Improving lesion volume measurements on digital mammograms.

Lesion volume is an important predictor for prognosis in breast cancer. However, it is currently impossible to compute lesion volumes accurately from digital mammography data, which is the most popular and readily available imaging modality for breast cancer. We make a step towards a more accurate lesion volume measurement on digital mammograms by developing a model that allows to estimate lesion volumes on processed mammogram. Processed mammograms are the images routinely used by radiologists in clinical practice as well as in breast cancer screening and are available in medical centers. Processed mammograms are obtained from raw mammograms, which are the X-ray data coming directly from the scanner, by applying certain vendor-specific non-linear transformations. At the core of our volume estimation method is a physics-based algorithm for measuring lesion volumes on raw mammograms. We subsequently extend this algorithm to processed mammograms via a deep learning image-to-image translation model that produces synthetic raw mammograms from processed mammograms in a multi-vendor setting. We assess the reliability and validity of our method using a dataset of 1778 mammograms with an annotated mass. Firstly, we investigate the correlations between lesion volumes computed from mediolateral oblique and craniocaudal views, with a resulting Pearson correlation of 0.93 [95% confidence interval (CI) 0.92 - 0.93]. Secondly, we compare the resulting lesion volumes from true and synthetic raw data, with a resulting Pearson correlation of 0.998 [95%CI 0.998 - 0.998] . Finally, for a subset of 100 mammograms with a malignant mass and concurrent MRI examination available, we analyze the agreement between lesion volume on mammography and MRI, resulting in an intraclass correlation coefficient of 0.81 [95%CI 0.73 - 0.87] for consistency and 0.78 [95%CI 0.66 - 0.86] for absolute agreement. In conclusion, we developed an algorithm to measure mammographic lesion volume that reached excellent reliability and good validity, when using MRI as ground truth. The algorithm may play a role in lesion characterization and breast cancer prognostication on mammograms.

The prognostic potential of mammographic growth rate of invasive breast cancer in the Nijmegen breast cancer screening cohort.

OBJECTIVES: Insight into the aggressiveness of potential breast cancers found in screening may optimize recall decisions. Specific growth rate (SGR), measured on mammograms, may provide valuable prognostic information. This study addresses the association of SGR with prognostic factors and overall survival in patients with invasive carcinoma of no special type (NST) from a screened population. METHODS: In this historic cohort study, 293 women with NST were identified from all participants in the Nijmegen screening program (2003-2007). Information on clinicopathological factors was retrieved from patient files and follow-up on vital status through municipalities. On consecutive mammograms, tumor volumes were estimated. After comparing five growth functions, SGR was calculated using the best-fitting function. Regression and multivariable survival analyses described associations between SGR and prognostic factors as well as overall survival. RESULTS: Each one standard deviation increase in SGR was associated with an increase in the Nottingham prognostic index by 0.34 [95% confidence interval (CI): 0.21-0.46]. Each one standard deviation increase in SGR increased the odds of a tumor with an unfavorable subtype (based on histologic grade and hormone receptors; odds ratio 2.14 [95% CI: 1.45-3.15]) and increased the odds of diagnosis as an interval cancer (versus screen-detected; odds ratio 1.57 [95% CI: 1.20-2.06]). After a median of 12.4 years of follow-up, 78 deaths occurred. SGR was not associated with overall survival (hazard ratio 1.12 [95% CI: 0.87-1.43]). CONCLUSIONS: SGR may indicate prognostically relevant differences in tumor aggressiveness if serial mammograms are available. A potential association with cause-specific survival could not be determined and is of interest for future research.

Deep learning-based breast region segmentation in raw and processed digital mammograms: generalization across views and vendors.

Purpose: We developed a segmentation method suited for both raw (for processing) and processed (for presentation) digital mammograms (DMs) that is designed to generalize across images acquired with systems from different vendors and across the two standard screening views. Approach: A U-Net was trained to segment mammograms into background, breast, and pectoral muscle. Eight different datasets, including two previously published public sets and six sets of DMs from as many different vendors, were used, totaling 322 screen film mammograms (SFMs) and 4251 DMs (2821 raw/processed pairs and 1430 only processed) from 1077 different women. Three experiments were done: first training on all SFM and processed images, second also including all raw images in training, and finally testing vendor generalization by leaving one dataset out at a time. Results: The model trained on SFM and processed mammograms achieved a good overall performance regardless of projection and vendor, with a mean (±std. dev.) dice score of 0.96±0.06 for all datasets combined. When raw images were included in training, the mean (±std. dev.) dice score for the raw images was 0.95±0.05 and for the processed images was 0.96±0.04. Testing on a dataset with processed DMs from a vendor that was excluded from training resulted in a difference in mean dice varying between -0.23 to +0.02 from that of the fully trained model. Conclusions: The proposed segmentation method yields accurate overall segmentation results for both raw and processed mammograms independent of view and vendor. The code and model weights are made available.

ANCA as a predictor of relapse: useful in patients with renal involvement but not in patients with nonrenal disease.

The value of measuring ANCA during follow-up to predict a relapse is controversial. On the basis of recently obtained pathophysiologic insights, we postulated that measuring ANCA is useful in patients with renal involvement but is less valuable in patients with nonrenal disease. One hundred sixty-six consecutive patients with ANCA-associated vasculitis, positive for either proteinase 3 (PR3)-ANCA or myeloperoxidase (MPO)-ANCA, were included in our study, followed at regular intervals, and tested for PR3-ANCA and MPO-ANCA. In this cohort, 104 patients had renal involvement (72 with PR3-ANCA, 32 with MPO-ANCA) and 62 patients had nonrenal disease (36 with PR3-ANCA, 26 with MPO-ANCA). During an average (±SD) follow-up of 49±33 months and 18±14 ANCA measurements, 89 ANCA rises and 74 relapses were recorded. ANCA rises correlated with relapses in patients who presented with renal involvement (hazard ratio [HR], 11.09; 95% confidence interval [95% CI], 5.01 to 24.55), but in comparison, associated only weakly with relapses in patients who presented with nonrenal disease (HR, 2.79; 95% CI, 1.30 to 5.98). In conclusion, longitudinal ANCA measurements may be useful in patients with renal involvement but is less valuable in patients with nonrenal disease.

Exercise in space: human skeletal muscle after 6 months aboard the International Space Station.

The aim of this investigation was to document the exercise program used by crewmembers (n = 9; 45 +/- 2 yr) while aboard the International Space Station (ISS) for 6 mo and examine its effectiveness for preserving calf muscle characteristics. Before and after spaceflight, we assessed calf muscle volume (MRI), static and dynamic calf muscle performance, and muscle fiber types (gastrocnemius and soleus). While on the ISS, crewmembers had access to a running treadmill, cycle ergometer, and resistance exercise device. The exercise regimen varied among the crewmembers with aerobic exercise performed approximately 5 h/wk at a moderate intensity and resistance exercise performed 3-6 days/wk incorporating multiple lower leg exercises. Calf muscle volume decreased (P < 0.05) 13 +/- 2% with greater (P < 0.05) atrophy of the soleus (-15 +/- 2%) compared with the gastrocnemius (-10 +/- 2%). Peak power was 32% lower (P < 0.05) after spaceflight. Force-velocity characteristics were reduced (P < 0.05) -20 to -29% across the velocity spectrum. There was a 12-17% shift in myosin heavy chain (MHC) phenotype of the gastrocnemius and soleus with a decrease (P < 0.05) in MHC I fibers and a redistribution among the faster phenotypes. These data show a reduction in calf muscle mass and performance along with a slow-to-fast fiber type transition in the gastrocnemius and soleus muscles, which are all qualities associated with unloading in humans. Future long-duration space missions should modify the current ISS exercise prescription and/or hardware to better preserve human skeletal muscle mass and function, thereby reducing the risk imposed to crewmembers.