016\/j.engappai.2026.114526_b18","doi-asserted-by":"crossref","first-page":"1655","DOI":"10.1109\/JPROC.2019.2921977","article-title":"Deep learning with edge computing: A review","volume":"107","author":"Chen","year":"2019","journal-title":"Proc. IEEE"},{"key":"10.1016\/j.engappai.2026.114526_b19","series-title":"2019 IEEE International Conference on Artificial Intelligence Circuits and Systems","first-page":"167","article-title":"NeuroPilot: A cross-platform framework for edge-AI","author":"Chen","year":"2019"},{"issue":"1","key":"10.1016\/j.engappai.2026.114526_b20","doi-asserted-by":"crossref","first-page":"126","DOI":"10.1109\/MSP.2017.2765695","article-title":"Model compression and acceleration for deep neural networks: The principles, progress, and challenges","volume":"35","author":"Cheng","year":"2018","journal-title":"IEEE Signal Process. Mag."},{"key":"10.1016\/j.engappai.2026.114526_b21","doi-asserted-by":"crossref","unstructured":"Chollet, F., 2017. Xception: Deep Learning With Depthwise Separable Convolutions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. CVPR.","DOI":"10.1109\/CVPR.2017.195"},{"key":"10.1016\/j.engappai.2026.114526_b22","series-title":"Rethinking attention with performers","author":"Choromanski","year":"2022"},{"issue":"1","key":"10.1016\/j.engappai.2026.114526_b23","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1186\/s40537-019-0197-0","article-title":"A survey on image data augmentation for deep learning","volume":"6","author":"Connor Shorten","year":"2019","journal-title":"J. Big Data"},{"key":"10.1016\/j.engappai.2026.114526_b24","series-title":"Proceedings of the 29th International Conference on Neural Information Processing Systems - Volume 2","first-page":"3123","article-title":"BinaryConnect: training deep neural networks with binary weights during propagations","author":"Courbariaux","year":"2015"},{"key":"10.1016\/j.engappai.2026.114526_b25","series-title":"2017 IEEE International Conference on Computer Vision","first-page":"764","article-title":"Deformable convolutional networks","author":"Dai","year":"2017"},{"key":"10.1016\/j.engappai.2026.114526_b26","doi-asserted-by":"crossref","DOI":"10.1016\/j.sysarc.2024.103114","article-title":"Adaptive approximate computing in edge AI and IoT applications: A review","volume":"150","author":"Damsgaard","year":"2024","journal-title":"J. Syst. Archit."},{"key":"10.1016\/j.engappai.2026.114526_b27","series-title":"2019 IEEE\/CVF International Conference on Computer Vision Workshop","first-page":"783","article-title":"Openvino deep learning workbench: Comprehensive analysis and tuning of neural networks inference","author":"Demidovskij","year":"2019"},{"issue":"8","key":"10.1016\/j.engappai.2026.114526_b28","doi-asserted-by":"crossref","first-page":"7457","DOI":"10.1109\/JIOT.2020.2984887","article-title":"Edge intelligence: The confluence of edge computing and artificial intelligence","volume":"7","author":"Deng","year":"2020","journal-title":"IEEE Internet Things J."},{"key":"10.1016\/j.engappai.2026.114526_b29","doi-asserted-by":"crossref","first-page":"7457","DOI":"10.1109\/JIOT.2020.2984887","article-title":"Edge intelligence: The confluence of edge computing and artificial intelligence","volume":"7","author":"Deng","year":"2020","journal-title":"IEEE Internet Things J."},{"key":"10.1016\/j.engappai.2026.114526_b30","series-title":"2021 Third IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications","first-page":"164","article-title":"Trustworthy and sustainable edge AI: A research agenda","author":"Ding","year":"2021"},{"key":"10.1016\/j.engappai.2026.114526_b31","series-title":"Roadmap for edge AI: A dagstuhl perspective","author":"Ding","year":"2021"},{"key":"10.1016\/j.engappai.2026.114526_b32","series-title":"An image is worth 16x16 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2020"},{"key":"10.1016\/j.engappai.2026.114526_b33","series-title":"Theory of Cryptography","first-page":"265","article-title":"Calibrating noise to sensitivity in private data analysis","author":"Dwork","year":"2006"},{"key":"10.1016\/j.engappai.2026.114526_b34","doi-asserted-by":"crossref","DOI":"10.1016\/j.artmed.2024.103012","article-title":"LSSF-net: Lightweight segmentation with self-awareness, spatial attention, and focal modulation","volume":"158","author":"Farooq","year":"2024","journal-title":"Artif. Intell. Med."},{"issue":"4","key":"10.1016\/j.engappai.2026.114526_b35","doi-asserted-by":"crossref","first-page":"419","DOI":"10.1109\/JETCAS.2020.3040300","article-title":"EdgeDRNN: Recurrent neural network accelerator for edge inference","volume":"10","author":"Gao","year":"2020","journal-title":"IEEE J. Emerg. Sel. Top. Circuits Syst."},{"key":"10.1016\/j.engappai.2026.114526_b36","series-title":"MobiZO: Enabling efficient LLM fine-tuning at the edge via inference engines","author":"Gao","year":"2025"},{"key":"10.1016\/j.engappai.2026.114526_b37","series-title":"Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing","first-page":"10786","article-title":"EdgeFormer: A parameter-efficient transformer for on-device seq2seq generation","author":"Ge","year":"2022"},{"key":"10.1016\/j.engappai.2026.114526_b38","series-title":"A survey of quantization methods for efficient neural network inference","author":"Gholami","year":"2021"},{"issue":"18","key":"10.1016\/j.engappai.2026.114526_b39","article-title":"Edge AI: A taxonomy, systematic review and future directions","volume":"28","author":"Gill","year":"2025","journal-title":"Clust. Comput."},{"key":"10.1016\/j.engappai.2026.114526_b40","series-title":"Proceedings of the 13th International Conference on Artificial Intelligence and Statistics","first-page":"249","article-title":"Understanding the difficulty of training deep feedforward neural networks","volume":"vol. 9","author":"Glorot","year":"2010"},{"issue":"6","key":"10.1016\/j.engappai.2026.114526_b41","doi-asserted-by":"crossref","first-page":"1789","DOI":"10.1007\/s11263-021-01453-z","article-title":"Knowledge distillation: A survey","volume":"129","author":"Gou","year":"2021","journal-title":"Int. J. Comput. Vis."},{"key":"10.1016\/j.engappai.2026.114526_b42","series-title":"2021 IEEE\/CVF International Conference on Computer Vision","first-page":"12239","article-title":"Levit: a vision transformer in ConvNet\u2019s clothing for faster inference","author":"Graham","year":"2021"},{"key":"10.1016\/j.engappai.2026.114526_b43","series-title":"Learning Both Weights and Connections for Efficient Neural Networks","first-page":"1135","author":"Han","year":"2015"},{"key":"10.1016\/j.engappai.2026.114526_b44","series-title":"2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"1577","article-title":"GhostNet: More features from cheap operations","author":"Han","year":"2020"},{"issue":"3","key":"10.1016\/j.engappai.2026.114526_b45","doi-asserted-by":"crossref","first-page":"6","DOI":"10.1109\/MWC.001.2000345","article-title":"Edge intelligence for empowering IoT-based healthcare systems","volume":"28","author":"Hayyolalam","year":"2021","journal-title":"IEEE Wirel. Commun."},{"key":"10.1016\/j.engappai.2026.114526_b46","series-title":"2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"4335","article-title":"Filter pruning via geometric median for deep convolutional neural networks acceleration","author":"He","year":"2019"},{"key":"10.1016\/j.engappai.2026.114526_b47","series-title":"2016 IEEE Conference on Computer Vision and Pattern Recognition","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"10.1016\/j.engappai.2026.114526_b48","series-title":"2017 IEEE International Conference on Computer Vision","first-page":"1398","article-title":"Channel pruning for accelerating very deep neural networks","author":"He","year":"2017"},{"key":"10.1016\/j.engappai.2026.114526_b49","series-title":"Edge AI for internet of energy: Challenges and perspectives","author":"Himeur","year":"2023"},{"key":"10.1016\/j.engappai.2026.114526_b50","series-title":"Training Recurrent Neural Networks","author":"Hinton","year":"2013"},{"key":"10.1016\/j.engappai.2026.114526_b51","series-title":"Distilling the knowledge in a neural network","author":"Hinton","year":"2015"},{"key":"10.1016\/j.engappai.2026.114526_b52","series-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"Howard","year":"2017"},{"key":"10.1016\/j.engappai.2026.114526_b53","series-title":"A first look at on-device models in iOS apps","author":"Hu","year":"2023"},{"key":"10.1016\/j.engappai.2026.114526_b54","series-title":"2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"7132","article-title":"Squeeze-and-excitation networks","author":"Hu","year":"2018"},{"key":"10.1016\/j.engappai.2026.114526_b55","series-title":"Proceedings of the 38th International Conference on Machine Learning","first-page":"4466","article-title":"Accurate post training quantization with small calibration sets","volume":"vol. 139","author":"Hubara","year":"2021"},{"key":"10.1016\/j.engappai.2026.114526_b56","series-title":"SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5mb model size","author":"Iandola","year":"2016"},{"key":"10.1016\/j.engappai.2026.114526_b57","doi-asserted-by":"crossref","DOI":"10.1016\/j.iot.2022.100674","article-title":"AI-based fog and edge computing: A systematic review, taxonomy and future directions","volume":"21","author":"Iftikhar","year":"2023","journal-title":"Internet Things"},{"key":"10.1016\/j.engappai.2026.114526_b58","series-title":"Proceedings of the 32nd International Conference on Machine Learning","first-page":"448","article-title":"Batch normalization: Accelerating deep network training by reducing internal covariate shift","author":"Ioffe","year":"2015"},{"key":"10.1016\/j.engappai.2026.114526_b59","article-title":"LDMRes-Net: A lightweight neural network for efficient medical image segmentation on IoT and edge devices","author":"Iqbal","year":"2023","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"10.1016\/j.engappai.2026.114526_b60","series-title":"2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"2704","article-title":"Quantization and training of neural networks for efficient integer-arithmetic-only inference","author":"Jacob","year":"2018"},{"key":"10.1016\/j.engappai.2026.114526_b61","series-title":"Advancing medical image segmentation with mini-net: A lightweight solution tailored for efficient segmentation of medical images","author":"Javed","year":"2024"},{"key":"10.1016\/j.engappai.2026.114526_b62","series-title":"2023 IEEE International Conference on Consumer Electronics","first-page":"1","article-title":"A real-time reconfigurable AI processor based on FPGA","author":"Jeong","year":"2023"},{"key":"10.1016\/j.engappai.2026.114526_b63","series-title":"Proceedings of the Twenty-Second International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS \u201917)","first-page":"615","article-title":"Neurosurgeon: Collaborative intelligence between the cloud and mobile edge","author":"Kang","year":"2017"},{"key":"10.1016\/j.engappai.2026.114526_b64","series-title":"5th International Conference on Learning Representations","article-title":"On large-batch training for deep learning: Generalization gap and sharp minima","author":"Keskar","year":"2017"},{"key":"10.1016\/j.engappai.2026.114526_b65","series-title":"International Conference on Digital Image Computing: Techniques and Applications","first-page":"1","article-title":"MKIS-net: a light-weight multi-kernel network for medical image segmentation","author":"Khan","year":"2022"},{"key":"10.1016\/j.engappai.2026.114526_b66","series-title":"2024 IEEE International Conference on Image Processing","first-page":"2807","article-title":"Lmbf-net: A lightweight multipath bidirectional focal attention network for multifeatures segmentation","author":"Khan","year":"2024"},{"key":"10.1016\/j.engappai.2026.114526_b67","doi-asserted-by":"crossref","DOI":"10.1016\/j.engappai.2024.107995","article-title":"ESDMR-net: A lightweight network with expand-squeeze and dual multiscale residual connections for medical image segmentation","volume":"133","author":"Khan","year":"2024","journal-title":"Eng. Appl. Artif. Intell."},{"key":"10.1016\/j.engappai.2026.114526_b68","series-title":"International Conference on Neural Information Processing","first-page":"612","article-title":"Neural network compression by joint sparsity promotion and redundancy reduction","author":"Khan","year":"2022"},{"key":"10.1016\/j.engappai.2026.114526_b69","doi-asserted-by":"crossref","unstructured":"Khan, T.M., Robles-Kelly, A., Naqvi, S.S., 2022c. T-Net: A resource-constrained tiny convolutional neural network for medical image segmentation. In: Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision. pp. 644\u2013653.","DOI":"10.1109\/WACV51458.2022.00186"},{"key":"10.1016\/j.engappai.2026.114526_b70","series-title":"Adam: A method for stochastic optimization","author":"Kingma","year":"2017"},{"key":"10.1016\/j.engappai.2026.114526_b71","series-title":"Quantizing deep convolutional networks for efficient inference: A whitepaper","author":"Krishnamoorthi","year":"2018"},{"issue":"6","key":"10.1016\/j.engappai.2026.114526_b72","doi-asserted-by":"crossref","first-page":"84","DOI":"10.1145\/3065386","article-title":"ImageNet classification with deep convolutional neural networks","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"Commun. ACM"},{"key":"10.1016\/j.engappai.2026.114526_b73","series-title":"CMSIS-NN: Efficient neural network kernels for arm cortex-m CPUs","author":"Lai","year":"2018"},{"issue":"7553","key":"10.1016\/j.engappai.2026.114526_b74","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"key":"10.1016\/j.engappai.2026.114526_b75","series-title":"On-device neural net inference with mobile GPUs","author":"Lee","year":"2019"},{"issue":"3","key":"10.1016\/j.engappai.2026.114526_b76","doi-asserted-by":"crossref","first-page":"729","DOI":"10.3390\/ai4030039","article-title":"A comprehensive review and a taxonomy of edge machine learning: Requirements, paradigms, and techniques","volume":"4","author":"Li","year":"2023","journal-title":"AI"},{"key":"10.1016\/j.engappai.2026.114526_b77","series-title":"Pruning filters for efficient ConvNets","author":"Li","year":"2017"},{"key":"10.1016\/j.engappai.2026.114526_b78","series-title":"World Conference on Information Systems and Technologies (WorldCIST 2020)","first-page":"35","article-title":"A survey of AI accelerators for edge environment","volume":"vol. 1160","author":"Li","year":"2020"},{"key":"10.1016\/j.engappai.2026.114526_b79","series-title":"Proceedings of the 36th International Conference on Neural Information Processing Systems","article-title":"EfficientFormer: vision transformers at MobileNet speed","author":"Li","year":"2022"},{"key":"10.1016\/j.engappai.2026.114526_b80","series-title":"AI on the edge: Rethinking AI-based IoT applications using specialized edge architectures","author":"Liang","year":"2020"},{"key":"10.1016\/j.engappai.2026.114526_b81","series-title":"AI on the edge: Rethinking AI-based IoT applications using specialized edge architectures","author":"Liang","year":"2020"},{"key":"10.1016\/j.engappai.2026.114526_b82","series-title":"2017 IEEE Conference on Computer Vision and Pattern Recognition","first-page":"936","article-title":"Feature pyramid networks for object detection","author":"Lin","year":"2017"},{"issue":"2","key":"10.1016\/j.engappai.2026.114526_b83","doi-asserted-by":"crossref","first-page":"318","DOI":"10.1109\/TPAMI.2018.2858826","article-title":"Focal loss for dense object detection","volume":"42","author":"Lin","year":"2020","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"10.1016\/j.engappai.2026.114526_b84","series-title":"Advances in Neural Information Processing Systems","article-title":"Runtime neural pruning","volume":"vol. 30","author":"Lin","year":"2017"},{"issue":"7","key":"10.1016\/j.engappai.2026.114526_b85","doi-asserted-by":"crossref","first-page":"1489","DOI":"10.1109\/LCOMM.2022.3167088","article-title":"Privacy-preserving federated edge learning: Modeling and optimization","volume":"26","author":"Liu","year":"2022","journal-title":"IEEE Commun. Lett."},{"key":"10.1016\/j.engappai.2026.114526_b86","doi-asserted-by":"crossref","first-page":"297","DOI":"10.1016\/j.neucom.2021.04.141","article-title":"Bringing AI to edge: From deep learning\u2019s perspective","volume":"485","author":"Liu","year":"2022","journal-title":"Neurocomputing"},{"key":"10.1016\/j.engappai.2026.114526_b87","doi-asserted-by":"crossref","unstructured":"Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., Guo, B., 2021. Swin Transformer: Hierarchical Vision Transformer Using Shifted Windows. In: Proceedings of the IEEE\/CVF International Conference on Computer Vision. ICCV, pp. 10012\u201310022.","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"10.1016\/j.engappai.2026.114526_b88","series-title":"Rethinking the value of network pruning","author":"Liu","year":"2019"},{"key":"10.1016\/j.engappai.2026.114526_b89","series-title":"SGDR: Stochastic gradient descent with warm restarts","author":"Loshchilov","year":"2017"},{"key":"10.1016\/j.engappai.2026.114526_b90","series-title":"The impact of cloud computing and AI on industry dynamics and concentration","author":"Lu","year":"2024"},{"key":"10.1016\/j.engappai.2026.114526_b91","series-title":"Luxonis documentation \u2013 opencv AI kit (OAK)","author":"Luxonis","year":"2025"},{"key":"10.1016\/j.engappai.2026.114526_b92","series-title":"LVS-net: A lightweight vessels segmentation network for retinal image analysis","author":"Mehmood","year":"2024"},{"key":"10.1016\/j.engappai.2026.114526_b93","series-title":"MobileViT: Light-weight, general-purpose, and mobile-friendly vision transformer","author":"Mehta","year":"2022"},{"issue":"4","key":"10.1016\/j.engappai.2026.114526_b94","doi-asserted-by":"crossref","first-page":"621","DOI":"10.1007\/s11227-025-07107-y","article-title":"Deep learning inference optimisation for IoT: Conv2D-ReLU-BN layer fusion and quantisation","volume":"81","author":"Mestre","year":"2025","journal-title":"J. Supercomput."},{"issue":"4","key":"10.1016\/j.engappai.2026.114526_b95","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1109\/MIC.2024.3383758","article-title":"Revisiting edge AI: Opportunities and challenges","volume":"28","author":"Meuser","year":"2024","journal-title":"IEEE Internet Comput."},{"issue":"4","key":"10.1016\/j.engappai.2026.114526_b96","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1109\/MIC.2024.3383758","article-title":"Revisiting edge AI: Opportunities and challenges","volume":"28","author":"Meuser","year":"2024","journal-title":"IEEE Internet Comput."},{"issue":"4","key":"10.1016\/j.engappai.2026.114526_b97","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1109\/MIC.2024.3383758","article-title":"Revisiting edge AI: Opportunities and challenges","volume":"28","author":"Meuser","year":"2024","journal-title":"IEEE Internet Comput."},{"key":"10.1016\/j.engappai.2026.114526_b98","series-title":"Azure IoT edge documentation","author":"Microsoft","year":"2025"},{"key":"10.1016\/j.engappai.2026.114526_b99","series-title":"2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"11256","article-title":"Importance estimation for neural network pruning","author":"Molchanov","year":"2019"},{"key":"10.1016\/j.engappai.2026.114526_b100","first-page":"189","article-title":"Lightweight deep learning models for edge devices\u2014A survey","volume":"17","author":"Musa","year":"2025","journal-title":"Int. J. Comput. Inf. Syst. Ind. Manag. Appl."},{"key":"10.1016\/j.engappai.2026.114526_b101","series-title":"Understanding edge AI","author":"Nutanix","year":"2024"},{"key":"10.1016\/j.engappai.2026.114526_b102","series-title":"Increasing intelligence at the edge with AI","author":"NVIDIA Corporation","year":"2023"},{"key":"10.1016\/j.engappai.2026.114526_b103","series-title":"Onnx\/onnx","author":"ONNX","year":"2025"},{"key":"10.1016\/j.engappai.2026.114526_b104","series-title":"Advances in Neural Information Processing Systems","first-page":"8024","article-title":"PyTorch: An imperative style, high-performance deep learning library","volume":"vol. 32","author":"Paszke","year":"2019"},{"key":"10.1016\/j.engappai.2026.114526_b105","doi-asserted-by":"crossref","first-page":"79","DOI":"10.22331\/q-2018-08-06-79","article-title":"Quantum computing in the NISQ era and beyond","volume":"2","author":"Preskill","year":"2018","journal-title":"Quantum"},{"key":"10.1016\/j.engappai.2026.114526_b106","series-title":"Computer Vision \u2013 ECCV 2016","first-page":"525","article-title":"XNOR-net: ImageNet classification using binary convolutional neural networks","author":"Rastegari","year":"2016"},{"key":"10.1016\/j.engappai.2026.114526_b107","series-title":"2016 IEEE Conference on Computer Vision and Pattern Recognition","first-page":"779","article-title":"You only look once: Unified, real-time object detection","author":"Redmon","year":"2016"},{"key":"10.1016\/j.engappai.2026.114526_b108","series-title":"Advances in Neural Information Processing Systems","article-title":"Faster R-CNN: Towards real-time object detection with region proposal networks","volume":"vol. 28","author":"Ren","year":"2015"},{"key":"10.1016\/j.engappai.2026.114526_b109","series-title":"Data collection and utilization framework for edge AI applications","author":"Rexha","year":"2021"},{"key":"10.1016\/j.engappai.2026.114526_b110","series-title":"Medical Image Computing and Computer-Assisted Intervention \u2013 MICCAI 2015","first-page":"234","article-title":"U-net: Convolutional networks for biomedical image segmentation","author":"Ronneberger","year":"2015"},{"key":"10.1016\/j.engappai.2026.114526_b111","doi-asserted-by":"crossref","first-page":"71","DOI":"10.1016\/j.jpdc.2022.04.004","article-title":"Distributed intelligence on the edge-to-cloud continuum: A systematic literature review","volume":"166","author":"Rosendo","year":"2022","journal-title":"J. Parallel Distrib. Comput."},{"key":"10.1016\/j.engappai.2026.114526_b112","series-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","first-page":"4510","article-title":"MobileNetV2: Inverted residuals and linear bottlenecks","author":"Sandler","year":"2018"},{"key":"10.1016\/j.engappai.2026.114526_b113","series-title":"Attention-gated networks for improving ultrasound scan plane detection","author":"Schlemper","year":"2018"},{"key":"10.1016\/j.engappai.2026.114526_b114","doi-asserted-by":"crossref","DOI":"10.1016\/j.compeleceng.2025.110146","article-title":"Privacy and security vulnerabilities in edge intelligence: An analysis and countermeasures","volume":"123","author":"Shafee","year":"2025","journal-title":"Comput. Electr. Eng."},{"issue":"5","key":"10.1016\/j.engappai.2026.114526_b115","doi-asserted-by":"crossref","first-page":"637","DOI":"10.1109\/JIOT.2016.2579198","article-title":"Edge computing: Vision and challenges","volume":"3","author":"Shi","year":"2016","journal-title":"IEEE Internet Things J."},{"key":"10.1016\/j.engappai.2026.114526_b116","doi-asserted-by":"crossref","DOI":"10.1109\/TMI.2016.2528162","article-title":"Deep convolutional neural networks for computer-aided detection: Cnn architectures, dataset characteristics and transfer learning","author":"Shin","year":"2016","journal-title":"IEEE Trans. Med. Imaging"},{"key":"10.1016\/j.engappai.2026.114526_b117","series-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2015"},{"key":"10.1016\/j.engappai.2026.114526_b118","doi-asserted-by":"crossref","first-page":"71","DOI":"10.1016\/j.iotcps.2023.02.004","article-title":"Edge AI: A survey","volume":"3","author":"Singh","year":"2023","journal-title":"Internet Things Cyber-Physical Syst."},{"key":"10.1016\/j.engappai.2026.114526_b119","series-title":"2022 31st Conference of Open Innovations Association","first-page":"320","article-title":"Artificial intelligence in the IoT era: A review of edge AI hardware and software","author":"Sipola","year":"2022"},{"key":"10.1016\/j.engappai.2026.114526_b120","series-title":"2022 31st Conference of Open Innovations Association","first-page":"320","article-title":"Artificial intelligence in the IoT era: A review of edge AI hardware and software","author":"Sipola","year":"2022"},{"key":"10.1016\/j.engappai.2026.114526_b121","series-title":"Artificial Intelligence for Edge Computing","author":"Springer","year":"2023"},{"issue":"2","key":"10.1016\/j.engappai.2026.114526_b122","first-page":"776","article-title":"The evolution of AI cloud computing and the future it holds","volume":"4","author":"Sriram","year":"2022","journal-title":"Int. Res. J. Mod. Eng. Technol. Sci."},{"issue":"1","key":"10.1016\/j.engappai.2026.114526_b123","first-page":"1929","article-title":"Dropout: a simple way to prevent neural networks from overfitting","volume":"15","author":"Srivastava","year":"2014","journal-title":"J. Mach. Learn. Res."},{"key":"10.1016\/j.engappai.2026.114526_b124","series-title":"Deep learning on edge TPUs","author":"Sun","year":"2022"},{"issue":"3","key":"10.1016\/j.engappai.2026.114526_b125","doi-asserted-by":"crossref","first-page":"1279","DOI":"10.3390\/s23031279","article-title":"A survey on optimization techniques for edge artificial intelligence (AI)","volume":"23","author":"Surianarayanan","year":"2023","journal-title":"Sensors"},{"key":"10.1016\/j.engappai.2026.114526_b126","doi-asserted-by":"crossref","unstructured":"Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., Rabinovich, A., 2015. Going Deeper With Convolutions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. CVPR.","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"10.1016\/j.engappai.2026.114526_b127","series-title":"EfficientNet: Rethinking model scaling for convolutional neural networks","author":"Tan","year":"2019"},{"issue":"7","key":"10.1016\/j.engappai.2026.114526_b128","first-page":"1","article-title":"Efficient transformers: A survey","volume":"54","author":"Tay","year":"2021","journal-title":"ACM Comput. Surv."},{"key":"10.1016\/j.engappai.2026.114526_b129","series-title":"Computer Vision \u2013 ECCV 2022","first-page":"459","article-title":"MaxViT: Multi-axis vision transformer","author":"Tu","year":"2022"},{"key":"10.1016\/j.engappai.2026.114526_b130","series-title":"Linformer: Self-attention with linear complexity","author":"Wang","year":"2020"},{"key":"10.1016\/j.engappai.2026.114526_b131","series-title":"2021 IEEE\/CVF International Conference on Computer Vision","first-page":"548","article-title":"Pyramid vision transformer: A versatile backbone for dense prediction without convolutions","author":"Wang","year":"2021"},{"key":"10.1016\/j.engappai.2026.114526_b132","series-title":"ESRGAN: Enhanced super-resolution generative adversarial networks","author":"Wang","year":"2018"},{"key":"10.1016\/j.engappai.2026.114526_b133","series-title":"Learning structured sparsity in deep neural networks","author":"Wen","year":"2016"},{"key":"10.1016\/j.engappai.2026.114526_b134","series-title":"Fastformer: Additive attention can be all you need","author":"Wu","year":"2021"},{"key":"10.1016\/j.engappai.2026.114526_b135","series-title":"2021 IEEE\/CVF International Conference on Computer Vision","first-page":"22","article-title":"Cvt: Introducing convolutions to vision transformers","author":"Wu","year":"2021"},{"key":"10.1016\/j.engappai.2026.114526_b136","series-title":"Computer Vision \u2013 ECCV 2022","first-page":"68","article-title":"TinyViT: Fast pretraining distillation for small vision transformers","author":"Wu","year":"2022"},{"key":"10.1016\/j.engappai.2026.114526_b137","series-title":"2017 IEEE Conference on Computer Vision and Pattern Recognition","first-page":"5987","article-title":"Aggregated residual transformations for deep neural networks","author":"Xie","year":"2017"},{"key":"10.1016\/j.engappai.2026.114526_b138","series-title":"A survey on deep neural network partition over cloud, edge and end devices","author":"Xu","year":"2023"},{"issue":"3","key":"10.1016\/j.engappai.2026.114526_b139","doi-asserted-by":"crossref","first-page":"93","DOI":"10.1007\/s10462-024-11033-5","article-title":"Edge deep learning in computer vision and medical diagnostics: a comprehensive survey","volume":"58","author":"Xu","year":"2025","journal-title":"Artif. Intell. Rev."},{"issue":"11","key":"10.1016\/j.engappai.2026.114526_b140","doi-asserted-by":"crossref","first-page":"1778","DOI":"10.1109\/JPROC.2021.3119950","article-title":"Edge intelligence: Empowering intelligence to the edge of network","volume":"109","author":"Xu","year":"2021","journal-title":"Proc. IEEE"},{"key":"10.1016\/j.engappai.2026.114526_b141","series-title":"Focal attention for long-range interactions in vision transformers","author":"Yang","year":"2021"},{"key":"10.1016\/j.engappai.2026.114526_b142","series-title":"How Transferable are Features in Deep Neural Networks?","first-page":"3320","author":"Yosinski","year":"2014"},{"key":"10.1016\/j.engappai.2026.114526_b143","series-title":"Proceedings of the Web Conference 2021","first-page":"3111","article-title":"Autodidactic neurosurgeon: Collaborative deep inference for mobile edge intelligence via online learning","author":"Zhang","year":"2021"},{"key":"10.1016\/j.engappai.2026.114526_b144","series-title":"Proceedings of the ACM Web Conference 2022 (WWW \u201922)","first-page":"3298","article-title":"A comprehensive benchmark of deep learning libraries on mobile devices","author":"Zhang","year":"2022"},{"key":"10.1016\/j.engappai.2026.114526_b145","doi-asserted-by":"crossref","first-page":"276","DOI":"10.1016\/j.eng.2024.04.020","article-title":"Physics guided deep learning-based model for short-term origin-destination demand prediction in urban rail transit systems under pandemic","volume":"41","author":"Zhang","year":"2024","journal-title":"Engineering"},{"issue":"8","key":"10.1016\/j.engappai.2026.114526_b146","first-page":"3231","article-title":"Lite transformer network with long\u2013short range attention for real-time fire detection","volume":"59","author":"Zhang","year":"2023","journal-title":"Fire Technol."},{"key":"10.1016\/j.engappai.2026.114526_b147","doi-asserted-by":"crossref","unstructured":"Zhang, X., Zhou, X., Lin, M., Sun, J., 2018. ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. CVPR.","DOI":"10.1109\/CVPR.2018.00716"},{"key":"10.1016\/j.engappai.2026.114526_b148","doi-asserted-by":"crossref","DOI":"10.1155\/2016\/8356294","article-title":"Multiscale CNNs for brain tumor segmentation and diagnosis","volume":"2016","author":"Zhao","year":"2016","journal-title":"Comput. Math. Methods Med."},{"key":"10.1016\/j.engappai.2026.114526_b149","series-title":"ICLR 2020 Workshop on Efficient Methods for Deep Neural Networks (Or Similar Workshop Proceedings)","article-title":"Linear symmetric quantization of neural networks for low-precision integer hardware","author":"Zhao","year":"2020"},{"key":"10.1016\/j.engappai.2026.114526_b150","series-title":"2022 IEEE 24th Int Conf on High Performance Computing & Communications; 8th Int Conf on Data Science & Systems; 20th Int Conf on Smart City; 8th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC\/DSS\/SmartCity\/DependSys)","first-page":"2011","article-title":"Exploring tensorrt to improve real-time inference for deep learning","author":"Zhou","year":"2022"}],"container-title":["Engineering Applications of Artificial Intelligence"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0952197626008079?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0952197626008079?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,4,11]],"date-time":"2026-04-11T22:05:48Z","timestamp":1775945148000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0952197626008079"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6]]},"references-count":150,"alternative-id":["S0952197626008079"],"URL":"https:\/\/doi.org\/10.1016\/j.engappai.2026.114526","relation":{},"ISSN":["0952-1976"],"issn-type":[{"value":"0952-1976","type":"print"}],"subject":[],"published":{"date-parts":[[2026,6]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Edge-based artificial intelligence: Understanding the evolution of hardware and software and future trends","name":"articletitle","label":"Article Title"},{"value":"Engineering Applications of Artificial Intelligence","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.engappai.2026.114526","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}],"article-number":"114526"}}