Correlating mammographic and pathologic findings in clinical decision support using natural language processing and data mining methods

Patel, Tejal; Puppala, Mamta; Ogunti, Richard; Ensor, Joe; He, Tiancheng; Shewale, Jitesh B.; Ankerst, Donna P.; Kaklamani, Virginia; Rodriguez, Angel Augusto; Wong, Stephen T.C.; Chang, Jenny C.

doi:10.1002/cncr.30245

Cited by 64 publications

(32 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…also found that HER2-enriched tumors were more likely to have heterogeneous and pleomorphic microcalcifications on mammogram. [ 12 17 ] Cen et al . also demonstrated that amorphous and heterogeneous coarse calcifications were associated with a higher incidence of LA subtype.…”

Section: Discussionmentioning

confidence: 99%

Predicting the molecular subtype of breast cancer based on mammography and ultrasound findings

Rashmi

Sannapareddy

Murthy

et al. 2018

Indian Journal of Radiology and Imaging

View full text Add to dashboard Cite

Aim:To determine the correlation between mammography and ultrasound features of breast cancer with molecular subtypes and to calculate the predictive value of these features.Materials and Method:This is a prospective study of consecutive patients with breast cancer presenting between January 2016 and July 2017, who underwent mammography and/or ultrasound of breast and excision of the breast mass. Patients with contralateral breast mass, metastases, h/o prior cancer treatment, and other malignancies were excluded. On mammography, the presence or absence of microcalcification was noted. On ultrasound examination size, margins, microcalcification, posterior acoustic features, vascularity, and axillary nodes were assessed. Margins were categorized into circumscribed and non-circumscribed. Posterior acoustic features were classified into four categories: shadowing, enhancement, mixed, and no changes. Vascularity was assessed based on Adler's index into grades 0, 1, 2, and 3. Grades 0 and 1 were considered low and 2 and 3 were high.Results:Tumors with non-circumscribed margins and posterior acoustic shadowing were likely to be luminal A or B subtype of breast cancer [odds ratio (OR) 5.78; 95% confidence interval (CI) 3.68–9.80; P < 0.0001]. Tumors with non-circumscribed margins, posterior acoustic shadowing, and high vascularity were more likely to be luminal B subtype (OR 2.88; 95% CI 2–4.14; P- <0.0001). Tumors with microcalcification and posterior mixed acoustic pattern were strongly associated to be HER2-positive (OR 5.48; 95% CI 3.06–9.80; P < 0.0001). Tumors with circumscribed margins and posterior acoustic enhancement were highly suggestive of triple-negative breast cancer (OR 7.06; 95% CI 4.64–10.73; P < 0.0001).Conclusion:Microcalcification detected on mammography and certain ultrasound features such as circumscribed or non-circumscribed margins, posterior acoustic features, and vascularity are strongly correlated in predicting the molecular subtypes of breast cancer, and thus may further expand the role of conventional breast imaging.

show abstract

Section: Discussionmentioning

confidence: 99%

Predicting the molecular subtype of breast cancer based on mammography and ultrasound findings

Rashmi

Sannapareddy

Murthy

et al. 2018

Indian Journal of Radiology and Imaging

View full text Add to dashboard Cite

show abstract

“…For example, Seo et al [19] found that calcifications were more frequent in tumors with HER2 overexpression than in those without it. Patel and coworkers [20] found that patients with tumors that overexpressed HER2 were more likely to have heterogeneous and pleomorphic calcifications. However, they did not measure the range, diameter, or density of the calcifications.…”

Section: Discussionmentioning

confidence: 99%

BI-RADS 3-5 microcalcifications can preoperatively predict breast cancer HER2 and Luminal a molecular subtype

Cen

Chen

et al. 2017

Oncotarget

View full text Add to dashboard Cite

PurposeTo investigate associations between breast cancer molecular subtype and the patterns of mammographically detected calcifications.ResultsIdentified were 93 (19.1%) Luminal A, 242 (49.9%) Luminal B, 108 (22.2%) HER2 and 42 (8.7%) basal subtypes. In univariate analysis, the clinicopathological parameters and BI-RADS 3–5 microcalcifications, which consisted 9 selected features was significantly associated with breast cancer molecular subtype (all P < 0.05). Among subtypes, multivariate analysis showed that calcification >2 cm in range (OR: 1.878, 95% CI: 1.150 to 3.067) and calcification > 0.5 mm in diameter (OR:2.206, 95% CI: 1.235 to 3.323) was independently predictive of HER2 subtype. The model showed good discrimination for predicting HER2 subtype, with a C-index of 0.704. In addition, multivariate analysis showed that calcification morphology (amorphour or coarse heterogenous calcifications OR: 2.847, 95% CI: 1.526 to 5.312) was independently predictive of Luminal A subtype. The model showed good discrimination for predicting Luminal A subtype, with a C-index of 0.74. And we demonstrated that amorphour or coarse heterogenous calcifications were associated with a higher incidence of Luminal A subtype than pleomorphic or fine linear or branching calcifications. There was no significant difference between breast cancer subtypes (Luminal B vs. other; Basal vs. other) and the patterns of mammographically detected calcifications.Materials and MethodsMammographic images of 485 female patients were included. The correlation between mammographic imaging features and breast cancer subtype was analyzed using Chi-square test, univariate and binary logistic regression analysis.ConclusionsThis study shows that BI-RADS 3–5 microcalcifications can be conveniently used to facilitate the preoperative prediction of HER2 and Luminal A molecular subtype in patients with infiltrating ductal carcinoma.

show abstract

“…Moreover, using NLP techniques for mining electronic health records is another trending research topic. Examples include using NLP to correlate mammographic reports with pathologic findings and identifying the local recurrence of breast cancer …”

Section: Implementations Of Nlp In Clinical Genetics and Breast Cancementioning

confidence: 99%

Natural language processing to facilitate breast cancer research and management

et al. 2019

View full text Add to dashboard Cite

The medical literature has been growing exponentially, and its size has become a barrier for physicians to locate and extract clinically useful information. As a promising solution, natural language processing (NLP), especially machine learning (ML)-based NLP is a technology that potentially provides a promising solution. ML-based NLP is based on training a computational algorithm with a large number of annotated examples to allow the computer to "learn" and "predict" the meaning of human language.Although NLP has been widely applied in industry and business, most physicians still are not aware of the huge potential of this technology in medicine, and the implementation of NLP in breast cancer research and management is fairly limited. With a real-world successful project of identifying penetrance papers for breast and other cancer susceptibility genes, this review illustrates how to train and evaluate an NLPbased medical abstract classifier, incorporate it into a semiautomatic meta-analysis procedure, and validate the effectiveness of this procedure. Other implementations of NLP technology in breast cancer research, such as parsing pathology reports and mining electronic healthcare records, are also discussed. We hope this review will help breast cancer physicians and researchers to recognize, understand, and apply this technology to meet their own clinical or research needs. K E Y W O R D Sbreast cancer, genetics, medical literature, natural language processing

show abstract

Correlating mammographic and pathologic findings in clinical decision support using natural language processing and data mining methods

Cited by 64 publications

References 32 publications

Predicting the molecular subtype of breast cancer based on mammography and ultrasound findings

Predicting the molecular subtype of breast cancer based on mammography and ultrasound findings

BI-RADS 3-5 microcalcifications can preoperatively predict breast cancer HER2 and Luminal a molecular subtype

Natural language processing to facilitate breast cancer research and management

Contact Info

Product

Resources

About