About this Research Topic
Information technology (IT) is playing an integral role in healthcare worldwide. With the use of predictive analytics, IT has become increasingly useful in operational personalized medicine, epidemiology, and in classifying complex care data with inferences that go beyond the imagination of the caregiver. Predictive analysis has been used mainly in the area of disease diagnosis to provide general information on illnesses based on open and anonymized datasets. However, in clinical practice, data sets are not open and are collected based on standard protocols. Protocols are not only important for data collection, but also for providing the context of the data. The way predictive analytics has been used in diagnosis strips away context and cannot bring to light the qualitative nature embedded in the medical practice in which decision-making activities should be supported by explanations. Clinical data can be further affected by several factors including variations in medical practice that have been widely documented and evidence shows that clinical decision making varies according to the patient characteristics. A prognostic analysis attempts to assess the relative importance of several predictor variables so as to inform patients how their disease will progress after their diagnosis. One reason for studying prognostic analysis is to learn the relative importance of several variables that might affect, or be associated with, disease outcome. For example, prognostic analysis can be used to find whether certain elderly patients will develop septic shocks when they are diagnosed with COVID-19.
Diagnosis requires plenty of data to uncover clinical patterns at a large scale while prognosis requires a relatively small size of data to cover the context of patient-centered patterns in greater depth. Prognosis relies on specific patient data extracted using qualitative methods and techniques, while diagnosis relies on machine learning techniques based on quantitative methods. Thus, prognosis focuses on the patient context by revealing the connections between data points related to different classes of the patient characteristics while diagnosis provides insights with a particular range of quantified data points. A prognostic analysis, therefore, attempts to assess the relative importance of several disease progression predictor variables based on patient characteristics. Its aim is to provide the data thickness required by clinical practice to identify the progression of a given disease for a particular patient group.
This Research Topic focuses on the role of healthcare data analytics for prognoses of patient cases. Subtopics of particular interest include, but are not limited to:
• Features Engineering for Prognosis
• Machine Learning for Prognosis
• Deep Learning for Prognosis
• Prognosis Progression Outcome Metrics
• Interlinking Diagnosis with Prognosis
• Prognosis for Value-Based Healthcare
• Thick Data for Prognosis
• Graphical Databases for Prognosis
• Prognosis and Clinical Trials
• Statistical Analysis for Prognosis
Keywords: Disease Prognosis, Patient Outcome, Disease Progression, Healthcare Data Analytics
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.