Representing Social Network Patient Data as Evidence-Based Knowledge to Support Decision Making in Disease Progression for Comorbidities

Social network patient data for comorbid studies is a sparsely explored avenue. This can provide unprecedented insight into disease conditions and their progression, hence facilitating improvement of healthcare and public health services. Structuring scattered social network data and mapping with standard disease ontologies to build reference-able knowledge base can be used in evidence-based decision support systems. In this paper, we attempt to address this direction of application where patient and time relationships are established between conditions to predict disease progression trends for comorbidities. Our prediction analysis is based on statistical modeling of the constructed knowledge base. It can be utilized towards driving personalized healthcare by applying life streams to patient journals that can provide a timed pattern of disease progression which can be used relatively and statistically for decision making and educational insight. We present and validate our approach using case-study of Brain Aneurysm with performance in terms of sensitivity and time-probability measures.