Exome Sequencing Identifies Carriers of the Autosomal Dominant Cancer Predisposition Disorders Beyond Current Practice Guideline Recommendations.

PURPOSE: The autosomal dominant cancer predisposition disorders hereditary breast and ovarian cancer (HBOC) and Lynch syndrome (LS) are genetic conditions for which early identification and intervention have a positive effect on the individual and public health. The goals of this study were to determine whether germline genetic screening using exome sequencing could be used to efficiently identify carriers of HBOC and LS. METHODS: Participants were recruited from three geographically and racially diverse sites in the United States (Rochester, MN; Phoenix, AZ; Jacksonville, FL). Participants underwent Exome+ sequencing (Helix Inc, San Mateo, CA) and return of results for specific genetic findings: HBOC (BRCA1 and BRCA1) and LS (MLH1, MSH2, MSH6, PMS2, and EPCAM). Chart review was performed to collect demographics and personal and family cancer history. RESULTS: To date, 44,306 participants have enrolled in Tapestry. Annotation and interpretation of all variants in genes for HBOC and LS resulted in the identification of 550 carriers (prevalence, 1.24%), which included 387 with HBOC (27.2% BRCA1, 42.8% BRCA2) and 163 with LS (12.3% MSH6, 8.8% PMS2, 4.5% MLH1, 3.8% MSH2, and 0.2% EPCAM). More than half of these participants (52.1%) were newly diagnosed carriers with HBOC and LS. In all, 39.2% of HBOC/LS carriers did not satisfy National Comprehensive Cancer Network (NCCN) criteria for genetic evaluation. NCCN criteria were less commonly met in underrepresented minority populations versus self-reported White race (51.5% v 37.5%, P = .028). CONCLUSION: Our results emphasize the need for wider utilization of germline genetic sequencing for enhanced screening and detection of individuals who have LS and HBOC cancer predisposition syndromes.

Impact of SUSAN Denoising and ComBat Harmonization on Machine Learning Model Performance for Malignant Brain Neoplasms.

BACKGROUND AND PURPOSE: Feature variability in radiomics studies due to technical and magnet strength parameters is well known and may be addressed through various pre-processing methods. However, very few studies have evaluated downstream impact of variable pre-processing on model classification performance in a multi-class setting. We sought to evaluate the impact of SUSAN denoising and ComBat harmonization on model classification performance. MATERIALS AND METHODS: A total of 493 cases (410 internal and 83 external dataset) of glioblastoma (GB), intracranial metastatic disease (IMD) and primary CNS lymphoma (PCNSL) underwent semi-automated 3D-segmentation post baseline image processing (BIP) consisting of resampling, realignment, co-registration, skull stripping and image normalization. Post BIP, two sets were generated, one with and another without SUSAN denoising (SD). Radiomics features were extracted from both datasets and batch corrected to produce four datasets: (a) BIP, (b) BIP with SD, (c) BIP with ComBat and (d) BIP with both SD and ComBat harmonization. Performance was then summarized for models using a combination of six feature selection techniques and six machine learning models across four mask-sequence combinations with features derived from one-three (multi-parametric) MRI sequences. RESULTS: Most top performing models on the external test set used BIP+SD derived features. Overall, use of SD and ComBat harmonization led to a slight but generally consistent improvement in model performance on the external test set. CONCLUSIONS: The use of image pre-processing steps such as SD and ComBat harmonization may be more useful in a multiinstitutional setting and improve model generalizability. Models derived from only T1-CE images showed comparable performance to models derived from multiparametric MRI.

Long-Term Ultrasound Twinkling Detectability and Safety of a Polymethyl Methacrylate Soft Tissue Marker Compared to Conventional Breast Biopsy Markers-A Preclinical Study in a Porcine Model.

OBJECTIVE: We have studied the use of polymethyl methacrylate (PMMA) as an alternative biopsy marker that is readily detectable with ultrasound Doppler twinkling in cases of in vitro, ex vivo, or limited duration in vivo settings. This study investigates the long-term safety and ultrasound Doppler twinkling detectability of a PMMA breast biopsy marker following local perturbations and different dwell times in a 6-mo animal experiment. METHODS: This study, which was approved by our Institutional Animal Care and Use Committee, involved three pigs and utilized various markers, including PMMA (Zimmer Biomet), 3D-printed, and Tumark Q markers. Markers were implanted at different times for each pig. Mesh material or ethanol was used to induce a local inflammatory reaction near certain markers. A semiquantitative twinkling score assessed twinkling for actionable localization during monthly ultrasounds. At the primary endpoint, ultrasound-guided localization of lymph nodes with detectable markers was performed. Following surgical resection of the localized nodes, histomorphometric analysis was conducted to evaluate for tissue ingrowth and the formation of a tissue rind around the markers. RESULTS: No adverse events occurred. Twinkling scores of all markers for all three pigs decreased gradually over time. The Q marker exhibited the highest mean twinkling score followed by the PMMA marker, PMMA with mesh, and Q with ethanol. The 3D-printed marker with mesh and PMMA with ethanol had the lowest scores. All wire-localized lymph nodes were successfully resected. Despite varying percentages of tissue rind around the markers and a significant reduction in overall twinkling (p < 0.001) over time, mean PMMA twinkling scores remained clinically actionable at 6 and 5 mo using a General Electric C1-6 probe and 9L-probe, respectively. CONCLUSIONS: In this porcine model, the PMMA marker demonstrates an acceptable safety profile. Clinically actionable twinkling aids PMMA marker detection even after 6 mo of dwell time in porcine lymph nodes. The Q marker maintained the greatest twinkling over time compared to all the other markers studied.

Quantitative Biomarkers Derived from a Novel, Contrast-Free Ultrasound, High-Definition Microvessel Imaging for Differentiating Choroidal Tumors.

Angiogenesis has an essential role in the de novo evolution of choroidal melanoma as well as choroidal nevus transformation into melanoma. Differentiating early-stage melanoma from nevus is of high clinical importance; thus, imaging techniques that provide objective information regarding tumor microvasculature structures could aid accurate early detection. Herein, we investigated the feasibility of quantitative high-definition microvessel imaging (qHDMI) for differentiation of choroidal tumors in humans. This new ultrasound-based technique encompasses a series of morphological filtering and vessel enhancement techniques, enabling the visualization of tumor microvessels as small as 150 microns and extracting vessel morphological features as new tumor biomarkers. Distributional differences between the malignant melanomas and benign nevi were tested on 37 patients with choroidal tumors using a non-parametric Wilcoxon rank-sum test, and statistical significance was declared for biomarkers with p-values < 0.05. The ocular oncology diagnosis was choroidal melanoma (malignant) in 21 and choroidal nevus (benign) in 15 patients. The mean thickness of benign and malignant masses was 1.70 ± 0.40 mm and 3.81 ± 2.63 mm, respectively. Six HDMI biomarkers, including number of vessel segments (p = 0.003), number of branch points (p = 0.003), vessel density (p = 0.03), maximum tortuosity (p = 0.001), microvessel fractal dimension (p = 0.002), and maximum diameter (p = 0.003) exhibited significant distributional differences between the two groups. Contrast-free HDMI provided noninvasive imaging and quantification of microvessels of choroidal tumors. The results of this pilot study indicate the potential use of qHDMI as a complementary tool for characterization of small ocular tumors and early detection of choroidal melanoma.

A Force-Matched Approach to Large-Strain Nonlinearity in Elasticity Imaging for Breast Lesion Characterization.

OBJECTIVE: Ultrasound elasticity imaging is a class of ultrasound techniques with applications that include the detection of malignancy in breast lesions. Although elasticity imaging traditionally assumes linear elasticity, the large strain elastic response of soft tissue is known to be nonlinear. This study evaluates the nonlinear response of breast lesions for the characterization of malignancy using force measurement and force-controlled compression during ultrasound imaging. METHODS: 54 patients were recruited for this study. A custom force-instrumented compression device was used to apply a controlled force during ultrasound imaging. Motion tracking derived strain was averaged over lesion or background ROIs and matched with compression force. The resulting force-matched strain was used for subsequent analysis and curve fitting. RESULTS: Greater median differences between malignant and benign lesions were observed at higher compressional forces (p-value < 0.05 for compressional forces of 2-6N). Of three candidate functions, a power law function produced the best fit to the force-matched strain. A statistically significant difference in the scaling parameter of the power function between malignant and benign lesions was observed (p-value = 0.025). CONCLUSIONS: We observed a greater separation in average lesion strain between malignant and benign lesions at large compression forces and demonstrated the characterization of this nonlinear effect using a power law model. Using this model, we were able to differentiate between malignant and benign breast lesions. SIGNIFICANCE: With further development, the proposed method to utilize the nonlinear elastic response of breast tissue has the potential for improving non-invasive lesion characterization for potential malignancy.

AI-based classification of three common malignant tumors in neuro-oncology: A multi-institutional comparison of machine learning and deep learning methods.

PURPOSE: To determine if machine learning (ML) or deep learning (DL) pipelines perform better in AI-based three-class classification of glioblastoma (GBM), intracranial metastatic disease (IMD) and primary CNS lymphoma (PCNSL). METHODOLOGY: Retrospective analysis included 502 cases for training (208 GBM, 67 PCNSL and 227 IMD), with external validation on 86 cases (27:27:32). Multiparametric MRI images (T1W, T2W, FLAIR, DWI and T1-CE) were co-registered, resampled, denoised and intensity normalized, followed by semiautomatic 3D segmentation of the enhancing tumor (ET) and peritumoral region (PTR). Model performance was assessed using several ML pipelines and 3D-convolutional neural networks (3D-CNN) using sequence specific masks, as well as combination of masks. All pipelines were trained and evaluated with 5-fold nested cross-validation on internal data followed by external validation using multi-class AUC. RESULTS: Two ML models achieved similar performance on test set, one using T2-ET and T2-PTR masks (AUC: 0.885, 95% CI: [0.816, 0.935] and another using T1-CE-ET and FLAIR-PTR mask (AUC: 0.878, CI: [0.804, 0.930]). The best performing DL models achieved an AUC of 0.854, (CI [0.774, 0.914]) on external data using T1-CE-ET and T2-PTR masks, followed by model derived from T1-CE-ET, ADC-ET and FLAIR-PTR masks (AUC: 0.851, CI [0.772, 0.909]). CONCLUSION: Both ML and DL derived pipelines achieved similar performance. T1-CE mask was used in three of the top four overall models. Additionally, all four models had some mask derived from PTR, either T2WI or FLAIR.

Non-invasive Measurement of the Viscoelasticity of the Optic Nerve and Sclera for Assessing Papilledema: A Pilot Clinical Study.

OBJECTIVE: The purpose of this study was to evaluate our novel ultrasound vibro-elastography (UVE) technique for assessing patients with papilledema by non-invasively measuring shear wave speed (SWS), elasticity and viscosity properties of the optic nerve and sclera. METHODS: Shear wave speeds were measured at three frequencies-100, 150 and 200 Hz-on the optic nerve and sclera tissues for assessing patients with papilledema resulting from idiopathic intracranial hypertension (IIH). The method was evaluated in six papilledema patients and six controls on two separate locations for each participant (i.e., optic nerve and posterior sclera). SWSs of the optic nerve and sclera were analyzed by using a 2-D speed map technique within a circular region of interest (ROI) (i.e., the diameter of the ROI was 1.5 mm × 3.0 mm at the optic nerve and sclera, respectively). Elasticity and viscosity were then analyzed using the wave speed dispersion over the three frequencies. RESULTS: We measured values of SWS at both locations, optic nerve and sclera, of the right eye and left eye at three different frequencies in IIH patients and controls. The SWS (mean ± standard deviation [m/s]) of the right eye was significantly higher at the sclera in IIH patients compared with controls (i.e., patients vs. controls: 5.91 ± 0.54 vs. 3.86 ± 0.56, p < 0.0001 at 100 Hz), but there was no significant difference at the optic nerve (i.e., patients vs. controls: 3.62 ± 0.39 vs. 3.36 ± 0.35, p = 0.1100 at 100Hz). We observed increased elasticity (kPa) in IIH patients, indicating there are significant differences in elasticity between patients and controls at the optic nerve and sclera (i.e., right eye [patients vs. controls]: 14.42 ± 6.59 vs. 6.5 ± 5.71, p = 0.0065 [optic nerve]; 33.04 ± 10.62 vs. 9.16 ± 7.15, p < 0.0001 [sclera]). Viscosity was also (Pa·s) higher in the sclera and optic nerve of the left eye (i.e., left eye [patient vs. control]: 8.89 ± 4.37 vs. 7.27 ± 5.01, p = 0.3790 (optic nerve); 16.05 ± 10.79 vs. 8.49 ± 6.09, p < 0.0194 [sclera]). CONCLUSION: This research illustrates the feasibility of using our UVE system to evaluate stiffness of different tissues in the eye non-invasively. It suggests that the viscoelasticity of the posterior sclera is higher than that of the optic nerve. We found that the posterior sclera is stiffer than the optic nerve in patients with papilledema resulting from IIH, making UVE a potential non-invasive technique for assessing papilledema.

Deep Learning in Different Ultrasound Methods for Breast Cancer, from Diagnosis to Prognosis: Current Trends, Challenges, and an Analysis.

Breast cancer is the second-leading cause of mortality among women around the world. Ultrasound (US) is one of the noninvasive imaging modalities used to diagnose breast lesions and monitor the prognosis of cancer patients. It has the highest sensitivity for diagnosing breast masses, but it shows increased false negativity due to its high operator dependency. Underserved areas do not have sufficient US expertise to diagnose breast lesions, resulting in delayed management of breast lesions. Deep learning neural networks may have the potential to facilitate early decision-making by physicians by rapidly yet accurately diagnosing and monitoring their prognosis. This article reviews the recent research trends on neural networks for breast mass ultrasound, including and beyond diagnosis. We discussed original research recently conducted to analyze which modes of ultrasound and which models have been used for which purposes, and where they show the best performance. Our analysis reveals that lesion classification showed the highest performance compared to those used for other purposes. We also found that fewer studies were performed for prognosis than diagnosis. We also discussed the limitations and future directions of ongoing research on neural networks for breast ultrasound.

Noninvasive prediction of axillary lymph node breast cancer metastasis using morphometric analysis of nodal tumor microvessels in a contrast-free ultrasound approach.

PURPOSE: Changes in microcirculation of axillary lymph nodes (ALNs) may indicate metastasis. Reliable noninvasive imaging technique to quantify such variations is lacking. We aim to develop and investigate a contrast-free ultrasound quantitative microvasculature imaging technique for detection of metastatic ALN in vivo. EXPERIMENTAL DESIGN: The proposed ultrasound-based technique, high-definition microvasculature imaging (HDMI) provides superb images of tumor microvasculature at sub-millimeter size scales and enables quantitative analysis of microvessels structures. We evaluated the new HDMI technique on 68 breast cancer patients with ultrasound-identified suspicious ipsilateral axillary lymph nodes recommended for fine needle aspiration biopsy (FNAB). HDMI was conducted before the FNAB and vessel morphological features were extracted, analyzed, and the results were correlated with the histopathology. RESULTS: Out of 15 evaluated quantitative HDMI biomarkers, 11 were significantly different in metastatic and reactive ALNs (10 with P << 0.01 and one with 0.01 < P < 0.05). We further showed that through analysis of these biomarkers, a predictive model trained on HDMI biomarkers combined with clinical information (i.e., age, node size, cortical thickness, and BI-RADS score) could identify metastatic lymph nodes with an area under the curve of 0.9 (95% CI [0.82,0.98]), sensitivity of 90%, and specificity of 88%. CONCLUSIONS: The promising results of our morphometric analysis of HDMI on ALNs offer a new means of detecting lymph node metastasis when used as a complementary imaging tool to conventional ultrasound. The fact that it does not require injection of contrast agents simplifies its use in routine clinical practice.

Morphometric analysis of tumor microvessels for detection of hepatocellular carcinoma using contrast-free ultrasound imaging: A feasibility study.

Introduction: A contrast-free ultrasound microvasculature imaging technique was evaluated in this study to determine whether extracting morphological features of the vascular networks in hepatic lesions can be beneficial in differentiating benign and malignant tumors (hepatocellular carcinoma (HCC) in particular). Methods: A total of 29 lesions from 22 patients were included in this work. A post-processing algorithm consisting of clutter filtering, denoising, and vessel enhancement steps was implemented on ultrasound data to visualize microvessel structures. These structures were then further characterized and quantified through additional image processing. A total of nine morphological metrics were examined to compare different groups of lesions. A two-sided Wilcoxon rank sum test was used for statistical analysis. Results: In the malignant versus benign comparison, six of the metrics manifested statistical significance. Comparing only HCC cases with the benign, only three of the metrics were significantly different. No statistically significant distinction was observed between different malignancies (HCC versus cholangiocarcinoma and metastatic adenocarcinoma) for any of the metrics. Discussion: Obtained results suggest that designing predictive models based on such morphological characteristics on a larger sample size may prove helpful in differentiating benign from malignant liver masses.

Variability in Lipid Levels and Risk for Cardiovascular Disease: An Electronic Health Record-Based Population Cohort Study.

Background Larger within-patient variability of lipid levels has been associated with increased risk of cardiovascular disease (CVD); however, measures of lipid variability require ≥3 measurements and are not currently used clinically. We investigated the feasibility of calculating lipid variability within a large electronic health record-based population cohort and assessed associations with incident CVD. Methods and Results We identified all individuals ≥40 years of age who resided in Olmsted County, MN, on January 1, 2006 (index date), without prior CVD, defined as myocardial infarction, coronary artery bypass graft surgery, percutaneous coronary intervention, or CVD death. Patients with ≥3 measurements of total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, or triglycerides during the 5 years before the index date were retained. Lipid variability was calculated using variability independent of the mean. Patients were followed through December 31, 2020 for incident CVD. We identified 19 652 individuals (mean age 61 years; 55% female), who were CVD-free and had variability independent of the mean calculated for at least 1 lipid type. After adjustment, those with highest total cholesterol variability had a 20% increased risk of CVD (Q5 versus Q1 hazard ratio, 1.20 [95% CI, 1.06-1.37]). Results were similar for low-density lipoprotein cholesterol and high-density lipoprotein cholesterol. Conclusions In a large electronic health record-based population cohort, high variability in total cholesterol, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol was associated with an increased risk of CVD, independent of traditional risk factors, suggesting it may be a possible risk marker and target for intervention. Lipid variability can be calculated in the electronic health record environment, but more research is needed to determine its clinical utility.

Quantitative Biomarkers Derived from a Novel Contrast-Free Ultrasound High-Definition Microvessel Imaging for Distinguishing Thyroid Nodules.

Low specificity in current ultrasound modalities for thyroid cancer detection necessitates the development of new imaging modalities for optimal characterization of thyroid nodules. Herein, the quantitative biomarkers of a new high-definition microvessel imaging (HDMI) were evaluated for discrimination of benign from malignant thyroid nodules. Without the help of contrast agents, this new ultrasound-based quantitative technique utilizes processing methods including clutter filtering, denoising, vessel enhancement filtering, morphological filtering, and vessel segmentation to resolve tumor microvessels at size scales of a few hundred microns and enables the extraction of vessel morphological features as new tumor biomarkers. We evaluated quantitative HDMI on 92 patients with 92 thyroid nodules identified in ultrasound. A total of 12 biomarkers derived from vessel morphological parameters were associated with pathology results. Using the Wilcoxon rank-sum test, six of the twelve biomarkers were significantly different in distribution between the malignant and benign nodules (all p < 0.01). A support vector machine (SVM)-based classification model was trained on these six biomarkers, and the receiver operating characteristic curve (ROC) showed an area under the curve (AUC) of 0.9005 (95% CI: [0.8279,0.9732]) with sensitivity, specificity, and accuracy of 0.7778, 0.9474, and 0.8929, respectively. When additional clinical data, namely TI-RADS, age, and nodule size were added to the features, model performance reached an AUC of 0.9044 (95% CI: [0.8331,0.9757]) with sensitivity, specificity, and accuracy of 0.8750, 0.8235, and 0.8400, respectively. Our findings suggest that tumor vessel morphological features may improve the characterization of thyroid nodules.

Pilot study to determine whether reduced-dose photon-counting detector chest computed tomography can reliably display Brody II score imaging findings for children with cystic fibrosis at radiation doses that approximate radiographs.

BACKGROUND: The Brody II score uses chest CT to guide therapeutic changes in children with cystic fibrosis; however, patients and providers are often reticent to undergo chest CT given concerns about radiation. OBJECTIVE: We sought to determine the ability of a reduced-dose photon-counting detector (PCD) chest CT protocol to reproducibly display pulmonary disease severity using the Brody II score for children with cystic fibrosis (CF) scanned at radiation doses similar to those of a chest radiograph. MATERIALS AND METHODS: Pediatric patients with CF underwent non-contrast reduced-dose chest PCD-CT. Volumetric inspiratory and expiratory scans were obtained without sedation or anesthesia. Three pediatric radiologists with Certificates of Added Qualification scored each scan on an ordinal scale and assigned a Brody II score to grade bronchiectasis, peribronchial thickening, parenchymal opacity, air trapping and mucus plugging. We report image-quality metrics using descriptive statistics. To calculate inter-rater agreement for Brody II scoring, we used the Krippendorff alpha and intraclass correlation coefficient (ICC). RESULTS: Fifteen children with CF underwent reduced-dose PCD chest CT in both inspiration and expiration (mean age 8.9 years, range, 2.5-17.5 years; 4 girls). Mean volumetric CT dose index (CTDIvol) was 0.07 ± 0.03 mGy per scan. Mean effective dose was 0.12 ± 0.04 mSv for the total examination. All three readers graded spatial resolution and noise as interpretable on lung windows. The average Brody II score was 12.5 (range 4-19), with moderate inter-reader reliability (ICC of 0.61 [95% CI=0.27, 0.84]). Inter-rater reliability was moderate to substantial for bronchiectasis (0.52), peribronchial thickening (0.55), presence of opacity (0.62) and air trapping (0.70) and poor for mucus plugging (0.09). CONCLUSION: Reduced-dose PCD-CT permits diagnostic image quality and reproducible identification of Brody II scoring imaging findings at radiation doses similar to those for chest radiography.

Factors Associated With Ultrasound Color Doppler Twinkling by Breast Biopsy Markers: In Vitro and Ex Vivo Evaluation of 35 Commercially Available Markers.

BACKGROUND. Targeted axillary lymph node dissection after neoadjuvant systemic therapy (NST) for breast cancer depends on identifying marked metastatic lymph nodes. However, ultrasound visualization of biopsy markers is challenging. OBJECTIVE. The purpose of our study was to identify biopsy markers that show actionable twinkling in cadaveric breast and to assess the association of actionable twinkling with markers' surface roughness. METHODS. Commercial breast biopsy markers were evaluated for twinkling artifact in various experimental conditions relating to scanning medium (solid gel phantom, ultrasound coupling gel, cadaveric breast), transducer (ML6-15, 9L, C1-6), and embedding material (present vs absent). Markers were assigned twinkling scores from 0 (confident in no twinkling) to 4 (confident in exuberant twinkling); a score of 3 or greater represented actionable twinkling (sufficient confidence to rely solely on twinkling for target localization). Markers were hierarchically advanced to evaluation with increasingly complex media if showing at least minimal twinkling for a given medium. A 3D coherence optical profiler measured marker surface roughness. Mixed-effects proportional odds regression models assessed associations between twinkling scores and transducer and embedding material; Wilcoxon rank sum test evaluated associations between actionable twinkling and surface roughness. RESULTS. Thirty-five markers (21 with embedding material) were evaluated. Ten markers without embedding material advanced to evaluation in cadaveric breast. Higher twinkling scores were associated with presence of embedding material (odds ratio [OR] = 5.05 in solid gel phantom, 9.84 in coupling gel) and transducer (using the C1-6 transducer as reference; 9L transducer: OR = 0.36, 0.83, and 0.04 in solid gel phantom, ultrasound coupling gel, and cadaveric breast; ML6-15 transducer: OR = 0.07, 0.18, and 0.00 respectively; post hoc p between 9L and ML6-15: p < .001, p = .02, and p = .04). In cadaveric breast, three markers (Cork, Professional Q, MRI [Flex]) exhibited actionable twinkling for two or more transducers; surface roughness was significantly higher for markers with than without actionable twinkling for C1-6 (median values: 0.97 vs 0.35, p = .02) and 9L (1.75 vs 0.36; p = .002) transducers. CONCLUSION. Certain breast biopsy markers exhibited actionable twinkling in cadaveric breast. Twinkling was observed with greater confidence for the C1-6 and 9L transducers than the ML6-15 transducer. Actionable twinkling was associated with higher marker surface roughness. CLINICAL IMPACT. Use of twinkling for marker detection could impact preoperative or intraoperative localization after NST.

Photon-counting Detector CT with Deep Learning Noise Reduction to Detect Multiple Myeloma.

Background Photon-counting detector (PCD) CT and deep learning noise reduction may improve spatial resolution at lower radiation doses compared with energy-integrating detector (EID) CT. Purpose To demonstrate the diagnostic impact of improved spatial resolution in whole-body low-dose CT scans for viewing multiple myeloma by using PCD CT with deep learning denoising compared with conventional EID CT. Materials and Methods Between April and July 2021, adult participants who underwent a whole-body EID CT scan were prospectively enrolled and scanned with a PCD CT system in ultra-high-resolution mode at matched radiation dose (8 mSv for an average adult) at an academic medical center. EID CT and PCD CT images were reconstructed with Br44 and Br64 kernels at 2-mm section thickness. PCD CT images were also reconstructed with Br44 and Br76 kernels at 0.6-mm section thickness. The thinner PCD CT images were denoised by using a convolutional neural network. Image quality was objectively quantified in two phantoms and a randomly selected subset of participants (10 participants; median age, 63.5 years; five men). Two radiologists scored PCD CT images relative to EID CT by using a five-point Likert scale to detect findings reflecting multiple myeloma. The scoring for the matched reconstruction series was blinded to scanner type. Reader-averaged scores were tested with the null hypothesis of equivalent visualization between EID and PCD. Results Twenty-seven participants (median age, 68 years; IQR, 61-72 years; 16 men) were included. The blinded assessment of 2-mm images demonstrated improvement in viewing lytic lesions, intramedullary lesions, fatty metamorphosis, and pathologic fractures for PCD CT versus EID CT (P < .05 for all comparisons). The 0.6-mm PCD CT images with convolutional neural network denoising also demonstrated improvement in viewing all four pathologic abnormalities and detected one or more lytic lesions in 21 of 27 participants compared with the 2-mm EID CT images (P < .001). Conclusion Ultra-high-resolution photon-counting detector CT improved the visibility of multiple myeloma lesions relative to energy-integrating detector CT. © RSNA, 2022 Online supplemental material is available for this article.

Rethinking blood eosinophil counts: Epidemiology, associated chronic diseases, and increased risks of cardiovascular disease.

Background: The distribution and determinants of blood eosinophil counts in the general population are unclear. Furthermore, whether elevated blood eosinophil counts increase risk for cardiovascular disease (CVD) and other chronic diseases, other than atopic conditions, remains uncertain. Objective: We sought to describe the distribution of eosinophil counts in the general population and determine the association of eosinophil count with prevalent chronic disease and incident CVD. Methods: A population-based adult cohort was followed from January 1, 2006, to December 31, 2020. Electronic health record data regarding demographic characteristics, prevalent clinical characteristics, and incident CVD were extracted. Associations between blood eosinophil counts and demographic characteristics, chronic diseases, laboratory values, and risks of incident CVD were assessed using chi-square test, ANOVA, and Cox proportional hazards regression. Results: Blood eosinophil counts increased with age, body mass index, and reported smoking and tobacco use. The prevalence of chronic obstructive pulmonary disease, hypertension, cardiac arrhythmias, hyperlipidemia, diabetes mellitus, chronic kidney disease, and cancer increased as eosinophil counts increased. Eosinophil counts were significantly associated with coronary heart disease (hazard ratio [HR], 1.44; 95% CI, 1.12-1.84) and heart failure (HR, 1.62; 95% CI, 1.30-2.01) in fully adjusted models and with stroke/transient ischemic attack (HR, 1.37; 95% CI, 1.16-1.61) and CVD death (HR, 1.49; 95% CI, 1.10-2.00) in a model adjusting for age, sex, race, and ethnicity. Conclusions: Blood eosinophil counts differ by demographic and clinical characteristics as well as by prevalent chronic disease. Moreover, elevated eosinophil counts are associated with risk of CVD. Further prospective investigations are needed to determine the utility of eosinophil counts as a biomarker for CVD risk.

Volumetric imaging and morphometric analysis of breast tumor angiogenesis using a new contrast-free ultrasound technique: a feasibility study.

BACKGROUND: There is a strong correlation between the morphological features of new tumor vessels and malignancy. However, angiogenic heterogeneity necessitates 3D microvascular data of tumor microvessels for more reliable quantification. To provide more accurate information regarding vessel morphological features and improve breast lesion characterization, we introduced a quantitative 3D high-definition microvasculature imaging (q3D-HDMI) as a new easily applicable and robust tool to morphologically characterize microvasculature networks in breast tumors using a contrast-free ultrasound-based imaging approach. METHODS: In this prospective study, from January 2020 through December 2021, a newly developed q3D-HDMI technique was evaluated on participants with ultrasound-identified suspicious breast lesions recommended for core needle biopsy. The morphological features of breast tumor microvessels were extracted from the q3D-HDMI. Leave-one-out cross-validation (LOOCV) was applied to test the combined diagnostic performance of multiple morphological parameters of breast tumor microvessels. Receiver operating characteristic (ROC) curves were used to evaluate the prediction performance of the generated pooled model. RESULTS: Ninety-three participants (mean age 52 ± 17 years, 91 women) with 93 breast lesions were studied. The area under the ROC curve (AUC) generated with q3D-HDMI was 95.8% (95% CI 0.901-1.000), yielding a sensitivity of 91.7% and a specificity of 98.2%, that was significantly higher than the AUC generated with the q2D-HDMI (p = 0.02). When compared to q2D-HDMI, the tumor microvessel morphological parameters obtained from q3D-HDMI provides distinctive information that increases accuracy in differentiating breast tumors. CONCLUSIONS: The proposed quantitative volumetric imaging technique augments conventional breast ultrasound evaluation by increasing specificity in differentiating malignant from benign breast masses.

Ultrasonographic Detection and Surgical Retrieval of a Nonmetallic Twinkle Marker in Breast Cancer: Pilot Study.

Purpose To evaluate the short-term safety of a nonmetallic twinkle marker and compare its conspicuity at color Doppler US with that of standard breast biopsy clips and radioactive seeds by using B-mode US in axillary lymph nodes. Materials and Methods This prospective study (November 2020-July 2021) of participants with node-positive breast cancer who completed chemotherapy involved placing a twinkle marker at the time of preoperative radioactive seed localization. A five-point scoring system (1 = easiest, 5 = most difficult) was used to rate the ease of identifying the clip, seed, and twinkle marker on postlocalization sonograms, mammograms, specimen radiographs, and gross pathologic specimens. Descriptive statistics were used. Results Eight women (mean age, 57 years ± 16 [SD]) were enrolled. The median scores for US conspicuity of each device were 3.9 (range, 3.7-5.0) for the radioactive seed, 2.4 (range, 1.0-5.0) for the clip, and 2.0 (range, 1.0-4.3) for the twinkle marker. In six of eight participants, the twinkle marker was the most identifiable at US. The seeds, clips, and twinkle markers were scored "very easy" to identify on seven of eight postlocalization mammograms. The surgeon retrieved all eight twinkle markers 1-3 days after localization. In all 16 interpretations, the seeds, clips, and twinkle markers were rated as very easy to identify on specimen radiographs. The clip was the most difficult device to identify at pathologic examination in all participants, and the twinkle marker was the easiest to identify in seven of eight participants. Conclusion This pilot study demonstrates that the safety and ease of US detection of a twinkling tissue marker may be comparable to a biopsy clip. Keywords: Ultrasonography, US-Doppler, Breast, Localization, Surgery Clinical trial registration no. NCT04674852 © RSNA, 2022.

Magnetic resonance radiomic feature performance in pulmonary nodule classification and impact of segmentation variability on radiomics.

OBJECTIVE: Investigate the performance of multiparametric MRI radiomic features, alone or combined with current standard-of-care methods, for pulmonary nodule classification. Assess the impact of segmentation variability on feature reproducibility and reliability. METHODS: Radiomic features were extracted from 74 pulmonary nodules of 68 patients who underwent nodule resection or biopsy after MRI exam. The MRI features were compared with histopathology and conventional quantitative imaging values (maximum standardized uptake value [SUVmax] and mean Hounsfield unit [HU]) to determine whether MRI radiomic features can differentiate types of nodules and associate with SUVmax and HU using Wilcoxon rank sum test and linear regression. Diagnostic performance of features and four machine learning (ML) models were evaluated with area under the receiver operating characteristic curve (AUC) and 95% confidence intervals (CIs). Concordance correlation coefficient (CCC) assessed the segmentation variation impact on feature reproducibility and reliability. RESULTS: Elevn diffusion-weighted features distinguished malignant from benign nodules (adjusted p < 0.05, AUC: 0.73-0.81). No features differentiated cancer types. Sixty-seven multiparametric features associated with mean CT HU and 14 correlated with SUVmax. All significant MRI features outperformed traditional imaging parameters (SUVmax, mean HU, apparent diffusion coefficient [ADC], T1, T2, dynamic contrast-enhanced imaging values) in distinguishing malignant from benign nodules with some achieving statistical significance (p < 0.05). Adding ADC and smoking history improved feature performance. Machine learning models demonstrated strong performance in nodule classification, with extreme gradient boosting (XGBoost) having the highest discrimination (AUC = 0.83, CI=[0.727, 0.932]). We found good to excellent inter- and intrareader feature reproducibility and reliability (CCC≥0.80). CONCLUSION: Eleven MRI radiomic features differentiated malignant from benign lung nodules, outperforming traditional quantitative methods. MRI radiomic ML models demonstrated good nodule classification performances with XGBoost superior to three others. There was good to excellent inter- and intrareader feature reproducibility and reliability. ADVANCES IN KNOWLEDGE: Our study identified MRI radiomic features that successfully differentiated malignant from benign lung nodules and demonstrated high performance of our MR radiomic feature-based ML models for nodule classification. These new findings could help further establish thoracic MRI as a non-invasive and radiation-free alternative to standard practice for pulmonary nodule assessment.

Radiomics-based Machine-learning Models Can Detect Pancreatic Cancer on Prediagnostic Computed Tomography Scans at a Substantial Lead Time Before Clinical Diagnosis.

BACKGROUND & AIMS: Our purpose was to detect pancreatic ductal adenocarcinoma (PDAC) at the prediagnostic stage (3-36 months before clinical diagnosis) using radiomics-based machine-learning (ML) models, and to compare performance against radiologists in a case-control study. METHODS: Volumetric pancreas segmentation was performed on prediagnostic computed tomography scans (CTs) (median interval between CT and PDAC diagnosis: 398 days) of 155 patients and an age-matched cohort of 265 subjects with normal pancreas. A total of 88 first-order and gray-level radiomic features were extracted and 34 features were selected through the least absolute shrinkage and selection operator-based feature selection method. The dataset was randomly divided into training (292 CTs: 110 prediagnostic and 182 controls) and test subsets (128 CTs: 45 prediagnostic and 83 controls). Four ML classifiers, k-nearest neighbor (KNN), support vector machine (SVM), random forest (RM), and extreme gradient boosting (XGBoost), were evaluated. Specificity of model with highest accuracy was further validated on an independent internal dataset (n = 176) and the public National Institutes of Health dataset (n = 80). Two radiologists (R4 and R5) independently evaluated the pancreas on a 5-point diagnostic scale. RESULTS: Median (range) time between prediagnostic CTs of the test subset and PDAC diagnosis was 386 (97-1092) days. SVM had the highest sensitivity (mean; 95% confidence interval) (95.5; 85.5-100.0), specificity (90.3; 84.3-91.5), F1-score (89.5; 82.3-91.7), area under the curve (AUC) (0.98; 0.94-0.98), and accuracy (92.2%; 86.7-93.7) for classification of CTs into prediagnostic versus normal. All 3 other ML models, KNN, RF, and XGBoost, had comparable AUCs (0.95, 0.95, and 0.96, respectively). The high specificity of SVM was generalizable to both the independent internal (92.6%) and the National Institutes of Health dataset (96.2%). In contrast, interreader radiologist agreement was only fair (Cohen's kappa 0.3) and their mean AUC (0.66; 0.46-0.86) was lower than each of the 4 ML models (AUCs: 0.95-0.98) (P < .001). Radiologists also recorded false positive indirect findings of PDAC in control subjects (n = 83) (7% R4, 18% R5). CONCLUSIONS: Radiomics-based ML models can detect PDAC from normal pancreas when it is beyond human interrogation capability at a substantial lead time before clinical diagnosis. Prospective validation and integration of such models with complementary fluid-based biomarkers has the potential for PDAC detection at a stage when surgical cure is a possibility.