SBIR Phase I: Artificial Intelligence (AI) chatbot providing flare-up support for patients with endometriosis

这个小企业创新研究（SBIR）第一阶段项目的更广泛的影响/商业潜力是一种为子宫内膜异位症患者提供支持的新型聊天机器人。子宫内膜异位症是一种使人衰弱的慢性疼痛疾病，影响美国约十分之一的女性-导致生产力显著下降，生活质量低下，每年经济负担超过800亿美元。这个人工智能（AI）聊天机器人旨在对话式地教育用户在疼痛的子宫内膜异位症发作等情况下的临床接受的护理模式和选择。该项目的最终目的是提高患者的自我认识和生活质量，扩大农村社区患者的就诊机会，改善子宫内膜异位症患者的康复，并减少由于缺乏知识或对适当的自我护理犹豫不决而造成的疾病负担。这个小企业创新研究（SBIR）第一阶段项目将使用先进的自然语言处理（NLP）开发一个按需子宫内膜异位症发作管理聊天机器人。理解（NLU）。许多子宫内膜异位症患者每天都会经历疼痛或症状，但在发作期间症状可能变得难以忍受，导致患者几乎没有支持寻求急诊室护理。对话聊天机器人将接受培训，以回答用户的问题，指导他们进行减轻疼痛的练习，如骨盆底治疗，并为专业资源提供方向。该框架使用内部自然语言理解单元来创建用户意图、变量实体、会话上下文和会话恢复槽填充。该技术结合了情感理解和移情镜像，允许聊天机器人与用户进行自然的自由文本对话，鼓励他们自由表达自己，并在危机时刻感到被理解。这个聊天机器人将作为患者的数字伴侣，既作为支持的朋友（倾听和验证），又作为虚拟教练（指导患者通过锻炼来缓解症状）。该技术项目将在有限的患者试验中设计、开发和验证聊天机器人。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is a novel chatbot that provides support to endometriosis patients. Endometriosis is a debilitating chronic pain disorder that impacts approximately one in ten women in the United States - resulting in significant loss of productivity, low quality of life, and greater than $80 billion annual economic burden. This artificial intelligence (AI) chatbot aims to conversationally educate users regarding clinically accepted paradigms of care and options during scenarios including painful endometriosis flare-ups. This project ultimately aims to improve patients' self-knowledge and quality of life, expand access to patients in rural communities to improve endometriosis patient recovery, and reduce the burden of the disease due to lack of knowledge or hesitancy for proper self-care.This Small Business Innovation Research (SBIR) Phase I project will develop an on-demand endometriosis flare-up management chatbot using advanced Natural Language Processing (NLP) and Understanding (NLU). Many endometriosis patients experience pain or symptoms on a daily basis but symptoms can become unbearable during flare-ups, causing patients with little support to seek emergency room care. The conversational chatbot will be trained to answer user’s questions, guide them through pain-reducing exercises such as pelvic floor therapy, and provide direction to professional resources. This framework uses an internal Natural Language Understanding unit to create user intents, variable entities, conversation context, and conversation resuming slot filling. The technology incorporates emotional understanding and empathetic mirroring allowing the chatbot to have a natural freeform text conversation with the user to encourage them to freely express themselves and feel understood during moments of crisis. This chatbot will serve as a digital companion for patients, acting as both a supportive friend (listening and validating) and as a virtual coach (guiding the patient through exercises to alleviate symptoms). This technology project will design, develop and validate the chatbot in a limited patient pilot.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.