EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

YOLOv8 integrates LSTM intelligent monitoring and early warning research on the resilience of the traditional Chinese medicine industry chain

Meng Lv, Ke Chen, Lina He

Discover Artificial Intelligence · 16.05.2026 · 6 (1)

DOI: 10.1007/s44163-026-01292-2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Discover Artificial Intelligence, online erschienen am 16.05.2026. Verfasst von Meng Lv und 2 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: he Henan University of Animal Husbandry and Economy, Henan Provincial Higher Education Institutions Young Backbone Teacher Training Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper utilizes the fusion model of YOLOv8 and LSTM to construct an intelligent monitoring and early warning system for the elasticity of the traditional Chinese medicine industry chain. This system combines all the efficient target detection capabilities of the YOLOv8 algorithm with the advantages of the LSTM neural network in processing time series data, and proposes a brand-new intelligent monitoring method. This system can conduct real-time monitoring of key links in the traditional Chinese medicine industry chain. It can also provide accurate risk warnings and predictions by analyzing time series data. This can enhance the stability and flexibility of the traditional Chinese medicine industry chain. In the process of data analysis, the experiment utilized multimodal datasets from various links of the traditional Chinese medicine industry chain. The visual data is processed by the YOLOv8 model, while the digital time series data is processed by the LSTM network. Moreover, this system was compared with traditional models such as YOLOv4 and YOLOv5. The experimental results showed that the accuracy rate of the YOLOv8 model reached 90%, the recall rate was 88%, and the F1 score was 89%. In terms of accuracy rate, recall rate, and F1 score, it is significantly better than other versions. When YOLOv8 and LSTM are integrated, the performance of the YOLOv8-LSTM model becomes even more prominent. Its accuracy rate can reach 92%, the recall rate can reach 91%, and the F1 score can also reach 91%. These data demonstrate that The YOLOv8-LSTM fusion model has significant advantages in dynamically monitoring risk factors in the traditional Chinese medicine industry chain. The YOLOv8-LSTM model shows high accuracy when predicting the resilience of the traditional Chinese medicine industry chain, especially in long-term prediction, the model can maintain a prediction accuracy of up to 75%, while the accuracy of YOLOv8 and YOLOv4-LSTM is significantly reduced. Through these experimental data, this paper proves that the integration of YOLOv8 and LSTM can greatly improve the intelligent monitoring and early warning capabilities of the traditional Chinese medicine industry chain, and provide strong decision support for governments and enterprises.

Förderung

  • he Henan University of Animal Husbandry and Economy
  • Henan Provincial Higher Education Institutions Young Backbone Teacher Training Program

Zitieren

APA

Lv, M., Chen, K., & He, L. (2026). YOLOv8 integrates LSTM intelligent monitoring and early warning research on the resilience of the traditional Chinese medicine industry chain. Discover Artificial Intelligence, 6(1). https://doi.org/10.1007/s44163-026-01292-2

Vancouver

Lv M, Chen K, He L. YOLOv8 integrates LSTM intelligent monitoring and early warning research on the resilience of the traditional Chinese medicine industry chain. Discover Artificial Intelligence. 2026;6(1). doi:10.1007/s44163-026-01292-2

BibTeX
@article{Lv2026,
  author = {Lv, Meng and Chen, Ke and He, Lina},
  title = {YOLOv8 integrates LSTM intelligent monitoring and early warning research on the resilience of the traditional Chinese medicine industry chain},
  journal = {Discover Artificial Intelligence},
  year = {2026},
  volume = {6},
  number = {1},
  doi = {10.1007/s44163-026-01292-2}
}
Kostenlos starten →