Evaluation of the diagnostic performance of infrared imaging of the breast: a preliminary study

Abstract

Background: The study was conducted to investigate the diagnostic performance of infrared (IR) imaging of the breast using an interpretive model derived from a scoring system.

Methods: The study was approved by the Institutional Review Board of our hospital. A total of 276 women (mean age = 50.8 years, SD 11.8) with suspicious findings on mammograms or ultrasound received IR imaging of the breast before excisional biopsy. The interpreting radiologists scored the lesions using a scoring system that combines five IR signs. The ROC (receiver operating characteristic) curve and AUC (area under the ROC curve) were analyzed by the univariate logistic regression model for each IR sign and an age-adjusted multivariate logistic regression model including 5 IR signs. The cut-off values and corresponding sensitivity, specificity, Youden's Index (Index = sensitivity+specificity-1), positive predictive value (PPV), negative predictive value (NPV) were estimated from the age-adjusted multivariate model. The most optimal cut-off value was determined by the one with highest Youden's Index.

Results: For the univariate model, the AUC of the ROC curve from five IR signs ranged from 0.557 to 0.701, and the AUC of the ROC from the age-adjusted multivariate model was 0.828. From the ROC derived from the multivariate model, the sensitivity of the most optimal cut-off value would be 72.4% with the corresponding specificity 76.6% (Youden's Index = 0.49), PPV 81.3% and NPV 66.4%.

Conclusions: We established an interpretive age-adjusted multivariate model for IR imaging of the breast. The cut-off values and the corresponding sensitivity and specificity can be inferred from the model in a subpopulation for diagnostic purpose.

Trial registration: NCT00166998.

Figure 1

A 76-year-old woman with left breast cancer. IR imaging reveals focal increased surface temperature (positive IR1 sign with dT = 1.5°C compared with the contralateral mirror image site; positive IR2 sign with dT = 2°C compared with the remaining breast tissue at the ipsilateral side), abnormal vascular pattern (IR3 signs including closed vascular pattern, and vascular completeness) (arrows) and asymmetric vascular pattern (IR5 sign), and subtle focal bulging with back heat (IR4 sign) in left lower breast (arrowheads). Surgical pathological finding revealed a 4 cm infiltrating ductal carcinoma.

Figure 2

The corresponding breast ultrasound of the patient in Figure 1 shows a lobular mass at left lower breast, measuring about 3.1 × 2.1 cm in diameter with heterogeneous echogenicity (arrows).

Figure 3

A 48-year-old woman with right breast cancer. IR imaging shows abnormally increased focal surface temperature (positive IR1 sign with dT = 1.2°C compared with the contralateral mirror image site; positive IR2 sign with dT = 2°C compared with the remaining breast tissue in the ipsilateral breast), abnormal vascular pattern (IR3 signs including bifurcated vascular pattern, transverse vascular pattern, vascular completeness) and an asymmetric vascular pattern (IR5) in the right upper breast (arrows). The ln(OD) value for this finding scored by the interpreting radiologist was: -5.463+0.0872(48)+0.3783(1)+1.9157(0)+0.1728(1)+0.1578(3)+1.0278(0)+1.0363(1) = 0.7834, which is higher than the most optimal cut-off point (0.30) we selected in Table 4, and this is test-positive based on this cut-off point. (The radiologist scored the IR1 scale for this lesion as 1, therefore, the IR1_A = 1 and IR1_B = 0).

Figure 4

The right magnified mammogram of the patient in Figure 3 shows segmental pleomorphic microcalcifications in right upper breast (arrows).

Figure 5

The ROC of an age-adjusted multivariate regression model. AUC = 0.828.