Decision quality support in diagnostic breast ultrasound through artificial Intelligence

Lev Barinov, Ajit Jairaj, Lina Paster, William Hulbert, Richard Mammone, Christine Podilchuk

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Medical Ultrasonography is a valuable imaging technology for medical diagnostics and, more recently, as a screening alternative to mammography for women with dense breasts. However, ultrasound imaging within the contexts of both diagnostic and screening mammography suffers from inter-operator and intra-operator variability. Consequently, there is a broad distribution of performance profiles, even for radiologists of similar training. Typically, these profiles tend to err on the side of caution, preferring false positive errors to false negative errors. While this approach may lead to a higher Cancer Detection Rate (CDR), it also lowers the Positive Predictive Value (PPV3) of performed biopsies. A lower PPV3 translates to an increase in benign biopsies, the annual cost of which are estimated to be on the order of $1 - $3 billion USD (not including pathological workups). And, of course, there is the immeasurable cost of pain, worry, and suffering borne by women undergoing these potentially unnecessary procedures. In this paper, we evaluate the ability of the ClearView cCAD algorithms to increase overall performance and reduce the inter-operator variance on a set of imaged lesions. The cCAD system provides an automated assessment of some ACR BI-RADs criteria and calculates a preliminary BI-RADs assessment, given as BI-RADS categorical bucket (1-3) or (4-5). Through the evaluation of 1300 breast lesion images, 3 MQSA certified radiologists were asked to determine both a Likelihood of Malignancy (LoM) and a BI-RADs assessment, from which their ROC curve AUC as well as PPV3 could be calculated. The cCAD system was also evaluated, on the same set of lesions, by a similar set of metrics. From this analysis we have been able to show that the cCAD system outperforms radiologists at all operating points within the scope of this study design. Furthermore, we've shown that through simple fusion schemes we are able to increase performance beyond that of either the cCAD system or the radiologist alone by all typically tracked quality metrics, and significantly reduce inter-operator variance.

Original languageEnglish (US)
Title of host publication2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509067138
DOIs
StatePublished - Feb 7 2017
Event2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Philadelphia, United States
Duration: Dec 3 2016 → …

Publication series

Name2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings

Other

Other2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016
CountryUnited States
CityPhiladelphia
Period12/3/16 → …

Fingerprint

Mammography
Biopsy
Artificial intelligence
Screening
Ultrasonics
Ultrasonography
Imaging techniques
Costs
Fusion reactions

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Biomedical Engineering

Keywords

  • computer aided diagnosis
  • inter-operator variability
  • ultrasound imaging

Cite this

Barinov, L., Jairaj, A., Paster, L., Hulbert, W., Mammone, R., & Podilchuk, C. (2017). Decision quality support in diagnostic breast ultrasound through artificial Intelligence. In 2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings [7846873] (2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SPMB.2016.7846873
Barinov, Lev ; Jairaj, Ajit ; Paster, Lina ; Hulbert, William ; Mammone, Richard ; Podilchuk, Christine. / Decision quality support in diagnostic breast ultrasound through artificial Intelligence. 2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. (2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings).
@inproceedings{0108ff4d339041b0b59bdcd8ff11340f,
title = "Decision quality support in diagnostic breast ultrasound through artificial Intelligence",
abstract = "Medical Ultrasonography is a valuable imaging technology for medical diagnostics and, more recently, as a screening alternative to mammography for women with dense breasts. However, ultrasound imaging within the contexts of both diagnostic and screening mammography suffers from inter-operator and intra-operator variability. Consequently, there is a broad distribution of performance profiles, even for radiologists of similar training. Typically, these profiles tend to err on the side of caution, preferring false positive errors to false negative errors. While this approach may lead to a higher Cancer Detection Rate (CDR), it also lowers the Positive Predictive Value (PPV3) of performed biopsies. A lower PPV3 translates to an increase in benign biopsies, the annual cost of which are estimated to be on the order of $1 - $3 billion USD (not including pathological workups). And, of course, there is the immeasurable cost of pain, worry, and suffering borne by women undergoing these potentially unnecessary procedures. In this paper, we evaluate the ability of the ClearView cCAD algorithms to increase overall performance and reduce the inter-operator variance on a set of imaged lesions. The cCAD system provides an automated assessment of some ACR BI-RADs criteria and calculates a preliminary BI-RADs assessment, given as BI-RADS categorical bucket (1-3) or (4-5). Through the evaluation of 1300 breast lesion images, 3 MQSA certified radiologists were asked to determine both a Likelihood of Malignancy (LoM) and a BI-RADs assessment, from which their ROC curve AUC as well as PPV3 could be calculated. The cCAD system was also evaluated, on the same set of lesions, by a similar set of metrics. From this analysis we have been able to show that the cCAD system outperforms radiologists at all operating points within the scope of this study design. Furthermore, we've shown that through simple fusion schemes we are able to increase performance beyond that of either the cCAD system or the radiologist alone by all typically tracked quality metrics, and significantly reduce inter-operator variance.",
keywords = "computer aided diagnosis, inter-operator variability, ultrasound imaging",
author = "Lev Barinov and Ajit Jairaj and Lina Paster and William Hulbert and Richard Mammone and Christine Podilchuk",
year = "2017",
month = "2",
day = "7",
doi = "10.1109/SPMB.2016.7846873",
language = "English (US)",
series = "2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings",
address = "United States",

}

Barinov, L, Jairaj, A, Paster, L, Hulbert, W, Mammone, R & Podilchuk, C 2017, Decision quality support in diagnostic breast ultrasound through artificial Intelligence. in 2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings., 7846873, 2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings, Institute of Electrical and Electronics Engineers Inc., 2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016, Philadelphia, United States, 12/3/16. https://doi.org/10.1109/SPMB.2016.7846873

Decision quality support in diagnostic breast ultrasound through artificial Intelligence. / Barinov, Lev; Jairaj, Ajit; Paster, Lina; Hulbert, William; Mammone, Richard; Podilchuk, Christine.

2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. 7846873 (2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Decision quality support in diagnostic breast ultrasound through artificial Intelligence

AU - Barinov, Lev

AU - Jairaj, Ajit

AU - Paster, Lina

AU - Hulbert, William

AU - Mammone, Richard

AU - Podilchuk, Christine

PY - 2017/2/7

Y1 - 2017/2/7

N2 - Medical Ultrasonography is a valuable imaging technology for medical diagnostics and, more recently, as a screening alternative to mammography for women with dense breasts. However, ultrasound imaging within the contexts of both diagnostic and screening mammography suffers from inter-operator and intra-operator variability. Consequently, there is a broad distribution of performance profiles, even for radiologists of similar training. Typically, these profiles tend to err on the side of caution, preferring false positive errors to false negative errors. While this approach may lead to a higher Cancer Detection Rate (CDR), it also lowers the Positive Predictive Value (PPV3) of performed biopsies. A lower PPV3 translates to an increase in benign biopsies, the annual cost of which are estimated to be on the order of $1 - $3 billion USD (not including pathological workups). And, of course, there is the immeasurable cost of pain, worry, and suffering borne by women undergoing these potentially unnecessary procedures. In this paper, we evaluate the ability of the ClearView cCAD algorithms to increase overall performance and reduce the inter-operator variance on a set of imaged lesions. The cCAD system provides an automated assessment of some ACR BI-RADs criteria and calculates a preliminary BI-RADs assessment, given as BI-RADS categorical bucket (1-3) or (4-5). Through the evaluation of 1300 breast lesion images, 3 MQSA certified radiologists were asked to determine both a Likelihood of Malignancy (LoM) and a BI-RADs assessment, from which their ROC curve AUC as well as PPV3 could be calculated. The cCAD system was also evaluated, on the same set of lesions, by a similar set of metrics. From this analysis we have been able to show that the cCAD system outperforms radiologists at all operating points within the scope of this study design. Furthermore, we've shown that through simple fusion schemes we are able to increase performance beyond that of either the cCAD system or the radiologist alone by all typically tracked quality metrics, and significantly reduce inter-operator variance.

AB - Medical Ultrasonography is a valuable imaging technology for medical diagnostics and, more recently, as a screening alternative to mammography for women with dense breasts. However, ultrasound imaging within the contexts of both diagnostic and screening mammography suffers from inter-operator and intra-operator variability. Consequently, there is a broad distribution of performance profiles, even for radiologists of similar training. Typically, these profiles tend to err on the side of caution, preferring false positive errors to false negative errors. While this approach may lead to a higher Cancer Detection Rate (CDR), it also lowers the Positive Predictive Value (PPV3) of performed biopsies. A lower PPV3 translates to an increase in benign biopsies, the annual cost of which are estimated to be on the order of $1 - $3 billion USD (not including pathological workups). And, of course, there is the immeasurable cost of pain, worry, and suffering borne by women undergoing these potentially unnecessary procedures. In this paper, we evaluate the ability of the ClearView cCAD algorithms to increase overall performance and reduce the inter-operator variance on a set of imaged lesions. The cCAD system provides an automated assessment of some ACR BI-RADs criteria and calculates a preliminary BI-RADs assessment, given as BI-RADS categorical bucket (1-3) or (4-5). Through the evaluation of 1300 breast lesion images, 3 MQSA certified radiologists were asked to determine both a Likelihood of Malignancy (LoM) and a BI-RADs assessment, from which their ROC curve AUC as well as PPV3 could be calculated. The cCAD system was also evaluated, on the same set of lesions, by a similar set of metrics. From this analysis we have been able to show that the cCAD system outperforms radiologists at all operating points within the scope of this study design. Furthermore, we've shown that through simple fusion schemes we are able to increase performance beyond that of either the cCAD system or the radiologist alone by all typically tracked quality metrics, and significantly reduce inter-operator variance.

KW - computer aided diagnosis

KW - inter-operator variability

KW - ultrasound imaging

UR - http://www.scopus.com/inward/record.url?scp=85016076108&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85016076108&partnerID=8YFLogxK

U2 - 10.1109/SPMB.2016.7846873

DO - 10.1109/SPMB.2016.7846873

M3 - Conference contribution

AN - SCOPUS:85016076108

T3 - 2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings

BT - 2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Barinov L, Jairaj A, Paster L, Hulbert W, Mammone R, Podilchuk C. Decision quality support in diagnostic breast ultrasound through artificial Intelligence. In 2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. 7846873. (2016 IEEE Signal Processing in Medicine and Biology Symposium, SPMB 2016 - Proceedings). https://doi.org/10.1109/SPMB.2016.7846873