Incidental Pulmonary Nodules on CT

Purpose

Detection and characterization of incidental pulmonary nodules on computed tomography (CT)

Tag(s)

 

Panel

Thoracic

Define-AI ID

08190003

Originator

Carol C. Wu, STR Big Data Subcommittee

Panel Chair

Eric J. Stern

Panel Reviewers

Thoracic Panel

License

Creative Commons 4.0

Status

Published

Clinical Implementation


Value Proposition

Incidental pulmonary nodules are commonly seen on computed tomography (CT) studies that include the lungs. A recent study shows that incidental nodules are seen in 13.9% of CT coronary angiogram (Robertson J et al. Heart Lung Circ 2017). Even though the majority of these nodules eventually prove to be benign, a small percentage of them are malignant. Follow-up chest CTs are often required to evaluate for growth, which is an indicator of malignancy. Automated identification and characterization of these nodules can result in significant improvement in workflow efficiency for radiologists. Subsequent recommendations of appropriate follow-up of these incidental pulmonary nodules based on patients’ risk factors and comorbidities are key to patient care.

Narrative(s)

A 68 year old smoker presents after a motor vehicle accident and undergoes a CT of the chest, abdomen and pelvis in the emergency room. Algorithm evaluates images of the lungs, detects and characterizes any incidental pulmonary nodules to allow radiologists to quickly review and insert these descriptions into the imaging report. The algorithm might also suggest appropriate follow-up recommendation based on Fleischer Society Guidelines. If follow-up imaging is required, the algorithm can prompt the referring provider to order the study prior to patient discharge from the acute care setting. If the algorithm or radiologist is able to determine that the patient meets criteria for low-dose CT lung screening based on the electronic medical record, a recommendation can be made to enroll the patient in a screening program.

A 70 year old admitted to ICU with COPD exacerbation and chest CT was performed for evaluation of respiratory distress. Algorithm evaluates image and detects small solid incidental pulmonary nodules which are stable since the patient’s oldest available comparison CT from 4 years prior. The algorithm describes the nodules and notes that the nodules are stable and do not require further imaging follow-up. The radiologist can insert these findings into the imaging report and focus on other more relevant and urgent CT findings.
A 45 year old otherwise healthy man presents to the emergency center with right lower quadrant abdominal pain and fever. A CT abdomen and pelvis reveals findings consistent with appendicitis and three cavitary nodules in the lung bases measuring up to 8mm. Algorithm detects these concerning incidental pulmonary nodules and alerts the radiologist to evaluate the lung bases. The algorithm inserts these findings into the imaging report.

Workflow Description

CT images sent to PACS and the AI engine. Image analyzed by engine. System detects and characterizes the incidental pulmonary nodules. A message is sent to PACS from the engine with the information, highlighting each incidental pulmonary nodules with a list of nodule characteristics.

Considerations for Dataset Development


Procedures(s): {Chest CT w/ or w/o intravenous contrast*, including High-resolution chest CT and low dose chest CT, cardiac CT, abdominal CT, neck CT, cervical/thoracic/lumbar spine CT, serial CTs over time}

Clinical Note: CTs that include any part of the lung

 

Slice: Continuous thin sections preferred (≤ 1.5 mm)

 

Sex at Birth: {Male, Female}

 

Age: [18, no limit] 

 

Nodule Attenuation: {solid, part-solid, groundglass, internal fat density, calcification, cavitary} 

 

Nodule Size (mm): [3,30] 

 

Nodule Shape : {2D: round, oval, triangular, lobular, irregular; 3D: spherical, flat} 

 

Nodule Margin: {smooth, irregular, spiculated, ill-defined} 

 

Nodule Location: {subpleural, peri-fissural, parenchymal, intraluminal} 

 

Comorbidities: {Smokers, non-smokers, emphysema, bronchitis, bronchiolitis, focal inflammatory lesions, usual interstitial pneumonia and other diffuse lung diseases <other diffuse lung diseases>}, other primary malignancy or history of primary malignancy

Clinical Note: Focal inflammatory lesions, particular with surrounding emphysema, can mimic lung cancers

Pathologic Diagnosis: {Fine needle aspiration, core needle biopsy, surgical specimen, if not available, obtain information regarding stability over time}

Technical Specifications


Inputs

DICOM Study

Procedure

CT

Views

Axial, multiplanar reformats, MIPS

Data Type

DICOM

Modality

CT

Body Region

Chest

Anatomic Focus

Lung

Pharmaceutical

N/A

Scenario

N/A

 

Primary Outputs

Nodule Detection

RadElement ID

 

Definition

The center x, y, and z coordinate of a candidate nodule bounding box with reference to the superior, anterior, and right-most pixel in the volume (referencing the patient for sidedness, zero indexed)

Data Type

Array

Value Set

 

Units

N/A

 

Nodule Localization

RadElement ID

 

Definition

dimensions of a bounding box in pixels (x, y, z).

Data Type

Array

Value Set

 

Units

N/A

 

Nodule Attenuation

RadElement ID

 

Definition

Determine the mean and range of HU density of the nodule

Data Type

Categorical

Value Set

0-fat density

1-groundglass

2-part solid

3-solid

4-calcification

5-cavitation

6-cystic lucencies

7-air bronchograms

Units

N/A

 

Nodule Diameter

RadElement ID

 

Definition

Measure largest diameters of nodule(s) in any plane, including 3D. Optional: For part solid nodule, return both the overall diameter and solid component diameter.

Data Type

Numerical

Value Set

 

Units

mm

 

Nodule Volume

RadElement ID

 

Definition

Measure the volume of the nodule(s). Optional: Measure volume of nodule(s) from all planes. Also optional: if part solid nodule, return both the overall diameter and solid component diameter in mm.

Data Type

Numerical

Value Set

 

Units

Mm3

 

Secondary Outputs

Nodule Shape

RadElement ID

 

Definition

Describe nodule shape

Data Type

Categorical

Value Set

0-Unknown

1-round

2-oval

3-triangular

4-irregular

5-lobular

6-other

Units

N/A

 

Nodule Margin

RadElement ID

 

Definition

Classify shape/margin of nodule

Data Type

Categorical

Value Set

1-smooth

2-lobulated

3-spiculated

Units

N/A

 

Nodule Location

RadElement ID

 

Definition

State on which lung lobe nodule is located. Ideally pull the image with the incidental nodule

Data Type

Categorical

Value Set

0-Unknown

1-Right upper lobe

2-Right upper lobe

3-Right lower lobe

4-Left upper lobe

5-Left lower lobe

6-Fissural

7-intraluminal

Units

N/A

 

Nodule Growth

RadElement ID

 

Definition

Change in diameter, volume, or attenuation over time if comparison CTs are available

Data Type

Categorical

Value Set

0-Unknown

1-Unchanged (to the nearest 0.1 mm)

2-Growth

3-Shrinkage

4-Decrease attenuation

5-Increase attenuation

Units

N/A

 

Probability of Malignancy

RadElement ID

 

Definition

Likelihood the nodule is malignant based on nodule characteristics, patient demographics and smoking history

Data Type

Numerical

Value Set

[0,1]
0-Benign

1-Malignant

Units

N/A

Future Development Ideas


Based on nodule attributes and probability of malignancy, Return follow-up recommendations according to Fleischer and/or LUNG-RADS guidelines.

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.