US 12,249,424 B2
Decision-support tools for pediatric obesity
Andrew Roberts, Overland Park, KS (US); and Sasanka Are, Kansas City, MO (US)
Assigned to CERNER INNOVATION, INC., Kansas City, MO (US)
Filed by CERNER INNOVATION, INC., Kansas City, KS (US)
Filed on Oct. 5, 2018, as Appl. No. 16/153,328.
Claims priority of provisional application 62/568,487, filed on Oct. 5, 2017.
Prior Publication US 2019/0108916 A1, Apr. 11, 2019
Int. Cl. G16H 50/30 (2018.01); G16H 10/60 (2018.01); G16H 20/10 (2018.01); G16H 20/30 (2018.01); G16H 20/60 (2018.01); G16H 40/20 (2018.01); G16H 40/63 (2018.01); G16H 50/50 (2018.01)
CPC G16H 50/30 (2018.01) [G16H 10/60 (2018.01); G16H 20/10 (2018.01); G16H 20/30 (2018.01); G16H 20/60 (2018.01); G16H 40/20 (2018.01); G16H 40/63 (2018.01); G16H 50/50 (2018.01)] 33 Claims
OG exemplary drawing
 
1. One or more non-transitory media having instructions that, when executed by one or more processors, cause the one or more processors to facilitate a plurality of operations for providing intervention for pediatric obesity, the operations comprising:
receiving patient data for a pediatric patient, the patient data including an age and a body mass index (BMI) value of the pediatric patient;
assigning an obesity risk level to the pediatric patient based on the age and the BMI value of the pediatric patient,
wherein:
the obesity risk level is defined by one or more obesity risk curves that are each generated using an age-dependent multiplier,
the age-dependent multiplier is formed based at least in part on a growth velocity value,
the growth velocity value is determined based at least in part on reference information for a set of reference pediatric information, and
the reference information comprises data associated with one or both of an age indication and a BMI indication for each of a set of reference individuals within the set of reference pediatric information;
predicting, for each growth velocity value, for a plurality of ages, a proxy for obesity based on a reference pediatric population and based further on a machine-learning model,
wherein:
(a) the machine-learning model is trained by inputting, to the machine-learning model, proxy content corresponding to instances of information selected from a group comprising: (i) a health care spend amount for an individual in the reference pediatric population and (ii) at least one chronic condition, for the individual, from a set of chronic conditions associated with the reference pediatric population, and
(b) the predicting comprises applying, to the trained machine-learning model, data associated with at least a portion of the proxy content or the instances of information to generate a model output that facilitates the prediction of the proxy for obesity; and
wherein, based on the obesity risk level for the pediatric patient, one or more of (a) electronically notifying a caregiver, (b) recommending a treatment plan, (c) modifying a healthcare software program, or (d) scheduling a healthcare resource is initiated.