Scholarlink

Lifestyle changes supported by continuous monitoring and timely expert advice can play a significant role in combating chronic diseases such as diabetes. Although technologies such as IoT, Big Data and Artificial Intelligence can enable such monitoring through the use of wearable devices and centralized intelligent data processing, achieving the robustness and trustworthiness required in the sensitive domain of healthcare remains a challenge. In this work, we propose a novel architecture for fusing the expert knowledge provided by a Healthcare Professional with a neural network that learns from the patient’s biometric data using NeuroSymbolic AI and specifically Logic Tensor Networks. In this new framework, both the AI expert and the Healthcare Professional are enabled to train the model in a collaborative manner, so as to ensure its alignment with established medical knowledge. The resulting system has been preliminarily evaluated on open-source healthcare datasets as well as a real-life use case of diabetic patients led by Innovation Sprint.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A NeuroSymbolic Human-in-the-Loop Approach Towards Fusing Medical Expert Knowledge with ANNs

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Spyros Theodoropoulos,
  • Georgios Makridis,
  • Aristodemos Pnevmatikakis,
  • Vrettos Moulos,
  • Dimosthenis Kyriazis,
  • Panayiotis Tsanakas

摘要

Lifestyle changes supported by continuous monitoring and timely expert advice can play a significant role in combating chronic diseases such as diabetes. Although technologies such as IoT, Big Data and Artificial Intelligence can enable such monitoring through the use of wearable devices and centralized intelligent data processing, achieving the robustness and trustworthiness required in the sensitive domain of healthcare remains a challenge. In this work, we propose a novel architecture for fusing the expert knowledge provided by a Healthcare Professional with a neural network that learns from the patient’s biometric data using NeuroSymbolic AI and specifically Logic Tensor Networks. In this new framework, both the AI expert and the Healthcare Professional are enabled to train the model in a collaborative manner, so as to ensure its alignment with established medical knowledge. The resulting system has been preliminarily evaluated on open-source healthcare datasets as well as a real-life use case of diabetic patients led by Innovation Sprint.