Developing an IoT-based conversational AI Recommender assistant for vital sign predicted anomalies

Abstract

In most real-time scenarios such as emergency first response or a patient self-monitoring using a wearable device, it is likely that accessing a healthcare physician for assessing potential vital sign anomalies and provide a recommendation will be impossible; thus potentially putting the patient at risk. Leveraging the latest advances in Natural Language Processing (NLP), this study presents a research-driven design and development of a cloud-based conversational AI platform trained to predict vital signs anomalies and provides recommendations from a dataset created by physicians. To reinforce the learning of the virtual assistant, the Conversation Driven Development (CDD) methodology has been adopted to involve end users in the testing process in the early phase. The proposed platform will help to manage the consequences of low physician-patient ratios especially in developing countries. A part from this thesis. I have already submitted my first paper about my research project. The paper was submitted to conference 8th International Conference on Machine Learning Technologies (ICMLT 2023) which has already been accepted for publication.