The current influenza forecasting models often use traditional influenza surveillance data as one-dimensional digital influenza sequences to predict. This method has limitations in feature extraction and cannot effectively mine deep information in sequences, thus affecting its prediction accuracy. To solve this problem, we introduced an end-to-end integrated model of univariate convolutional neural network based on image coding in this paper. The model encoded one-dimensional digital sequences into two-dimensional image sequences which provided unique image-based information features for the model. On this basis, a deep learning network model was designed and built by integrating swarm intelligence optimization algorithm, and the image processing method of convolutional neural network was utilized for feature extraction to complete the influenza prediction of the influenza data sets from 2013 to 2019 and 2019 to 2023. Compared with the standard model method, the experimental results show that MASE and sMAPE are 1.138, 2.307 and 10.505%, 13.881%, respectively. The model has better performance, good robustness and flexibility in predicting influenza. In addition, the experimental results also verify the advantages and practicability of convolutional neural network image processing in the field of influenza time series forecasting.
ZHONGFade, ZHANGLina, SUNHongyan. Investigation on an outbreak of influenza[J].Zhejiang Journal of Preventive Medicine, 2009, 21(1): 31-32. (in Chinese)
WUJialin. Prevention and control of nosocomial infection of influenza A(H1N1)[J].Journal of Clinical Pulmonary Medicine, 2009, 14(10): 1414. (in Chinese)
CHAIGuorong, WANGBin, SHAYongzhong.Public health risk forecasting with multiple machine learning methods combined: Case study of influenza forecasting in Lanzhou, China[J].Data Analysis and Knowledge Discovery, 2021, 5(1): 90-98. (in Chinese)
[7]
NICHOLK L, NORDINJ D, NELSOND B, et al.Effectiveness of influenza vaccine in the community-dwelling elderly[J].The New England Journal of Medicine, 2007, 357(14): 1373-1381.
[8]
AGORJ K, ÖZALTINO Y.Models for predicting the evolution of influenza to inform vaccine strain selection[J].Human Vaccines & Immunotherapeutics, 2018, 14(3): 678-683.
[9]
BIGGERSTAFFM, JOHANSSONM, ALPERD, et al. Results from the second year of a collaborative effort to forecast influenza seasons in the United States [J]. Epidemics, 2018, 24: 26-33.
[10]
WONM, MARQUES-PITAM, LOUROC, et al.Early and real-time detection of seasonal influenza onset[J].PLoS Computational Biology, 2017, 13(2): e1005330.
[11]
OSTHUSD, HICKMANNK S, CARAGEAP C, et al.Forecasting seasonal influenza with a state-space SIR model[J].The Annals of Applied Statistics, 2017, 11(1): 202-224.
[12]
ZHANGQ, PERRAN, PERROTTAD, et al. Forecasting seasonal influenza fusing digital indicators and a mechanistic disease model[C]// Proceedings of the 26th International Conference on World Wide Web, 2017: 311-319.
[13]
KEELINGM J, DYSONL, TILDESLEYM J, et al.Comparison of the 2021 COVID-19 roadmap projections against public health data in England[J].Nature Communications, 2022, 13(1): 4924.
[14]
SOEBIYANTOR P, KIANGR.Meteorological parameters as predictors for seasonal influenza[J].Geocarto International, 2014, 29(1): 39-47.
[15]
VOLKOVAS, AYTONE, PORTERFIELDK, et al.Forecasting influenza-like illness dynamics for military populations using neural networks and social media[J].PLoS One, 2017, 12(12): e0188941.
[16]
JUNGS, MOONJ, PARKS, et al.Self-attention-based deep learning network for regional influenza forecasting[J].IEEE Journal of Biomedical and Health Informatics, 2022, 26(2): 922-933.
[17]
LEEK, RAY J, SAFTAC.The predictive skill of convolutional neural networks models for disease forecasting[J].PLoS One, 2021, 16(7): e0254319.
[18]
SEMENOGLOUA A, SPILIOTISE, ASSIMAKOPOULOSV.Image-based time series forecasting: A deep convolutional neural network approach[J].Neural Networks, 2023, 157: 39-53.
[19]
WANGZ, OATEST.Encoding time series as images for visual inspection and classification using tiled convolutional neural networks[C]//AAAI-15: Twenty-Ninth Conference on Artificial Intelligence, 2015: 40-46.
WANGZ, OATEST.Imaging time-series to improve classification and imputation[DB/OL].(2015-06-01)[2024-07-25].
[22]
XUEJ, SHENB. A novel swarm intelligence optimization approach:Sparrow search algorithm[J].Systems Science & Control Engineering,2020,8(1):22-34.
[23]
HEK, ZHANGX, RENS, et al.Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR), 2016: 770-778.
[24]
NAIRV, HINTONG E.Rectified linear units improve restricted boltzmann machines[C]// Proceedings of the 27th International Conference on Machine Learning (ICML-10), 2010: 807-814.