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

MedFusion-LLM: A Blockchain Enabled Weighted Token-Attribute Mapping Framework for Intelligent Medical Documents Retrieval

Syeda Ayesha Siddiqha, Chandra Sekhar M

International Journal of Automation and Smart Technology · 25.07.2026 · 16 (1)

DOI: 10.5875/x4r24w25 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Automation and Smart Technology, online erschienen am 25.07.2026. Verfasst von Syeda Ayesha Siddiqha und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Medical documents retrieval and processing is a challenging task as it contains sensitive and classified information on the patient and treatment history. With modernizing technology and purpose of data processing has become the need of the hour. The urgency in large medical data processing across geographical servers can yield enormous meaningful information to upscale the regular medical practices. The proposed system is developed on the existing third-party Large Language Models (LLM) to extract and systematically categorize the medical records. The input driven from the user (doctor) is collected and token segmentation is appended with a novel interdependency weight token-attribute-weight (WTW) mapping for efficient customization of requesting queries. The outcome has demonstrated a higher accuracy in extracting and recommending documents. The blockchain based document inter-connectivity assures the privacy and integrity of the data shared via the proposed framework. The outcome has demonstrated an accuracy of 92.2% in document hit-rate and return-rate with respect to the medical document classification and categorization. The framework incorporates a Weighted Token-Attribute-Weight (WTW) mapping mechanism to enhance semantic alignment and contextual relevance during medical document retrieval and classification. The model is evaluated using multiple performance metrics, including precision, recall, and F1-score, along with accuracy, ensuring a comprehensive assessment of retrieval effectiveness.

Zitieren

APA

Siddiqha, S. A., & M, C. S. (2026). MedFusion-LLM: A Blockchain Enabled Weighted Token-Attribute Mapping Framework for Intelligent Medical Documents Retrieval. International Journal of Automation and Smart Technology, 16(1). https://doi.org/10.5875/x4r24w25

Vancouver

Siddiqha SA, M CS. MedFusion-LLM: A Blockchain Enabled Weighted Token-Attribute Mapping Framework for Intelligent Medical Documents Retrieval. International Journal of Automation and Smart Technology. 2026;16(1). doi:10.5875/x4r24w25

BibTeX
@article{Siddiqha2026,
  author = {Siddiqha, Syeda Ayesha and M, Chandra Sekhar},
  title = {MedFusion-LLM: A Blockchain Enabled Weighted Token-Attribute Mapping Framework for Intelligent Medical Documents Retrieval},
  journal = {International Journal of Automation and Smart Technology},
  year = {2026},
  volume = {16},
  number = {1},
  doi = {10.5875/x4r24w25}
}
Kostenlos starten →