Skip to main content

About this Research Topic

Submission closed.

This Research Topic is the second volume of "Machine Learning in Clinical Decision-Making". Please see the first volume here.

Clinicians make complex medical decisions under time constraints ...

This Research Topic is the second volume of "Machine Learning in Clinical Decision-Making". Please see the first volume here.

Clinicians make complex medical decisions under time constraints and uncertainty using highly variable hypothetical-deductive reasoning and individual judgement. Time constraints are imposed by acute diseases and high clinical workloads; uncertainty results from insufficient knowledge, data, and evidence regarding possible diagnoses and treatments. Under time constraints and uncertainty, clinicians often rely on heuristics, or cognitive shortcuts. Heuristics introduce bias, leading to error and preventable harm. Approximately 10-15% of all post-mortem examinations reveal major diagnostic errors. Clinical decision-support systems intend to prevent errors by providing evidence-based risk assessments and recommendations, but many require time-consuming manual data acquisition and entry, which impairs their clinical adoption. In addition, many traditional decision-support tools apply static variable thresholds to additive and linear models that misrepresent individual patient pathophysiology, compromising their accuracy.

Machine learning systems using automated electronic health record data inputs have the potential to revolutionize clinical decision-support by obviating manual data entry and providing accurate risk assessments and recommendations tailored to individual patients. Several significant barriers to clinical adoption remain:
1) The use of biased or misrepresentative training data could produce erroneous model outputs,
2) Large, robust datasets are needed to allow self-learning, minimize the impact of outliers in small data, and ensure model validity and generalizability, and
3) Many patients could be harmed in a short time frame if model outputs are not carefully monitored and interpreted.

These challenges could be addressed by applying data standardization frameworks to large-scale, multi-institutional electronic health record data and actionable, interpretable machine learning models that provide outputs to clinicians’ mobile devices. The goal of the Research Topic is to assimilate evidence and perspectives from researchers and thought leaders that are pursuing the safe, effective development and clinical application of machine learning systems to augment clinical decision-making.

This Research Topic will include the following topics, addressing promises and pitfalls for each:
• Clinical Decision Analysis;
• Machine Learning Predictive Analytics for Clinical Decision-Support;
• Markov Decision Processes and Reinforcement Learning;
• Machine learning with Electronic Health Record Data;
• Mobile Device Interfaces for Clinical Decision-Support Tools;
• Clinical Implementation and Investigation of Clinical Decision-Support Tools

Keywords: artificial intelligence, machine learning, reinforcement learning, decision analysis, decision-making, #CollectionSeries

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.

Topic Editors


Topic Coordinators


Recent Articles



Sort by:





total views views downloads topic views

Top countries
Top referring sites

Share on

About Frontiers Research Topics

With their unique mixes of varied contributions from Original Research to Review Articles, Research Topics unify the most influential researchers, the latest key findings and historical advances in a hot research area! Find out more on how to host your own Frontiers Research Topic or contribute to one as an author.