EP-4738381-A1 - MEDICAL IMAGING DEVICE PREDICTION AND RECOMMENDATION FOR OPTIMAL IMAGE QUALITY

EP4738381A1EP 4738381 A1EP4738381 A1EP 4738381A1EP-4738381-A1

Abstract

Techniques for predicting and recommending the best medical imaging device for fulling an order are presented. A system receives an order requesting performance of a medical imaging exam on a patient, the order comprising information identifying a modality, a clinical indication and an anatomical region for the exam. The system identifies different medical imaging devices capable of performing the exam based on the modality, and employs one or more first machine learning processes to estimate measures of quality of the exam capable of being generated by the different medical imaging devices based on the order information and historical information regarding historical medical imaging exams performed by the different medical imaging devices that satisfy the order information. The system selects a subset of the different medical imaging devices to recommend for fulfilling the order based on respective measures of quality for the subset satisfying a quality criterion.

Inventors

RODRIGUEZ, Ezra Nathaniel
NOLAN, Jason
HORBAIL, Suranjani
RANGAVAJHALA, VAMSEE

Assignees

GE Precision Healthcare LLC

Dates

Publication Date: 20260506
Application Date: 20251010

Claims (15)

A method, comprising: receiving, by a system comprising at least one processor, an order requesting performance of a medical imaging exam on a patient, the order comprising order information identifying a modality of the medical imaging exam, a clinical indication for the medical imaging exam, and an anatomical region to be captured in the medical imaging exam; identifying, by the system, different medical imaging devices capable of performing the medical imaging exam based on the modality; employing, by the system, one or more first machine learning processes to estimate measures of quality of the medical imaging exam capable of being generated by the different medical imaging devices based on the order information and historical imaging exam information regarding historical medical imaging exams performed by the different medical imaging devices that satisfy the order information; selecting, by the system, a subset of the different medical imaging devices based on respective measures of quality for the subset satisfying a quality criterion; generating, by the system, recommendation information identifying the subset; and providing, by the system, the recommendation information to at least one of, a device associated with the patient or a scheduling system configured to schedule the medical imaging exam for the patient in accordance with the recommendation information.
The method of claim 1, wherein the employing comprises: training, by the system, one or more machine learning models to estimate training measures of quality corresponding to the measures of quality based on the historical imaging exam information, resulting in one or more trained versions of the one or more machine learning models; and applying, by the system, the order information and device information identifying the different medical imaging devices as input to the one or more trained versions to generate the measures of quality.
The method of claim 1, wherein the quality criterion comprises an acceptable measure of quality for the medical imaging exam.
The method of claim 3, further comprising: employing, by the system, the one or more first machine learning processes or one or more second machine learning processes to estimate the acceptable measure of quality for the medical imaging exam based on the order information and the historical imaging exam information.
The method of claim 3, further comprising: employing, by the system, the one or more first machine learning processes or one or more second machine learning processes to estimate, based on the order information and the historical imaging exam information, radiation doses capable of being exposed to the patient via the different medical imaging devices in association with performing the medical imaging exam and achieving the acceptable measure of quality, wherein the selecting the subset is further based on respective radiation doses for the subset satisfying a radiation dose criterion.
The method of claim 5, wherein the employing comprises: training, by the system, one or more machine learning models to estimate training radiation doses corresponding to the radiation doses based on the historical imaging exam information, resulting in one or more trained versions of the one or more machine learning models; and applying, by the system, the order information, device information identifying the different medical imaging devices, and the acceptable measure of quality as input to the one or more trained versions to generate the radiation doses.
The method of claim 5, wherein the employing the one or more first machine learning processes or the one or more second machine learning processes further comprises determining respective acquisition protocols capable of being used by the subset in association with performing the medical imaging exam and achieving the acceptable measure of quality with the respective radiation doses for the subset that satisfy the radiation dose criterion, and wherein the generating the recommendation information further comprises including acquisition protocol information with the recommendation information identifying the respective acquisition protocols for the subset.
The method of claim 7, wherein the selecting the subset is further based on the respective acquisition protocols for the subset satisfying an acquisition protocol criterion.
The method of claim 5, wherein the employing the one or more first machine learning processes or the one or more second machine learning processes further comprises determining respective acquisition parameters capable of being used by the subset in association with performing the medical imaging exam and achieving the acceptable measure of quality with the respective radiation doses for the subset that satisfy the radiation dose criterion, and wherein the generating the recommendation information further comprises including acquisition parameter information with the recommendation information identifying the respective acquisition parameters for the subset.
The method of claim 9, wherein the selecting the subset is further based on the respective acquisition parameters for the subset satisfying an acquisition parameter criterion.
The method of claim 5, wherein the employing the one or more first machine learning processes or the one or more second machine learning processes further comprises determining respective technicians capable of being used by the subset in association with performing the medical imaging exam and achieving the acceptable measure of quality with the respective radiation doses for the subset that satisfy the radiation dose criterion, and wherein the generating the recommendation information further comprises including technician information with the recommendation information identifying the respective technicians for the subset.
The method of claim 1, further comprising: determining, by the system, one or more patient parameters relevant to the medical imaging exam included in an electronic medical record for the patient, wherein the employing further comprises employing the one or more first machine learning processes to estimate the measures of quality based on the one or more patient parameters.
A system, comprising: at least one memory that stores computer-executable components; and at least one processor that executes the computer-executable components stored in the at least one memory, wherein the computer-executable components comprise: a reception component that receives an order requesting performance of a medical imaging exam on a patient, the order comprising order information identifying a modality of the medical imaging exam, a clinical indication for the medical imaging exam, and an anatomical region to be captured in the medical imaging exam; a filtering component that identifies different medical imaging devices capable of performing the medical imaging exam based on the modality; an artificial intelligence component that employs more first machine learning processes to estimate measures of quality of the medical imaging exam capable of being generated by the different medical imaging devices based on the order information and historical imaging exam information regarding historical medical imaging exams performed by the different medical imaging devices that satisfy the order information; an assessment component that selects a subset of the different medical imaging devices based on respective measures of quality for the subset satisfying a quality criterion; a recommendation component that generates recommendation information identifying the subset and provides the recommendation information to at least one of, a device associated with the patient or a scheduling system configured to schedule the medical imaging exam for the patient in accordance with the recommendation information.
The system of claim 13, wherein the computer-executable components further comprise: a training component that trains one or more machine learning models to estimate training measures of quality corresponding to the measures of quality based on the historical imaging exam information, resulting in one or more trained versions of the one or more machine learning models; and an execution component that applies the order information and device information identifying the different medical imaging devices as input to the one or more trained versions to generate the measures of quality.
A non-transitory machine-readable storage medium, comprising executable instructions that, when executed by a processor, facilitate performance of operations, comprising: receiving an order requesting performance of a medical imaging exam on a patient, the order comprising order information identifying a modality of the medical imaging exam, a clinical indication for the medical imaging exam, and an anatomical region to be captured in the medical imaging exam; identifying different medical imaging devices capable of performing the medical imaging exam based on the modality; employing one or more first machine learning processes to estimate measures of quality of the medical imaging exam capable of being generated by the different medical imaging devices based on the order information and historical imaging exam information regarding historical medical imaging exams performed by the different medical imaging devices that satisfy the order information; selecting a subset of the different medical imaging devices based on respective measures of quality for the subset satisfying a quality criterion; generating recommendation information identifying the subset; and providing the recommendation information to at least one of, a device associated with the patient or a scheduling system configured to schedule the medical imaging exam for the patient in accordance with the recommendation information.

Description

TECHNICAL FIELD This disclosure relates generally to medical imaging, and more particularly to medical imaging device prediction and recommendation for optimal image quality (e.g., via employment of machine learning). BACKGROUND Medical imaging devices of the same modality are not created equally. Within a single imaging modality (e.g., magnetic resonance imaging (MRI), computed tomography (CT), X-ray, and others), there are significant differences in terms of technology, performance, and capabilities. These variations can affect the quality of the images, the type of information captured, patient comfort and safety in terms of amount of radiation dosage exposure. Poor image quality can create a repeat exam scenario causing the patient to go back for the same imaging exam and may increase the radiation dosage given to the patient depending on the modality. Furthermore, the need for repeat imaging may cause a delay in treatment, especially if the patient is an outpatient. In addition, it is possible for exams to be unable to be performed at all if the device is not equipped to do so. Accordingly, techniques for determining the best medical imaging device of amongst those of the same imaging modality available for fulfilling a medical imaging order that provides the best image quality while minimizing radiation exposure and the need for repeat imaging and while further accounting for a multitude of other scheduling considerations are desired. SUMMARY The following presents a simplified summary of the specification in order to provide a basic understanding of some aspects of the specification. This summary is not an extensive overview of the specification. It is intended to neither identify key or critical elements of the specification, nor delineate any scope of the particular implementations of the specification or any scope of the claims. Its sole purpose is to present some concepts of the specification in a simplified form as a prelude to the more detailed description that is presented later. According to an embodiment, a system includes at least one memory that stores computer-executable components, and at least one processor that executes the computer-executable components stored in the at least one memory. The computer-executable components can comprise a reception component that receives an order requesting performance of a medical imaging exam on a patient, the order comprising order information identifying a modality of the medical imaging exam, a clinical indication for the medical imaging exam, and an anatomical region to be captured in the medical imaging exam. The computer-executable components further comprise a filtering component that identifies different medical imaging devices capable of performing the medical imaging exam based on the modality, and an artificial intelligence (AI) component that employs more first machine learning processes to estimate measures of quality of the medical imaging exam capable of being generated by the different medical imaging devices based on the order information and historical imaging exam information regarding historical medical imaging exams performed by the different medical imaging devices that satisfy the order information. The computer-executable components further comprise an assessment component that selects a subset of the different medical imaging devices based on respective measures of quality for the subset satisfying a quality criterion, and a recommendation component that generates recommendation information identifying the subset and provides the recommendation information to at least one of, a device associated with the patient or a scheduling system configured to schedule the medical imaging exam for the patient in accordance with the recommendation information. In one or more implementations, the computer-executable components further comprise a training component that trains one or more machine learning models to estimate training measures of quality corresponding to the measures of quality based on the historical imaging exam information, resulting in one or more trained versions of the one or more machine learning models, and an execution component that applies the order information and device information identifying the different medical imaging devices as input to the one or more trained versions to generate the measures of quality. In some embodiments, the quality criterion comprises an acceptable measure of quality for the medical image exam. In an implementation of these embodiments, wherein the AI component further employs the one or more first machine learning processes or one or more second machine learning processes to estimate the acceptable measure of quality for the medical image exam based on the order information and the historical imaging exam information. In some embodiments, the AI component further employs the one or more first machine learning processes or one or more second machine learning processes to estimate, bas