ABSTRACT
From a high level, patient well-being, the healthy-aging process, and medical care is a spectrum of health supported by four medical pillars, including (1) staying healthy (generally being young); (2) identifying health problems early; (3) detecting, diagnosing, managing, and monitoring any diseases; and (4) identifying a proper and appropriate methodology for treating diseases. Across these pillars of health, as each patient proceeds on their journey, they generate a wealth of new data inside the health care system(s). This includes a wide array of radiologic imaging, pathology reports, cytogenetics, tissue banking, flow cytometry, physician documentation, multidisciplinary tumor board discussions, nursing and therapist notes on patient progression and status, and specialist input (Figure 12.1). During this process, massive amounts of data are collected and stored in a digital format largely because the demonstration of meaningful use has been tied to medical financial reimbursement under the Patient Protection and Affordable Care Act (ACA) and because the Centers for Medicare and Medicaid Services (CMS) has moved 90% of the fee-for-service requirements into quality metrics. Unfortunately, data collection is heterogeneous in terms of both input and output formatting (and access), rendering the data impossible to effectively manage, analyze, or data mine or use to generate clinical decisions from due to a lack of user interface, data volume, and required computational power (software and hardware). Programs to effectively manage this data within the radiation oncology setting are being developed, and they will require a system that provides adequate capacities for sorting through high data storage, with acceptance of data variety, as well as high speed for data processing and analyzing. The need for these high-level programs is why the radiation oncology field has introduced big data analytics, with the hopes of it improving clinical outcomes and optimizing the patient–physician experience and overall effectiveness.
