Detection of pulmonary nodules in CCTA acquisition

Purpose

The algorithm should identify the pulmonary nodules in a common CCTA.

Tag(s)

 

Panel

Cardiac Panel

Define-AI ID

19040038

Originator

Giuseppe Muscogiuri
Lead  Giuseppe Muscogiuri

Panel Chair

Carlo De Cecco

Panel Reviewers

Cardiac Panel

License

Creative Commons 4.0 
Status Public Commenting 
RadElement Set RDES111
                               

Clinical Implementation


Value Proposition


In clinical practice, automated detection of pulmonary nodules in coronary computed tomography angiography (CCTA) should speed up the reporting time. Patients who undergo cardiac CT could be evaluated in the same examination for presence of nodules with high probability of malignancy as well as presence of coronary artery disease.

Narrative(s)


Patients undergo to CCTA for rule out or follow up of CAD, commonly have lung disease. Patients undergo cardiac CT for the assessment of CAD or evaluation of calcium score. Images reconstructed with large field of view (FOV) can be analyzed using the AI algorithm.

Workflow Description


CCTA and CT for calcium score acquisition are acquired in clinical routine. Images with large field of view are elaborated and send to PACS system.
Subsequently the AI algorithm analyzes the lung, identifying nodules and alerting the physician in presence of pulmonary nodules.

Considerations for Dataset Development



Procedures

CT, Chest

Views

{Axial}

Age

[0,90]

Comorbidities

{presence of chronic obstructive pulmonary disease)

Breath artifacts

present, absent

Clinical Notes : Breath artifacts can provide both false positive and false negative.

Limphonodes

present, absent

Technical Specifications


Inputs

DICOM Study

Procedure

CT, Chest

Views

Axial

Data Type

DICOM

Modality

CT

Body Region

Chest

Anatomic Focus

Lung


Primary Outputs

 

Pulmonary Nodules Detection

RadElement ID

RDE706

Definition

Detection of pulmonary nodules during a CT acquisition

Data Type

Categorical

Value Set

  • Pulmonary nodule absent
  • Pulmonary nodule present
  • Unknown

Units

N/A


Secondary Outputs

 

Pulmonary Nodule Attenuation

 

RadElement ID

RDE707

Definition

Define the attenuation of pulmonary nodules.

Data Type

Categorical

 

Value Set

 

 

  • fat density
  • groundglass
  • part solid
  • solid
  • calcification
  • cavitation
  • cystic lucencies
  • air bronchograms

 

Units

N/A

 

 

Pleural Involvement

 

 

RadElement ID

RDE708

Definition

Evaluation of pleural involvement

Data Type

Categorical

Value Set

  • Pleural involvement absent
  • Pleural involvement present
  • Unknown

Units

N/A

Future Development Ideas


  • Evaluation of AI algorithm for pulmonary nodules detection can be  further developed for patients with history of carcinoma.