Scholarlink

Symptoms of post-traumatic stress disorder (PTSD) frequently occur following life-threatening emergencies such as strokes, heart attacks, and stays in the intensive care unit (ICU). In response, we are introducing “Heal-Mind”, an innovative mobile application currently in active development. This application aims to harness the potential of generative artificial intelligence (AI) to provide personalized support for individuals grappling with post-emergency PTSD. At the core of Heal-Mind lies a meticulously trained Seq2Seq Transformer model with 44.7 million parameters, which has been trained on a publicly available dataset. This generative model is specifically designed to offer empathetic and coherent responses tailored to users’ unique experiences, fostering a sense of understanding and support. While still in the developmental phase, Heal-Mind shows promising potential in addressing the distinct challenges posed by post-emergency PTSD. Users of the application will have access to an evolving range of services aimed at aiding their recovery. Notably, Heal-Mind is being developed with a strong focus on user experience, incorporating post-processing algorithms to identify contradictions, enhance coherency, and reduce repetitiveness in its responses. As our work continues to progress, we anticipate that Heal-Mind will emerge as a mental vital tool, providing accessible and compassionate assistance to individuals navigating the complexities of post-emergency PTSD.

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

Towards a Deep Learning Post-traumatic Stress Disorder Dialogue System Based on Transformers

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Nourchene Ouerhani,
  • Ahmed Maalel,
  • Dalibor P. Drljača,
  • Henda Ben Ghezala

摘要

Symptoms of post-traumatic stress disorder (PTSD) frequently occur following life-threatening emergencies such as strokes, heart attacks, and stays in the intensive care unit (ICU). In response, we are introducing “Heal-Mind”, an innovative mobile application currently in active development. This application aims to harness the potential of generative artificial intelligence (AI) to provide personalized support for individuals grappling with post-emergency PTSD. At the core of Heal-Mind lies a meticulously trained Seq2Seq Transformer model with 44.7 million parameters, which has been trained on a publicly available dataset. This generative model is specifically designed to offer empathetic and coherent responses tailored to users’ unique experiences, fostering a sense of understanding and support. While still in the developmental phase, Heal-Mind shows promising potential in addressing the distinct challenges posed by post-emergency PTSD. Users of the application will have access to an evolving range of services aimed at aiding their recovery. Notably, Heal-Mind is being developed with a strong focus on user experience, incorporating post-processing algorithms to identify contradictions, enhance coherency, and reduce repetitiveness in its responses. As our work continues to progress, we anticipate that Heal-Mind will emerge as a mental vital tool, providing accessible and compassionate assistance to individuals navigating the complexities of post-emergency PTSD.