Ultrasound (US) is widely
regarded as the adjunct procedure of choice to mammography, especially for
distinguishing cystic from solid masses where accuracy is 96-100%. However,
earlier studies, in which ultrasound was evaluated largely as a primary
screening tool, reported a wide variance in Positive Predictive Value (PPV) and
an unsettling range of False Negative (FN) rate ranging from 0.3-30%. These
results led to many recommendations still extant that breast ultrasound be used
only to determine cyst from solid and/or for needle guidance. Furthermore, even
with combined information from mammography and ultrasound, each radiologist may
apply a different decision threshold to recommend biopsy of a suspicious mass.
The American College of Radiology (ACR)
developed the Breast Imaging Reporting and Data System (BI-RADS) program to
guide interpretation and reporting of breast ultrasound exams. The ACR also
manages a program to accredit the clinical practice of breast ultrasound.
Similarly, in 1998 the American Institute of Ultrasound in Medicine launched a
program to accredit the breast ultrasound practices of radiologists and
sonographers. These steps by professional organizations are designed to improve
the quality and uniformity of care.
Acceptance and utilization of BI-RADS
for ultrasound is increasing but the research literature, including our own
reports, shows it is difficult to uniformly apply the method and there is
considerable variability of lesion description and assessment between
radiologists. The American College of Radiology Imaging Network (ACRIN) Research
Protocol 6666, entitled Breast Cancer: Ultrasound Screening, is examining many
aspects of breast ultrasound interpretation in whole-breast screening of
high-risk women that may eventually lead to more uniform practice. Early results
from this large study reveal that breast screening with sonography has
considerable merit but there are many practical impediments to its widespread
use. These include, among others, the great length of time needed to perform and
interpret the study, the need for a radiologist to be involved in the
performance of the exam, the large number of images that must be reviewed, and
the need for highly expert operators.
Work to improve the accuracy of diagnostic breast
ultrasound has led to the development by American
College of Radiologists of a well-defined system for
breast lesion Assessment based on
parameters describing the ultrasound appearance of
breast lesions (known as BI-RADS®). BI-RADS®
outlines six different possible mammogram results,
ranging from a Category 0 to a Category 6.
Overall Assessment is assigned by interpreting
radiologist based on the number of
benign and
malignant criteria found.
Almen Laboratories has developed and
marketed Image Companion® - a sophisticated general application image-processing
software system for imaging applications that provides extensive tools to
identify objects and image features of interest, analyze the information content
and then store, retrieve and compare different objects in images of
interest based on this information. During the last 12 years, in collaboration
with the University of California, San Diego (UCSD) School of Medicine and the
Veterans Affairs Medical Center (VAMC) San Diego, parts of this software system
were tailored to the needs of diagnostic breast ultrasound. The resulting breast
assessment assisting software was created and validated under IRB-approved
cohorts of cases with available confirmation of the “truth” via needle biopsy or
12-24 months follow-up for benign lesions. During 2010-2012 FDA PMA review the
developed and validated software was approved for diagnostic used by a
BI-RADS-trained physician.
Breast Companion®
(“BC”, “BC CADx”) is identical to BI-RADS Companion™ (K072258) with an added
optional scoring function (Computer-aided Lesion Assessment, or “CLA”, or “BC
CLA”). It is an adjunctive tool to be used in support of radiologists’ readings
and is intended to be used in the diagnostic breast work-up process. This is an
important distinction from computer-aided detection devices for screening. CLA
is intended to be used by radiologist as decision stratification tool.
The BC device includes a report-generating function that
is fully compliant with the American College of Radiology Breast Imaging
Reporting and Data System (ACR BI-RADS®). BC CLA is optionally used
by a Radiologist who is completing an ACR BI-RADS® assessment report
of a lesion that was detected or identified in a previous procedure using a
different modality such as screening mammography or physical examination, or
patient discovery during self-examination. This ACR BI-RADS® report
is a record of the sonographic clinical assessment and incorporates
recommendations by the radiologist for further action within the overall
diagnostic process.
The BC CLA function is designed to
work as an “upgrade” feature on already installed BI-RADS Companion™
CAD software (Class II, K072258). BI-RADS Companion™ may be purchased with the
BC CLA function activated – that configuration is called Breast Companion®.
Breast Companion®
functionality working as an enhancement of BI-RADS Companion™
computer-aided system can be summarized as follows:
When
invoked by the user, the BC CLA function receives calculated numeric imaging
parameters of the lesion previously defined (outlined) by the user and already
stored in computer memory
Based
on calculated numeric imaging parameters of the lesion in question, Breast
Companion®
compares and retrieves the 7 most similar lesions from the Reference Library of
cases with confirmed findings
BC CLA
function computes the CLA score based on confirmed pathology of the retrieved
most similar lesions and then Breast Companion®
displays the CLA score
