Browse Prior Art Database

System and Method for Real-time Predictive Analytics Guided Ultrasonography

IP.com Disclosure Number: IPCOM000245377D
Publication Date: 2016-Mar-04
Document File: 3 page(s) / 173K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a system that provides real time recommendations, based on predictive analytical models, to sonographers during ultrasonography procedures. The novel system uses a distributed design to push models and predictive processing to the network edge, close to the most current patient data in the electronic medical record (EMR), and to the real time exam data.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 49% of the total text.

Page 01 of 3

System and Method for Real-time Predictive Analytics Guided Ultrasonography

Sonographers depend on ultrasound skills, medical knowledge, and overall experience to perform ultrasonography (i.e. sonography) procedures. Sonographers must understand the associations between the information provided in the ultrasound images and possible abnormalities. This requires an in-depth knowledge of human anatomy and pathologies. Rapid improvements in ultrasound technology and advances in medical knowledge require sonographers to complete registry level assessments and ongoing education in sonography sub-specializations.

A method is needed to optimize, in terms of number, cost, time, etc., medical diagnostic procedures and improve patient outcomes.

The novel contribution is a system that provides real time recommendations, based on predictive analytical models, to sonographers during ultrasonography procedures. The novel system uses a distributed design to push models and predictive processing to the network edge, close to the most current patient data in the electronic medical record (EMR), and to the real time exam data.

The system enables dynamic workflows based on predictive modeling using patient data captured in real time during the exam, and then provides direction to the sonographer for collecting additional images and metrics. It aggregates patient information from different data silos to create a unique and most current view of the patient in real time. In addition, the system uses analytic and statistical modelling to suggest the next best action based on real-time input and the patient's previous history, compared to similar cases among the population.

Procedures Guided by Predictive Analytics

Using information from the patient's EMR, de-identified information from population health data repositories, guidance from medical journals on related pathologies, and real-time information collected during the exam, the system can apply statistical models and formulate recommendations in real time as new information is collected by the sonographer during the exam.

The system can provide multiple types of guidance. For example, to consider the likelihood of other pathologies, models can compare combinations of patient demographic information (e.g., age, ethnicity, gender, geographic location, etc.) as well as current and past problems and medical history from the EMR, along with the data collected in real-time and the patient's information, with that of many other individuals

as stored in the repository. Models can also look at common sequences of procedures based on a broad review of population patient profiles. Models may incorporate additional genetic information to consider for the scan workflow.

The system then recommends the next best actions based on the likelihood of abnormalities, emergence of chronic diseases, and other situations. The system may make recommendations for additional measurements within the scope of the supplied medic...