Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Large Language Model-Enhanced Interactive Agent for Public Education on Newborn Auricular Deformities

Shuyue Wang, Liujie Ren, Tianyao Zhou, Lili Chen, Tianyu Zhang, Yaoyao Fu, Shuo Wang

TL;DR

The paper addresses the need for early recognition and education regarding newborn auricular deformities, which can affect aesthetic and psychosocial well-being and, in some cases, hearing. It proposes a Baidu Ernie-based interactive agent (Ernie-fdear) that combines a PaddleDetection-based image-diagnosis module and a Retrieval-Augmented Generation (RAG) knowledge module to provide professional guidance and reduce barriers to initial assessment. Key findings include a 75% category accuracy for deformity typing and 90% accuracy for normal-vs-abnormal classification in image-based diagnosis, along with a user study showing near-expert educational performance when using the RAG-enabled agent. The approach demonstrates potential to extend remote, equity-focused pediatric diagnosis and education to other newborn illnesses, particularly in rural regions where access to specialists is limited.

Abstract

Auricular deformities are quite common in newborns with potential long-term negative effects of mental and even hearing problems.Early diagnosis and subsequent treatment are critical for the illness; yet they are missing most of the time due to lack of knowledge among parents. With the help of large language model of Ernie of Baidu Inc., we derive a realization of interactive agent. Firstly, it is intelligent enough to detect which type of auricular deformity corresponding to uploaded images, which is accomplished by PaddleDetection, with precision rate 75\%. Secondly, in terms of popularizing the knowledge of auricular deformities, the agent can give professional suggestions of the illness to parents. The above two effects are evaluated via tests on volunteers with control groups in the paper. The agent can reach parents with newborns as well as their pediatrician remotely via Internet in vast, rural areas with quality medical diagnosis capabilities and professional query-answering functions, which is good news for newborn auricular deformity and other illness that requires early intervention for better treatment.

Large Language Model-Enhanced Interactive Agent for Public Education on Newborn Auricular Deformities

TL;DR

The paper addresses the need for early recognition and education regarding newborn auricular deformities, which can affect aesthetic and psychosocial well-being and, in some cases, hearing. It proposes a Baidu Ernie-based interactive agent (Ernie-fdear) that combines a PaddleDetection-based image-diagnosis module and a Retrieval-Augmented Generation (RAG) knowledge module to provide professional guidance and reduce barriers to initial assessment. Key findings include a 75% category accuracy for deformity typing and 90% accuracy for normal-vs-abnormal classification in image-based diagnosis, along with a user study showing near-expert educational performance when using the RAG-enabled agent. The approach demonstrates potential to extend remote, equity-focused pediatric diagnosis and education to other newborn illnesses, particularly in rural regions where access to specialists is limited.

Abstract

Auricular deformities are quite common in newborns with potential long-term negative effects of mental and even hearing problems.Early diagnosis and subsequent treatment are critical for the illness; yet they are missing most of the time due to lack of knowledge among parents. With the help of large language model of Ernie of Baidu Inc., we derive a realization of interactive agent. Firstly, it is intelligent enough to detect which type of auricular deformity corresponding to uploaded images, which is accomplished by PaddleDetection, with precision rate 75\%. Secondly, in terms of popularizing the knowledge of auricular deformities, the agent can give professional suggestions of the illness to parents. The above two effects are evaluated via tests on volunteers with control groups in the paper. The agent can reach parents with newborns as well as their pediatrician remotely via Internet in vast, rural areas with quality medical diagnosis capabilities and professional query-answering functions, which is good news for newborn auricular deformity and other illness that requires early intervention for better treatment.
Paper Structure (7 sections, 2 figures, 3 tables)

This paper contains 7 sections, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Experiments and Results
  4. Expert diagnosis module: PaddleDetection
  5. Expert knowledge module: RAG
  6. Conclusion
  7. Appendix

Figures (2)

  • Figure 1: The substructures of the auricle and different types of ear deformities. (a) presents the morphology of a normal ear from one newborn. (b) shows examples of some sub-types of auricular deformities. The abnormal structures are marked with arrows. The illustration is also shown from paperren2024publicly.
  • Figure 2: Workflow of the project where the interactive agent accomplishes three functions via three modules.