Cholecystitis on US

Purpose	Detection of cholecystitis on ultrasound
Tag(s)
Panel	Abdominal
Define-AI ID	19020004
Originator	Luther B. Adair, II, MD
Lead	Luther B. Adair, II, MD
Panel Chair	Arun Krishnaraj, MD
Panel Reviewers	Abdominal Panel
License	Creative Commons 4.0
Status	Public Comment
RadElement Set	RDES143

Clinical Implementation

Value Proposition

Cholecystitis is a potentially life-threatening emergency, particularly if the gallbladder is perforated. AI meeting this use case would help in detecting and alerting both nonspecialists and radiologists to this potentially life-threatening abnormality. In clinical settings where a radiologist is not readily available, such detection would be of value to non-radiologists such as technologists who would perform an ultrasound for this disease entity. In clinical settings where radiologists are available, such detection could be used to alert the radiologist and prioritize its reporting and notification to the treating physicians. One possible extension to this use case is to detect cholecystitis on ultrasound and assess the need for further evaluation or intervention.

Narrative(s)

A 40-year old patient presents with constant right upper quadrant pain and is taken to the emergency room at which time ultrasound of the right upper quadrant abdomen is performed.
A 67-year old patient in the ICU with a history of diabetes presents with fever, but is intubated and is a poor candidate for an ultrasound. He is taken to the radiology department and an ultrasound of the right upper quadrant abdomen is performed.

Workflow Description

An algorithm evaluates the ultrasound and categorizes whether cholecystitis is present, equivocal, or complicated. If present, it then assesses whether a gallstone is present, absent, or undetermined; assesses whether pericholecystic fluid is present, absent, or undetermined; assesses whether the gallbladder wall is thickened or not thickened with a measurement; assesses whether there is a Sonographic Murphy’s sign which would require input from the technologist. For cases returned where the algorithm returns cholecystitis, the radiologist is made aware of the need for further evaluation or intervention. If the radiologist is not available, the non-radiologist is made aware of the need for further evaluation or intervention.

Considerations for Dataset Development

Procedure(s)	{Ultrasound limited; Ultrasound of the abdomen complete, Ultrasound of the right upper quadrant}
View(s)	{transverse, long} Clinical Note: Subtle gallbladder wall thickening can be overlooked
Sex at Birth	{Male, Female}
Age	[0,90]
Abdominal pain	{right upper quadrant pain}
Comorbidities	{female, over 40 years of age, diabetes, sickle cell anemia, obesity}
Cholecystostomy Tube	{ percutaneous drainage tube}
Other	{adenomyomatosis, cholesterol crystals, gallbladder carcinoma}

Technical Specifications

Inputs

DICOM Study

Primary Outputs

Cholecystitis Detection

 

Gallstone Detection

Secondary Outputs

Sconographic Murphy’s sign detection

Gallbladder wall thickness measurement

Pericholecystic fluid detection

Future Development Ideas

• Consider corroborating CT and ultrasound data

• Considering providing a scale of certainty based on findings that coincide with existing data to suggest a spectrum of specificity (i.e. presence of gallstones and sonographic Murphy’s sign are the most specific findings in cholecystitis). 

Related Datasets

No known related public datasets at this time,  please alert us if you know of any.