A noisy environment adaptive multi-agent medical accompanying system and method for elderly patients

By employing a robust automatic speech recognition module, a multimodal recognition and medical knowledge graph fusion module, and a multi-agent collaboration framework, the system addresses the challenges faced by elderly patients seeking medical care in noisy environments. It achieves efficient and accurate information conversion and end-to-end assistance, thereby enhancing the medical experience for elderly patients.

CN122454967APending Publication Date: 2026-07-24NORTHEASTERN UNIV CHINA
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-04-01
Publication Date
2026-07-24

AI Technical Summary

Technical Problem

Existing medical escort technologies suffer from low voice recognition accuracy, insufficient information conversion capabilities, limited system synergy, and a lack of emotional care in noisy environments, making it difficult to meet the medical needs of elderly patients.

Method used

It employs a robust automatic speech recognition module, a multimodal recognition and medical knowledge graph fusion module, and a multi-agent collaboration framework, combined with cross-attention mechanism, knowledge distillation technology, multi-layer knowledge graph, multi-agent collaboration framework, and emotion recognition and soothing agent to achieve real-time interaction and full-process assistance for elderly patients.

Benefits of technology

It improves the accuracy of speech recognition in noisy environments, enables in-depth analysis and transformation of medical information, provides proactive collaborative services throughout the entire process, and has emotional care, thus enhancing the medical experience for elderly patients.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122454967A_ABST
    Figure CN122454967A_ABST
Patent Text Reader

Abstract

The application relates to the technical field of artificial intelligence, and discloses a noisy environment adaptive multi-agent medical accompanying diagnosis system and method for old patients, which aims to solve the problems of poor adaptability of the medical accompanying diagnosis technology environment, insufficient information conversion capability and limited system collaboration. The method comprises the following steps: a dual-encoder architecture is fused with voice features and waveform language models, cross attention and knowledge distillation are combined to construct a robust speech recognition module; a plurality of heterogeneous medical knowledge graphs are integrated, and a rule and large model hybrid architecture is used to realize deep analysis and popularization of information; a collaboration framework comprising a routing decision maker and a plurality of special intelligent agents is constructed, a memory and retrieval enhancement generation mechanism is integrated to realize whole-process active assistance; and service interaction is performed through a digital human interface. The system comprises corresponding recognition, fusion and collaboration modules. The application can improve the recognition accuracy in a noisy environment, reduce the cognitive load of old patients, and realize whole-process intelligent guidance and old-age care.
Need to check novelty before this filing date? Find Prior Art