Calcification morphology and follow up


Purpose

Detect morphology of calcifications on a mammogram and recommend follow-up by risk stratification


Tag(s)


Panel

Breast Imaging

Define-AI

19060011

Originator

Kavita Mukund Patel

Lead

Kavita Mukund Patel

Panel Chair

Elizabeth Burnside

Panel Reviewers

Breast Imaging

License

Creative Commons 4.0

Status

Public Comment

Clinical Implementation


Value Proposition

An algorithm that detects morphology of calcifications of the breast would likely improve the reporting turnaround time. Furthermore this system could learn from outcomes of patients with similar calcification patterns in a manner to recommend appropriate follow up at the point of care. This would likely decrease unnecessary biopsies and follow-up recommendations.

Narrative(s)

Patient over 70 years old presents with round, punctate grouped calcifications on a baseline mammogram. Algorithm detects regional calcifications in the breast and the radiologist determines these are benign calcifications with the aid of the algorithm.

Workflow Description

Digital data from standard mammographic and tomosynthesis breast exams are sent to a server, which can be accessed by the AI algorithm. The algorithm then analyzes the data and dynamically generates breast calcification information for the interpreting radiologist, as well as automatically populating appropriate information into the radiologist’s report.

Considerations for Dataset Development


Procedure: Screening mammography and screening tomosynthesis

Views: Magnification views in the CC and lateral projection

Age:
Over 40 years

Breast Anatomy: No exclusion criteria

Biopsy-proven calcifications: benign, malignant

Technical Specifications


Inputs

DICOM Study

Procedure

Mammogram

Views

Bilateral CC and MLO mammography images

Data Type

DICOM

Modality

MAM

Body Region

Chest

Anatomic Focus

Breast

Pharmaceutical


Scenario



Primary Outputs

Breast calcification morphology

RadElement ID


Definition

morphology of breast calcification

Data Type

Categorical

Value Set

1a. Skin

1b. Vascular

1c. Coarse or “popcorn-like”

1d. Large rod-like

1e. Round

1f. Rim

1g. Dystrophic

1h. Milk of calcium

1i. Suture

2a. Amorphous

2b. Coarse heterogeneous

2c. Fine pleomorphic

2d. Fine linear or fine linear branching

3a.Diffuse

3b. Regional

3c. Grouped

3d. Linear

3e. Segmental

Units



 

Follow up recommendation

RadElement ID


Definition

Follow up recommendations from BI-RADS Fifth Edition

Data Type

Categorical

Value Set

0-Incomplete - need additional imaging

1-Negative

2-Benign

3-Probably benign - initial short-interval follow-up suggested

4-Suspicious abnormality - biopsy should be considered

5- Highly suggestive of malignancy- appropriate action should be taken

6- Known biopsy-proven malignancy- appropriate action should be taken

Units

N/A


Secondary Outputs

Calcification segmentation

RadElement ID


Definition

segment the calcifications

Data Type

Coordinates

Value Set


Units