Technical challenges in generalizing calibration techniques for breast density measurements

Abstract

Purpose We are developing a calibration methodology for full‐field digital mammography ( FFDM ). Calibration compensates for image acquisition technique influences on the pixel representation, ideally producing improved inter‐image breast density estimates. This approach relies on establishing references with rigid breast tissue‐equivalent phantoms ( BTE s) and requires an accurate estimate of the compressed breast thickness because the system readout is nominal. There is also an attenuation mismatch between adipose breast tissue and the adipose BTE that was noted in our previous work. It is referred to as the “attenuation anomaly” and addressed in this report. The objectives are to evaluate methods to correct for the compressed breast thickness and compensate for the attenuation anomaly. Methods Thickness correction surfaces were established with a deformable phantom ( DP ) using both image and physical measurements for three direct x‐ray conversion FFDM units. The Cumulative Sum serial quality control procedure was established to ensure the thickness correction measurements were stable over time by imaging and calibrating DP s biweekly in lieu of physical measurements. The attenuation anomaly was addressed by evaluating adipose image regions coupled with an optimization technique to adjust the adipose calibration data. We compared calibration consistency across matched left and right cranial caudal ( CC ) mammographic views (n = 199) with and without corrections using Bland–Altman plots. These plots were complemented by comparing the right and left breast calibrated average (μ a ) and population distribution mean (m a ) with 95% confidence intervals and difference distribution variances with the F‐test for uncorrected and corrected data. Results Thickness correction surfaces were well approximated as tilted planes and were dependent upon compression force. A correction was developed for the attenuation anomaly. All paddles (large and small paddles for all units) exhibited similar tilt as a function of force. Without correction, m a = 0.92 (−1.77, 3.62) was not significantly different from zero with many negative μ a samples. The thickness correction produced a significant shift in the μ a distribution in the positive direction with m a = 13.99 (11.17, 16.80) and reduced the difference distribution variance significantly ( P < 0.0001). Applying both corrections in tandem gave m a = 22.83 (20.32, 25.34), representing another significant positive shift in comparison with the thickness correction in isolation. Thickness correctio...

Artificial Intelligence in Breast Cancer Screening and Diagnosis

“…Proper segmentation using AI greatly improves the prognosis of the patient [30]. Breast density assessment is carried out using two-dimensional mammograms [31]. Breast cancer risk assessment is based on assessing risk factors such as age, family history, reproductive factors (menarche, menopause, age during the first pregnancy, parity, etc.…”

Artificial Intelligence in Medical Imaging of the Breast

“…Breast density is a strong risk factor for breast cancer and is usually evaluated by two-dimensional (2D) mammograms. Women with higher breast density have a two-to six-fold higher risk of developing breast cancer than women with low breast density (36). Mammographic density has traditionally been assessed as the absolute or relative amount (as percentage of the total breast size) occupied by dense tissue, which appears on a mammographic images as white "cotton-like" patches (37).…”

Calibrated Breast Density Measurements

“…Although these units are from the same manufacturer, there are design differences that could influence the calibration. The compression paddles attach differently, which could influence the compressed breast thickness correction (28) and erosion process parameters. Moreover, for the same target/filter combination, calibration curves also differ across the H and D systems (see Supplemental Table).…”

“…The calibration technique used in this report is an outgrowth of earlier work by Kaufhold et al (21).We extended this calibration work initially with phantom studies using a specific indirect X-ray detection full-field digital mammography (FFDM) unit (22À25) that produced several calibrated breast density metrics predictive of breast cancer (18,19,26,27). Subsequently, we modified this calibration approach to operate on direct X-ray conversion FFDM units and evaluated it with phantom studies (26,28,29); this approach was extended to patient mammograms for risk prediction purposes in this report using a matched caseÀcontrol study.…”