| CPC G06Q 10/0635 (2013.01) [G06N 20/00 (2019.01); G06Q 10/06393 (2013.01); G16H 40/20 (2018.01)] | 12 Claims |
|
1. A system for automatically deriving correlations between healthcare facility safety data and risk factors, the system comprising:
at least one non-volatile storage region configured to store
a plurality of facility incident records each corresponding to one of a plurality of healthcare facilities, each incident record regarding a respective liability claim or a respective safety incident,
healthcare facility data comprising a plurality of healthcare facility attributes of each healthcare facility of the plurality of healthcare facilities, and
a plurality of medical team records corresponding to a plurality of medical professionals each associated with one or more of the plurality of healthcare facilities; and
processing circuitry configured to perform operations, comprising
accessing, from the plurality of facility incident records, injury data corresponding to a plurality of injury outcomes, a plurality of injury events, and a plurality of injury causes, wherein the injury data comprises identification of a set of facility attributes of the plurality of healthcare facility attributes and a set of medical professionals of the plurality of medical professionals,
automatically applying a set of standardized labels to categorize each of the plurality of injury outcomes, the plurality of injury events, and the plurality of injury causes into a respective smaller set of actionable terms consistent across the plurality of healthcare facilities,
accessing, from the plurality of medical team records, medical team data corresponding to the set of medical professionals,
after applying the set of standardized labels, providing, to a feature learning engine comprising one or more machine learning models, the injury data, the medical team data, and the healthcare facility data, wherein
the feature learning engine is configured to identify, from the medical team data and/or the healthcare facility data, a set of risk correlation factors, wherein
each risk correlation factor of the set of risk correlation factors correlates to at least one respective increased injury risk of a set of injury risks according to the injury data, and
each risk correlation factor of the set of risk correlation factors identifies a respective attribute of a plurality of attributes and at least one value of the respective attribute, wherein
the plurality of attributes comprises the set of facility attributes, a set of medical team attributes, and a set of patient attributes,
obtaining, from the feature learning engine responsive to the providing, the set of risk correlation factors,
training, using the facility attributes, the medical team data, and a portion of the set of risk correlation factors, at least one machine learning model to predict risk of safety incidents due to the portion of the set of risk correlation factors based on a collection of facility attributes and a collection of medical team data of a target medical facility.
|