Decision Support Systems in Oncology

Abstract

Precision medicine is the future of health care: please watch the animation at https://vimeo.com/241154708 . As a technology-intensive and -dependent medical discipline, oncology will be at the vanguard of this impending change. However, to bring about precision medicine, a fundamental conundrum must be solved: Human cognitive capacity, typically constrained to five variables for decision making in the context of the increasing number of available biomarkers and therapeutic options, is a limiting factor to the realization of precision medicine. Given this level of complexity and the restriction of human decision making, current methods are untenable. A solution to this challenge is multifactorial decision support systems (DSSs), continuously learning artificial intelligence platforms that integrate all available data-clinical, imaging, biologic, genetic, cost-to produce validated predictive models. DSSs compare the personalized probable outcomes-toxicity, tumor control, quality of life, cost effectiveness-of various care pathway decisions to ensure optimal efficacy and economy. DSSs can be integrated into the workflows both strategically (at the multidisciplinary tumor board level to support treatment choice, eg, surgery or radiotherapy) and tactically (at the specialist level to support treatment technique, eg, prostate spacer or not). In some countries, the reimbursement of certain treatments, such as proton therapy, is already conditional on the basis that a DSS is used. DSSs have many stakeholders-clinicians, medical directors, medical insurers, patient advocacy groups-and are a natural consequence of big data in health care. Here, we provide an overview of DSSs, their challenges, opportunities, and capacity to improve clinical decision making, with an emphasis on the utility in oncology.

FIG 1.

Capacity versus complexity. The upsurge in data available for medical decision making threatens to overwhelm human cognitive capacity (maximum five variables per decision).

FIG 2.

Schematic diagrams of centralized and distributed approaches. In a centralized data and learning approach, multiple centers pool their data to enable learning, whereas, in a distributed data and learning approach, multiple centers link their systems to enable learning. A key aspect of a distributed approach is that it is privacy-by-design construction (ie, data remains at the source), whereas a key aspect of the centralized approach is that data can be directly accessed and scrutinized. These are the two competing tradeoffs between the approaches. DSS, decision support system; EMR, electronic medical record; PACS, picture archiving and communication system.

FIG 3.

Data-to-decision. Data sources in oncology (clinical, imaging, biologic, genetic, and costs) can be used via artificial intelligence (AI) methods in decision support systems (DSSs) to augment decision making in oncology (toxicity, tumor control, quality of life, cost effectiveness). This volume and complexity of data overload human cognitive capacity but can be mined and distilled by AI in rapid-learning health care frameworks.

FIG 4.

Decision support systems integrated into the workflow. This can be accomplished both strategically (at the multidisciplinary tumor board level to support treatment choice, eg, surgery or radiotherapy) and tactically (at the specialist level to support treatment technique, eg, prostate spacer or not).

FIG 5.

Decision support system (DSS) stakeholders. Clinicians, medical directors, medical insurers, and patient advocacy groups all share common interest in the adoption, use, evaluation, and improvement of DSSs in health care.