Citations

PURPOSE Background parenchymal uptake (BPU), which describes the level of radiotracer uptake in normal fibroglandular tissue on molecular breast imaging (MBI), has been identified as a breast cancer risk factor. Our objective was to develop and validate a deep learning model using image convolution to automatically categorize BPU on MBI. METHODS MBI examinations obtained for clinical and research purposes from 2004 to 2015 were reviewed to classify the BPU pattern using a standardized five-category scale. Two expert radiologists provided interpretations that were used as the reference standard for modeling. The modeling consisted of training and validating a convolutional neural network to predict BPU. Model performance was summarized in data reserved to test the performance of the algorithm at the per-image and per-breast levels. RESULTS Training was performed on 24,639 images from 3,133 unique patients. The model performance on the withheld testing data (6,172 images; 786 patients) was evaluated. Using direct matching on the predicted classification resulted in an accuracy of 69.4% (95% CI, 67.4% to 71.3%), and if prediction within one category was considered, accuracy increased to 96.0% (95% CI, 95.2% to 96.7%). When considering the breast-level prediction of BPU, the accuracy remained strong, with 70.3% (95% CI, 68.0% to 72.6%) and 96.2% (95% CI, 95.3% to 97.2%) for the direct match and allowance for one category, respectively. CONCLUSION BPU provided a robust target for training a convolutional neural network. A validated computer algorithm will allow for objective, reproducible encoding of BPU to foster its integration into risk-stratification algorithms.

BackgroundThere is no accurate method distinguishing different types of pulmonary nodules.PurposeTo investigate whether multiparametric 3T MRI biomarkers can distinguish malignant from benign pulmonary nodules, differentiate different types of neoplasms, and compare MRI‐derived measurements with values from commonly used noninvasive imaging modalities.Study TypeProspective.SubjectsSixty‐eight adults with pulmonary nodules undergoing resection.SequencesRespiratory triggered diffusion‐weighted imaging (DWI), periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) fat saturated T2‐weighted imaging, T1‐weighted 3D volumetric interpolated breath‐hold examination (VIBE) using CAIPIRINHA (controlled aliasing in parallel imaging results in a higher acceleration).Assessment/StatisticsApparent diffusion coefficient (ADC), T1, T2, T1 and T2 normalized to muscle (T1/M and T2/M), and dynamic contrast enhancement (DCE) values were compared with histology to determine whether they could distinguish malignant from benign nodules and discern primary from secondary malignancies using logistic regression. Predictability of primary neoplasm types was assessed using two‐sample t‐tests. MRI values were compared with positron emission tomography / computed tomography (PET/CT) to examine if they correlated with standardized uptake value (SUV) or CT Hounsfield unit (HU). Intra‐ and interreader agreements were assessed using intraclass correlations.ResultsForty‐nine of 74 nodules were malignant. There was a significant association between ADC and malignancy (odds ratio 4.47, P < 0.05). ADC ≥1.3 μm2/ms predicted malignancy. ADC, T1, and T2 together predicted malignancy (P = 0.003). No MRI parameter distinguished primary from metastatic neoplasms. T2 predicted PET positivity (P = 0.016). T2 and T1/M correlated with SUV (P < 0.05). Of 18 PET‐negative malignant nodules, 12 (67%) had an ADC ≥1.3 μm2/ms. With the exception of T2, all noncontrast MRI parameters distinguished adenocarcinomas from carcinoid tumors (P < 0.05). T1, T2, T1/M, and T2/M correlated with HU and therefore can predict nodule density. Combined with ADC, washout enhancement, arrival time (AT), peak enhancement intensity (PEI), Ktrans, Kep, Ve collectively were predictive of malignancy (P = 0.012). Combined washin, washout, time to peak (TTP), AT, and PEI values predicted malignancy (P = 0.043). There was good observer agreement for most noncontrast MRI biomarkers.Data ConclusionMRI can contribute to pulmonary nodule analysis. Multiparametric MRI might be better than individual MRI biomarkers in pulmonary nodule risk stratification.Level of Evidence: 1Technical Efficacy: Stage 2J. Magn. Reson. Imaging 2018.

Objectives: The purpose of this study is to compare diagnostic outcomes of digital breast tomosynthesis screen-detected masses worked up with mammography first with those evaluated with diagnostic ultrasound initially. Methods: All masses recalled from screening digital breast tomosynthesis between July 1, 2017 and December 31, 2017 that were sent either to diagnostic mammography or ultrasound were compared. Size, shape, margins, visibility on ultrasound, diagnostic assessment and pathology of all masses along with breast density were evaluated. Results: 102/212 digital breast tomosynthesis screen-detected masses were worked up with diagnostic mammography initially and 110/212 were worked up with ultrasound directly. There was no significant difference in ultrasound visibility of masses sent to diagnostic mammography first with those sent to ultrasound first (p = 0.42). 4 (4%) masses sent to mammogram first and 2 (2%) masses sent to ultrasound first were not visualized. There was a significant difference in size between masses that were visualized under ultrasound versus those that were not (p = 0.01), when masses in both groups were assessed cumulatively. Conclusions: 98% of digital breast tomosynthesis screen-detected masses sent to ultrasound directly were adequately assessed without diagnostic mammography. Advances in knowledge: There is potential for avoiding a diagnostic mammogram for evaluation of majority of digital breast tomosynthesis screen-detected masses.

AbstractSoluble ST2 (sST2) is a novel biomarker of inflammation and fibrosis. Elevated sST2 levels (≥35 ng/mL) are associated with worse outcomes in patients with heart failure (HF). There are sparse data regarding the significance of sST2 levels after heart transplantation (HTx). The study aims were to evaluate trends in soluble ST2 levels after the resolution of HF status with HTx and association between post‐HTx sST2 levels and outcomes. Plasma sST2 levels were measured at baseline (median [IQR] of 118 days pre‐HTx) and 12 months post‐HTx in 62 subjects who were stratified into two groups by post‐HTx sST2 levels < or ≥35 ng/mL: “Group 1” or “Group 2,” respectively. Plasma sST2 levels were elevated in 58% of patients pre‐HTx and in 50% of patients post‐HTx. There was no association between elevated sST2 levels before and after HTx, and no significant differences in baseline characteristics between Group 1 and Group 2 patients. Group 2 as compared to Group 1 HTx recipients had significantly higher incidence of antibody‐mediated rejection (AMR) for the entire post‐transplant follow‐up period (32% vs 4%, P = 0.006). There was no association between post‐HTx sST2 level status and other post‐HTx outcomes including survival. In conclusion, elevated plasma sST2 levels after HTx are associated with increased risk for AMR.

Objectives To (1) test whether patient attitudes toward intake forms at three Midwestern outpatient clinics are significantly more negative among those who are asked to complete SOGI questions versus those who are not; and (2) gain an in‐depth understanding of patient concerns about SOGI questions. Study Setting Data were collected between 6/29/2015 and 2/29/2016 from new patients ( N = 491) who presented at three outpatient clinics in a large academic medical center. This study was originally a quality improvement project, and later, institutional review board approval was obtained for secondary data analysis. Study Design Two‐stage mixed‐methods study. (1) Experimental: New patients at three sites were randomly assigned to complete either routine intake forms (control) or routine intake forms with SOGI questions (experimental); and (2) qualitative: interviews with patients who responded negatively to SOGI questions. Principal Findings There were no significant differences in patient attitudes between experimental and control groups ( p > .05). Of those who received SOGI questions, only 3 percent reported being distressed, upset, or offended by the SOGI questions. Conclusions Collection of SOGI data as a part of the routine clinical patient intake process is not distressing to 97 percent of patients who are heterosexual, cisgender, and older than 50 years.