Pediatric Elbow

Purpose

To assist radiologists in the interpretation of pediatric elbow radiographs

Tag(s)

 

Panel

 Musculoskeletal

TOUCH-AI ID

 TAI-05180008

Originator

 Jay Patti

Panel Chair

 Jay Patti

Panel Reviewers

 Musculoskeletal Panel

License

 Creative Commons 4.0

Status  Public Commenting

Clinical Implementation


Value Proposition

When interpreting pediatric elbow radiographs, the appearance of the osseous structures can vary significantly based on the patient’s age. Normal development dictates an ordered mineralization of the apophyses of the elbow, which depends on patient age. Fractures of the elbow bones can mimic normal apophyses. Knowing the order of expected mineralization and age of the patient can help prevent diagnostic error. An AI algorithm could refine the currently known expected age of mineralization and warn radiologists when pathology mimics a normal developmental appearance in the setting of a discordant patient age. The classic mnemonic device that is used to remember the order of ossification is CRITOE, which stands for Capitellum, Radial head, Internal epicondyle, Trochlea, Olecranon, External epicondyle. In normal development, these apophyses ossify at roughly ages 2, 4, 5, 9, and 11, respectively.

Narrative(s)

A pediatric (<15 years old) patient presents for elbow radiography after trauma. There is a fracture of the lateral humeral epiphyseal apophysis that mimics normal development in a patient 3 years older than the patient’s true age. This could present a diagnostic challenge or potential area for diagnostic oversight resulting in medical error.

Workflow Description

The radiograph is processed by the algorithm, which is tuned to identify the elbow apophyses and correlate them with patient age. The engine would send a message to PACS with the coordinates of each normal apophysis, which are then highlighted with a subtle green tint. Any apophyses that are inconsistent with the patient’s age are highlighted in yellow, which alerts the radiologist to examine these suspicious areas more closely.

Considerations for Dataset Development


Procedures(s): X-ray, Elbow

View(s): {AP, Lateral, Oblique}

 

Sex at Birth: {Male, Female}

 

Age (years): [0,15]

 

Data source: {pediatric facilities, general facilities}

 

Differential Diagnoses: Apophysis injury

 

Elbow: {Left, Right}

 

Fractures: supracondylar fractures

 

Other: {Casting, bracing material}

Technical Specifications


Inputs

DICOM Study

Procedure

XRAY, Elbow

Views

AP, Lateral, Oblique

Data Type

DICOM

Modality

XRAY

Body Region

Upper Extremity

Anatomic Focus

Elbow

Primary Outputs

Abnormal Apophyses

RadElement ID

RDED246

Definition

Detect and count abnormal apophyses

Data Type

Numeric

Value Set

[0,6]

Units

Apophyses

Secondary Outputs

Age

RadElement ID

RDE247

Definition

Patient age in years

Data Type

Numeric

Value Set

 

Units

Years

Future Development Ideas


Segment capitellum, radial head, internal epicondyle, trochlea, olecranon, and external epicondyle as well as highlight the abnormal apophyses.

Public Commenting


Use cases are meant to be a primary vehicle for distributing clinical information to the developer community. They pinpoint precise scenarios within radiology workflows where potential automation could add noticeable value and establish standards for interpreting and passing corresponding common data elements. Implementing effective standards requires the perspective from all stakeholders. So to that end, we encourage your feedback on use cases.

To submit comments, please email DSIUseCases@acr.org with the use case title(s) and relevant comments by January 1, 2019. If more convenient, you may also download this use case and comment directly on the PDF. Just attach the PDF copy on the email.