‘Precision Health’: Balancing Reactive Care and Proactive Care Through the Evidence Based Knowledge Graph Constructed from Real-World Electronic Health Records, Disease Trajectories, Diseasome, and Patholome

Abstract

Health care can be either ‘Reactive Care’ or ‘Proactive Care’. Reactive care is self-referral where a medical help is solicited by the person or family members on suspecting illness. In Proactive care an individual seeks medical help before the appearance of symptoms in order to prevent illness, or detect and treat it early before the disease progresses or becomes chronic. There are advantages and disadvantages for both of these approaches. Reactive approach relies on healing followed by self-referrals wherein the right care is often delayed, or even neglected, resulting in accelerated disease progression. Proactive care, on the other hand, takes into account the potential risk factors in a person’s health. Proactive care carries risks of overdiagnosis, overtreatment, and unnecessary interventions. In this paper we make a balance between the reactive care and the proactive care through the use of data driven algorithms, models, and knowledge graphs. We show how diseasome network constructed from tacit knowledge of Spatial Comorbidity and Temporal Comorbidity, and Patholome explicit knowledge can offer ‘Precision Health’. We looked at the real-world EHR data (mostly reactive diagnosis by hundreds of doctors) to construct spatial comorbidity knowledge network. We then combined disease trajectories data (temporal comorbidity) with the spatial comorbidity. This helped us understand how diseases manifest in a target population and their interrelationships. Finally we constructed patholome disease-diagnostic-test explicit knowledge and integrated with the diseasome knowledge network to form evidence based Knowledge Graph or a Clinical Expert System. We added a Semantic Engine (Reasoning Knowledge Network) on this statistically significant knowledge graph to help a health service provider to make an accurate informed decision on balancing the reactive care and the proactive care with a focus on ‘Right Care’ through explainable AI (XAI). To offer the knowledge driven right care at the right time at anywhere point-of-care we used Big-Data Analytics, Statistics, Artificial Intelligence, Knowledge Discovery & Management, WebRTC, and Smartphones.