JMI thanks everyone who served as a reviewer in 2017.

Breast cancer is the most common cancer among women worldwide and ranks second in terms of overall cancer deaths. One of the difficulties associated with treating breast cancer is that it is a heterogeneous disease with variations in benign and pathologic tissue composition, which contributes to disease development, progression, and treatment response. Many of these phenotypes are uncharacterized and their presence is difficult to detect, in part due to the sparsity of methods to correlate information between the cellular microscale and the whole-breast macroscale. Quantitative multiscale imaging of the breast is an emerging field concerned with the development of imaging technology that can characterize anatomic, functional, and molecular information across different resolutions and fields of view. It involves a diverse collection of imaging modalities, which touch large sections of the breast imaging research community. Prospective studies have shown promising results, but there are several challenges, ranging from basic physics and engineering to data processing and quantification, that must be met to bring the field to maturity. This paper presents some of the challenges that investigators face, reviews currently used multiscale imaging methods for preclinical imaging, and discusses the potential of these methods for clinical breast imaging.

Guidewires and catheters are used during minimally invasive interventional procedures to traverse in vascular system and access the desired position. Computer models are increasingly being used to predict the behavior of these instruments. This information can be used to choose the right instrument for each case and increase the success rate of the procedure. Moreover, a designer can test the performance of instruments before the manufacturing phase. A precise model of the instrument is also useful for a training simulator. Therefore, to identify the strengths and weaknesses of different approaches used to model guidewires and catheters, a literature review of the existing techniques has been performed. The literature search was carried out in Google Scholar and Web of Science and limited to English for the period 1960 to 2017. For a computer model to be used in practice, it should be sufficiently realistic and, for some applications, real time. Therefore, we compared different modeling techniques with regard to these requirements, and the purposes of these models are reviewed. Important factors that influence the interaction between the instruments and the vascular wall are discussed. Finally, different ways used to evaluate and validate the models are described. We classified the developed models based on their formulation into finite-element method (FEM), mass-spring model (MSM), and rigid multibody links. Despite its numerical stability, FEM requires a very high computational effort. On the other hand, MSM is faster but there is a risk of numerical instability. The rigid multibody links method has a simple structure and is easy to implement. However, as the length of the instrument is increased, the model becomes slower. For the level of realism of the simulation, friction and collision were incorporated as the most influential forces applied to the instrument during the propagation within a vascular system. To evaluate the accuracy, most of the studies compared the simulation results with the outcome of physical experiments on a variety of phantom models, and only a limited number of studies have done face validity. Although a subset of the validated models is considered to be sufficiently accurate for the specific task for which they were developed and, therefore, are already being used in practice, these models are still under an ongoing development for improvement. Realism and computation time are two important requirements in catheter and guidewire modeling; however, the reviewed studies made a trade-off depending on the purpose of their model. Moreover, due to the complexity of the interaction with the vascular system, some assumptions have been made regarding the properties of both instruments and vascular system. Some validation studies have been reported but without a consistent experimental methodology.

This guest editorial introduces the Special Section honoring Dr. Laurence P. Clarke.

The robustness of radiomic texture analysis across different manufacturers of mammography imaging systems is investigated. We quantified feature robustness across mammography manufacturers using a dataset of 111 women who underwent consecutive screening mammography on both general electric and Hologic systems. In each mammogram, a square region of interest (ROI) directly behind the nipple was manually selected. Radiomic features describing parenchymal patterns were automatically extracted on each ROI. Feature comparisons were conducted between manufacturers (and breast densities) using newly developed robustness metrics descriptive of correlation, equivalence, and variability. By examining the distribution of these metric values, we propose the following selection criteria to guide feature evaluation in this dataset: (1) 0.8<mean of feature ratios <1.2, (2) standard deviation of feature ratios <0.3, (3) correlation of features (ρF)<0.5, and (4) p<0.05. Statistically significant correlation coefficients ranged from 0.13 to 0.68 in comparisons between the two mammographic systems tested. Features describing spatial patterns tended to exhibit high correlation coefficients, while intensity- and directionality-based features had comparatively poor correlation. Our proposed robustness metrics may be used to evaluate other datasets, for which different ranges of metric values may be appropriate.

Diffusion weighted MRI has become ubiquitous in many areas of medicine, including cancer diagnosis and treatment response monitoring. Reproducibility of diffusion metrics is essential for their acceptance as quantitative biomarkers in these areas. We examined the variability in the apparent diffusion coefficient (ADC) obtained from both postprocessing software implementations utilized by the NCI Quantitative Imaging Network and online scan time-generated ADC maps. Phantom and in vivo breast studies were evaluated for two (ADC2) and four (ADC4) b-value diffusion metrics. Concordance of the majority of implementations was excellent for both phantom ADC measures and in vivo ADC2, with relative biases <0.1  %   (ADC2) and <0.5  %   (phantom ADC4) but with higher deviations in ADC at the lowest phantom ADC values. In vivo ADC4 concordance was good, with typical biases of ±2  %   to 3% but higher for online maps. Multiple b-value ADC implementations were separated into two groups determined by the fitting algorithm. Intergroup mean ADC differences ranged from negligible for phantom data to 2.8% for ADC4in vivo data. Some higher deviations were found for individual implementations and online parametric maps. Despite generally good concordance, implementation biases in ADC measures are sometimes significant and may be large enough to be of concern in multisite studies.

Multiparametric magnetic resonance imaging (MP-MRI), including diffusion-weighted imaging, is commonly used to diagnose prostate cancer. This radiology–pathology study correlates prostate cancer grade and morphology with common b-value combinations for calculating apparent diffusion coefficient (ADC). Thirty-nine patients undergoing radical prostatectomy were recruited for MP-MRI prior to surgery. Diffusion imaging was collected with seven b-values, and ADC was calculated. Excised prostates were sliced in the same orientation as the MRI using 3-D printed slicing jigs. Whole-mount slides were digitized and annotated by a pathologist. Annotated samples were aligned to the MRI, and ADC values were extracted from annotated peripheral zone (PZ) regions. A receiver operating characteristic (ROC) analysis was performed to determine accuracy of tissue type discrimination and optimal ADC b-value combination. ADC significantly discriminates Gleason (G) G4-5 cancer from G3 and other prostate tissue types. The optimal b-values for discriminating high from low-grade and noncancerous tissue in the PZ are 50 and 2000, followed closely by 100 to 2000 and 0 to 2000. Optimal ADC cut-offs are presented for dichotomized discrimination of tissue types according to each b-value combination. Selection of b-values affects the sensitivity and specificity of ADC for discrimination of prostate cancer.

Radiomic features characterize tumor imaging phenotype. Nonsmall cell lung cancer (NSCLC) tumors are known for their complexity in shape and wide range in density. We explored the effects of variable tumor contouring on the prediction of epidermal growth factor receptor (EGFR) mutation status by radiomics in NSCLC patients treated with a targeted therapy (Gefitinib). Forty-six early stage NSCLC patients (EGFR mutant:wildtype = 20:26) were included. Three experienced radiologists independently delineated the tumors using a semiautomated segmentation software on a noncontrast-enhanced baseline and three-week post-therapy CT scan images that were reconstructed using 1.25-mm slice thickness and lung kernel. Eighty-nine radiomic features were computed on both scans and their changes (radiomic delta-features) were calculated. The highest area under the curves (AUCs) were 0.87, 0.85, and 0.80 for the three radiologists and the number of significant features ( AUC&gt;0.8) was 3, 5, and 0, respectively. The AUCs of a single feature significantly varied among radiologists (e.g., 0.88, 0.75, and 0.73 for run-length primitive length uniformity). We conclude that a three-week change in tumor imaging phenotype allows identifying the EGFR mutational status of NSCLC. However, interobserver variability in tumor contouring translates into a significant variability in radiomic metrics accuracy.

This paper reports on results of a multisite collaborative project launched by the MRI subgroup of Quantitative Imaging Network to assess current capability and provide future guidelines for generating a standard parametric diffusion map Digital Imaging and Communication in Medicine (DICOM) in clinical trials that utilize quantitative diffusion-weighted imaging (DWI). Participating sites used a multivendor DWI DICOM dataset of a single phantom to generate parametric maps (PMs) of the apparent diffusion coefficient (ADC) based on two models. The results were evaluated for numerical consistency among models and true phantom ADC values, as well as for consistency of metadata with attributes required by the DICOM standards. This analysis identified missing metadata descriptive of the sources for detected numerical discrepancies among ADC models. Instead of the DICOM PM object, all sites stored ADC maps as DICOM MR objects, generally lacking designated attributes and coded terms for quantitative DWI modeling. Source-image reference, model parameters, ADC units and scale, deemed important for numerical consistency, were either missing or stored using nonstandard conventions. Guided by the identified limitations, the DICOM PM standard has been amended to include coded terms for the relevant diffusion models. Open-source software has been developed to support conversion of site-specific formats into the standard representation.

Multiple sclerosis (MS) is a neurological disease characterized by focal lesions and morphological changes in the brain captured on magnetic resonance (MR) images. However, extraction of the corresponding imaging markers requires accurate segmentation of normal-appearing brain structures (NABS) and the lesions in MR images. On MR images of healthy brains, the NABS can be accurately captured by MR intensity mixture models, which, in combination with regularization techniques, such as in Markov random field (MRF) models, are known to give reliable NABS segmentation. However, on MR images that also contain abnormalities such as MS lesions, obtaining an accurate and reliable estimate of NABS intensity models is a challenge. We propose a method for automated segmentation of normal-appearing and abnormal structures in brain MR images that is based on a locally adaptive NABS model, a robust model parameters estimation method, and an MRF-based segmentation framework. Experiments on multisequence brain MR images of 30 MS patients show that, compared to whole-brain MR intensity model and compared to four popular unsupervised lesion segmentation methods, the proposed method increases the accuracy of MS lesion segmentation.

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is sensitive but not specific to determining treatment response in early stage triple-negative breast cancer (TNBC) patients. We propose an efficient computerized technique for assessing treatment response, specifically the residual tumor (RT) status and pathological complete response (pCR), in response to neoadjuvant chemotherapy. The proposed approach is based on Riesz wavelet analysis of pharmacokinetic maps derived from noninvasive DCE-MRI scans, obtained before and after treatment. We compared the performance of Riesz features with the traditional gray level co-occurrence matrices and a comprehensive characterization of the lesion that includes a wide range of quantitative features (e.g., shape and boundary). We investigated a set of predictive models ( ∼96) incorporating distinct combinations of quantitative characterizations and statistical models at different time points of the treatment and some area under the receiver operating characteristic curve (AUC) values we reported are above 0.8. The most efficient models are based on first-order statistics and Riesz wavelets, which predicted RT with an AUC value of 0.85 and pCR with an AUC value of 0.83, improving results reported in a previous study by ∼13%. Our findings suggest that Riesz texture analysis of TNBC lesions can be considered a potential framework for optimizing TNBC patient care.

To create tumor “habitats” from the “signatures” discovered from multimodality metabolic and physiological images, we developed a framework of a processing pipeline. The processing pipeline consists of six major steps: (1) creating superpixels as a spatial unit in a tumor volume; (2) forming a data matrix {D} containing all multimodality image parameters at superpixels; (3) forming and clustering a covariance or correlation matrix {C} of the image parameters to discover major image “signatures;” (4) clustering the superpixels and organizing the parameter order of the {D} matrix according to the one found in step 3; (5) creating “habitats” in the image space from the superpixels associated with the “signatures;” and (6) pooling and clustering a matrix consisting of correlation coefficients of each pair of image parameters from all patients to discover subgroup patterns of the tumors. The pipeline was applied to a dataset of multimodality images in glioblastoma (GBM) first, which consisted of 10 image parameters. Three major image “signatures” were identified. The three major “habitats” plus their overlaps were created. To test generalizability of the processing pipeline, a second image dataset from GBM, acquired on the scanners different from the first one, was processed. Also, to demonstrate the clinical association of image-defined “signatures” and “habitats,” the patterns of recurrence of the patients were analyzed together with image parameters acquired prechemoradiation therapy. An association of the recurrence patterns with image-defined “signatures” and “habitats” was revealed. These image-defined “signatures” and “habitats” can be used to guide stereotactic tissue biopsy for genetic and mutation status analysis and to analyze for prediction of treatment outcomes, e.g., patterns of failure.

Blood flow-metabolism mismatch from dynamic positron emission tomography (PET) studies with O <mprescripts /> <none /> 15-labeled water ( H2O) and F <mprescripts /> <none /> 18-labeled fluorodeoxyglucose (FDG) has been shown to be a promising diagnostic for locally advanced breast cancer (LABCa) patients. The mismatch measurement involves kinetic analysis with the arterial blood time course (AIF) as an input function. We evaluate the use of a statistical method for AIF extraction (SAIF) in these studies. Fifty three LABCa patients had dynamic PET studies with H2O and FDG. For each PET study, two AIFs were recovered, an SAIF extraction and also a manual extraction based on a region of interest placed over the left ventricle (LV-ROI). Blood flow-metabolism mismatch was obtained with each AIF, and kinetic and prognostic reliability comparisons were made. Strong correlations were found between kinetic assessments produced by both AIFs. SAIF AIFs retained the full prognostic value, for pathologic response and overall survival, of LV-ROI AIFs.

This meta-analysis assesses the prognostic value of quantitative dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and diffusion-weighted MRI (DW-MRI) performed during neoadjuvant therapy (NAT) of locally advanced breast cancer. A systematic literature search was conducted to identify studies of quantitative DCE-MRI and DW-MRI performed during breast cancer NAT that report the sensitivity and specificity for predicting pathological complete response (pCR). Details of the study population and imaging parameters were extracted from each study for subsequent meta-analysis. Metaregression analysis, subgroup analysis, study heterogeneity, and publication bias were assessed. Across 10 studies that met the stringent inclusion criteria for this meta-analysis (out of 325 initially identified studies), we find that MRI had a pooled sensitivity of 0.91 [95% confidence interval (CI), 0.80 to 0.96] and specificity of 0.81(95% CI, 0.68 to 0.89) when adjusted for covariates. Quantitative DCE-MRI exhibits greater specificity for predicting pCR than semiquantitative DCE-MRI ( p&lt;0.001). Quantitative DCE-MRI and DW-MRI are able to predict, early in the course of NAT, the eventual response of breast tumors, with a high level of specificity and sensitivity. However, there is a high degree of heterogeneity in published studies highlighting the lack of standardization in the field.

Clinical trials that evaluate cancer treatments may benefit from positron emission tomography (PET) imaging, which for many cancers can discriminate between effective and ineffective treatments. However, the image metrics used to quantify disease and evaluate treatment may be biased by many factors related to clinical protocols and PET system settings, many of which are site- and/or manufacturer-specific. An observational study was conducted using two surveys that were designed to record key sources of bias and variability in PET imaging. These were distributed to hospitals across the United States. The first round of surveys was designed and distributed by the American College of Radiology’s Centers of Quantitative Imaging Excellence program in 2011. The second survey expanded on the first and was completed by the National Cancer Institute’s Quantitative Imaging Network. Sixty-three sites responded to the first survey and 36 to the second. Key imaging parameters varied across participating sites. The range of reported methods for image acquisition and reconstruction suggests that signal biases are not matched between sites. Patient preparation was also inconsistent, potentially contributing additional variability. For multicenter clinical trials, efforts to control biases through standardization of imaging procedures should precede patient measurements.

Large variability in computed tomography (CT) radiomics feature values due to CT imaging parameters can have subsequent implications on the prognostic or predictive significance of these features. Here, we investigated the impact of pitch, dose, and reconstruction kernel on CT radiomic features. Moreover, we introduced correction factors to reduce feature variability introduced by reconstruction kernels. The credence cartridge radiomics and American College of Radiology (ACR) phantoms were scanned on five different scanners. ACR phantom was used for 3-D noise power spectrum (NPS) measurements to quantify correlated noise. The coefficient of variation (COV) was used as the variability assessment metric. The variability in texture features due to different kernels was reduced by applying the NPS peak frequency and region of interest (ROI) maximum intensity as correction factors. Most texture features were dose independent but were strongly kernel dependent, which is demonstrated by a significant shift in NPS peak frequency among kernels. Percentage improvement in robustness was calculated for each feature from original and corrected %COV values. Percentage improvements in robustness of 19 features were in the range of 30% to 78% after corrections. We show that NPS peak frequency and ROI maximum intensity can be used as correction factors to reduce variability in CT texture feature values due to reconstruction kernels.

Although the role of cancer-activated stroma in malignant progression has been well investigated, the influence of an activated stroma in therapy response is not well understood. Using retrospective pilot cohorts, we previously observed that MRI detected stromal contrast enhancement was associated with proximity to the tumor and was predictive for relapse-free survival in patients with breast cancer receiving neoadjuvant chemotherapy. Here, to evaluate the association of stromal contrast enhancement to therapy, we applied an advanced tissue mapping technique to evaluate stromal enhancement patterns within 71 patients enrolled in the I-SPY 1 neoadjuvant breast cancer trial. We correlated MR stromal measurements with stromal protein levels involved in tumor progression processes. We found that stromal percent enhancement values decrease with distance from the tumor edge with the estimated mean change ranging −0.48 to −0.17 ( P≤0.001) for time points T2 through T4. While not statistically significant, we found a decreasing trend in global stromal signal enhancement ratio values with the use of chemotherapy. There were no statistically significant differences between MR enhancement measurements and stromal protein levels. Findings from this study indicate that stromal features characterized by MRI are impacted by chemotherapy and may have predictive value in a larger study.

Pathologic complete response following neoadjuvant therapy (NAT) is used as a short-term surrogate marker of eventual outcome in patients with breast cancer. Analyzing voxel-level heterogeneity in MRI-derived parametric maps, obtained before and after the first cycle of NAT (n=33), in conjunction with receptor status, may improve the predictive accuracy of tumor response to NAT. Toward that end, we incorporated two MRI-derived parameters, the apparent diffusion coefficient and efflux rate constant, with receptor status in a logistic ridge-regression model. The area under the curve (AUC) and Brier score of the model computed via 10-fold cross validation were 0.94 (95% CI: 0.85, 0.99) and 0.11 (95% CI: 0.06, 0.16), respectively. These two statistics strongly support the hypothesis that our proposed model outperforms the other models that we investigated (namely, models without either receptor information or voxel-level information). The contribution of the receptor information was manifested by an 8% to 15% increase in AUC and a 14% to 21% decrease in Brier score. These data indicate that combining multiparametric MRI with hormone receptor status has a high likelihood of improved prediction of pathologic response to NAT in breast cancer.

Positron emission tomography (PET) is a quantitative imaging modality, but the computation of standardized uptake values (SUVs) requires several instruments to be correctly calibrated. Variability in the calibration process may lead to unreliable quantitation. Sealed source kits containing traceable amounts of Ge68  /  Ga68 were used to measure signal stability for 19 PET scanners at nine hospitals in the National Cancer Institute’s Quantitative Imaging Network. Repeated measurements of the sources were performed on PET scanners and in dose calibrators. The measured scanner and dose calibrator signal biases were used to compute the bias in SUVs at multiple time points for each site over a 14-month period. Estimation of absolute SUV accuracy was confounded by bias from the solid phantoms’ physical properties. On average, the intrascanner coefficient of variation for SUV measurements was 3.5%. Over the entire length of the study, single-scanner SUV values varied over a range of 11%. Dose calibrator bias was not correlated with scanner bias. Calibration factors from the image metadata were nearly as variable as scanner signal, and were correlated with signal for many scanners. SUVs often showed low intrascanner variability between successive measurements but were also prone to shifts in apparent bias, possibly in part due to scanner recalibrations that are part of regular scanner quality control. Biases of key factors in the computation of SUVs were not correlated and their temporal variations did not cancel out of the computation. Long-lived sources and image metadata may provide a check on the recalibration process.

A clinical validation of the bone scan lesion area (BSLA) as a quantitative imaging biomarker was performed in metastatic castration-resistant prostate cancer (mCRPC). BSLA was computed from whole-body bone scintigraphy at baseline and week 12 posttreatment in a cohort of 198 mCRPC subjects (127 treated and 71 placebo) from a clinical trial involving a different drug from the initial biomarker development. BSLA computation involved automated image normalization, lesion segmentation, and summation of the total area of segmented lesions on bone scan AP and PA views as a measure of tumor burden. As a predictive biomarker, treated subjects with baseline BSLA &lt;200  cm2 had longer survival than those with higher BSLA ( HR=0.4 and p&lt;0.001). As a surrogate outcome biomarker, subjects were categorized as progressive disease (PD) if the BSLA increased by a prespecified 30% or more from baseline to week 12 and non-PD otherwise. Overall survival rates between PD and non-PD groups were statistically different ( HR=0.64 and p=0.007). Subjects without PD at week 12 had longer survival than subjects with PD: median 398 days versus 280 days. BSLA has now been demonstrated to be an early surrogate outcome for overall survival in different prostate cancer drug treatments.

The growth of multiparametric imaging protocols has paved the way for quantitative imaging phenotypes that predict treatment response and clinical outcome, reflect underlying cancer molecular characteristics and spatiotemporal heterogeneity, and can guide personalized treatment planning. This growth has underlined the need for efficient quantitative analytics to derive high-dimensional imaging signatures of diagnostic and predictive value in this emerging era of integrated precision diagnostics. This paper presents cancer imaging phenomics toolkit (CaPTk), a new and dynamically growing software platform for analysis of radiographic images of cancer, currently focusing on brain, breast, and lung cancer. CaPTk leverages the value of quantitative imaging analytics along with machine learning to derive phenotypic imaging signatures, based on two-level functionality. First, image analysis algorithms are used to extract comprehensive panels of diverse and complementary features, such as multiparametric intensity histogram distributions, texture, shape, kinetics, connectomics, and spatial patterns. At the second level, these quantitative imaging signatures are fed into multivariate machine learning models to produce diagnostic, prognostic, and predictive biomarkers. Results from clinical studies in three areas are shown: (i) computational neuro-oncology of brain gliomas for precision diagnostics, prediction of outcome, and treatment planning; (ii) prediction of treatment response for breast and lung cancer, and (iii) risk assessment for breast cancer.

Comparative preliminary analysis of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) data collected in the International Breast MR Consortium 6883 multicenter trial was performed to distinguish benign and malignant breast tumors. Prebiopsy DCE-MRI data from 45 patients with suspicious breast lesions were obtained. Semiquantitative mean signal-enhancement ratio ( SERmean) was calculated for all lesions, and quantitative pharmacokinetic, parameters Ktrans, kep, and ve, were calculated for the subset with available T1 maps ( n=35). Diagnostic performance was estimated for DCE-MRI parameters and compared to standard clinical MRI assessment. Quantitative and semiquantitative metrics discriminated benign and malignant lesions, with receiver operating characteristic area under the curve (AUC) values of 0.71, 0.70, and 0.82 for Ktrans, kep, and SERmean, respectively ( p&lt;0.05). At equal 94% sensitivity, the specificity and positive predictive value of SERmean (53% and 63%, respectively) and K^trans (42% and 58%) were higher than clinical MRI assessment (32% and 54%).

High-dimensional imaging features extracted from diagnostic imaging, called radiomics, are increasingly reported for diagnosis, prognosis, and response to therapy. Establishing the sensitivity of radiomic features to variation in scan protocols is necessary because acquisition and reconstruction parameters can vary widely across and within institutions. Our objective was to assess the reproducibility of radiomic features derived from computed tomography (CT) images by varying tube current (mA), noise index, and reconstruction [adaptive statistical iterative reconstruction (ASiR)], parameters increasingly varied by institutions seeking to reduce radiation dose in their patients. We extracted radiomic features from CT images of a uniform water phantom, anthropomorphic phantom, and a human scan. Scans were acquired from the phantoms with six tube currents (50, 100, 200, 300, 400, and 500 mA) and five noise index levels (12, 14, 16, 18, and 20), respectively. Scans of the phantoms and patient were reconstructed from 0% ASiR (i.e., filtered back projection) to 100% ASiR in increments of 10%. Two hundred and forty-eight well-known radiomic features were extracted from all scans. The concordance correlation coefficient was used to assess agreement of features. Our analysis suggests that image acquisition parameters (tube current, noise index) as well as the reconstruction technique strongly influence radiomic feature reproducibility and demonstrate a subset of reproducible features potentially usable in clinical practice.

Lung cancer has a high incidence and mortality rate. Early detection and diagnosis of lung cancers is best achieved with low-dose computed tomography (CT). Classical radiomics features extracted from lung CT images have been shown as able to predict cancer incidence and prognosis. With the advancement of deep learning and convolutional neural networks (CNNs), deep features can be identified to analyze lung CTs for prognosis prediction and diagnosis. Due to a limited number of available images in the medical field, the transfer learning concept can be helpful. Using subsets of participants from the National Lung Screening Trial (NLST), we utilized a transfer learning approach to differentiate lung cancer nodules versus positive controls. We experimented with three different pretrained CNNs for extracting deep features and used five different classifiers. Experiments were also conducted with deep features from different color channels of a pretrained CNN. Selected deep features were combined with radiomics features. A CNN was designed and trained. Combinations of features from pretrained, CNNs trained on NLST data, and classical radiomics were used to build classifiers. The best accuracy (76.79%) was obtained using feature combinations. An area under the receiver operating characteristic curve of 0.87 was obtained using a CNN trained on an augmented NLST data cohort.

The purpose of this study was to develop a dynamic physical cardiac phantom with a realistic coronary plaque to investigate stenosis measurement accuracy under clinically relevant heart-rates. The coronary plaque model (5 mm diameter, 50% stenosis, and 32 mm long) was designed and 3D-printed with tissue equivalent materials (calcified plaque with iodine-enhanced lumen). Realistic cardiac motion was modeled by converting computational cardiac motion vectors into compression and rotation profiles executed by a commercial base cardiac phantom. The phantom was imaged on a dual-source CT system applying a retrospective gated coronary CT angiography (CCTA) protocol using synthesized motion-synchronized electrocardiogram (ECG) waveforms. Multiplanar reformatted images were reconstructed along vessel centerlines. Enhanced lumens were segmented by five independent operators. On average, stenosis measurement accuracy was 0.9% positively biased for the motion-free condition. Average measurement accuracy monotonically decreased from 0.9% positive bias for the motion-free condition to 18.5% negative bias at 90 beats per minute. Contrast-to-noise ratio, lumen circularity, and segmentation conformity also decreased monotonically with increasing heart-rate. These results demonstrate successful implementation of a base cardiac phantom with a 3D-printed coronary plaque model, relevant motion profile, and coordinated ECG waveform. They further show the utility of the model to ascertain metrics of CCTA accuracy and image quality under realistic plaque, motion, and acquisition conditions.

Stationary digital breast tomosynthesis (sDBT) is an emerging technology in which the single rotating x-ray tube is replaced by a fixed array of multiple carbon nanotube-enabled sources, providing a higher spatial and temporal resolution. As such, sDBT offers a promising platform for contrast-enhanced (CE) imaging. However, given the minimal enhancement above background with standard operational tube settings and iodine dosing, CE breast imaging requires additional acquisition steps to isolate the iodine signal, using either temporal or dual energy subtraction (TS or DES) protocols. Also, correcting for factors that limit contrast is critical, and scatter and noise pose unique challenges during tomosynthesis. This phantom-based study of CE sDBT compared different postacquisition scatter correction approaches on the quality of the reconstructed image slices. Beam-pass collimation was used to sample scatter indirectly, from which an interpolated scatter map was obtained for each projection image. Scatter-corrected projections provided the information for reconstruction. Comparison between the application of different scatter maps demonstrated the significant effect that processing has on the contrast-to-noise ratio and feature detectability (d′) in the final displayed images and emphasized the critical importance of scatter correction during DES.

Two dosimetric quantities [mean glandular dose (MGD) and entrance surface air kerma (ESAK)] and the diagnostic performance of phase-contrast mammography with synchrotron radiation (MSR) are compared to conventional digital mammography (DM). Seventy-one patients (age range, 41 to 82 years) underwent MSR after a DM examination if questionable or suspicious breast abnormalities were not clarified by ultrasonography. The MGD and the ESAK delivered in both examinations were evaluated and compared. Two on-site radiologists rated the images in consensus according to the Breast Imaging Reporting and Data System assessment categories, which were then correlated with the final diagnoses by means of statistical generalized linear models (GLMs). Receiver operating characteristic curves were also used to assess the diagnostic performance by comparing the area under the curve (AUC). An important MGD and ESAK reduction was observed in MSR due to the monoenergetic beam. In particular, an average 43% reduction was observed for the MGD and a reduction of more than 50% for the ESAK. GLM showed higher diagnostic accuracy, especially in terms of specificity, for MSR, confirmed by AUC analysis (p<0.001). The study design implied that the population was characterized by a high prevalence of disease and that the radiologists, who read the DM images before referring the patient to MSR, could have been influenced in their assessments. Within these limitations, the use of synchrotron radiation with the phase-contrast technique applied to mammography showed an important dose reduction and a higher diagnostic accuracy compared with DM. These results could further encourage research on the translation of x-ray phase-contrast imaging into the clinics.

This work concerns the validation of the Kyoto-Kagaku thorax anthropomorphic phantom Lungman for use in chest radiography optimization. The equivalence in terms of polymethyl methacrylate (PMMA) was established for the lung and mediastinum regions of the phantom. Patient chest examination data acquired under automatic exposure control were collated over a 2-year period for a standard x-ray room. Parameters surveyed included exposure index, air kerma area product, and exposure time, which were compared with Lungman values. Finally, a voxel model was developed by segmenting computed tomography images of the phantom and implemented in PENELOPE/penEasy Monte Carlo code to compare phantom tissue-equivalent materials with materials from ICRP Publication 89 in terms of organ dose. PMMA equivalence varied depending on tube voltage, from 9.5 to 10.0 cm and from 13.5 to 13.7 cm, for the lungs and mediastinum regions, respectively. For the survey, close agreement was found between the phantom and the patients’ median values (deviations lay between 8% and 14%). Differences in lung doses, an important organ for optimization in chest radiography, were below 13% when comparing the use of phantom tissue-equivalent materials versus ICRP materials. The study confirms the value of the Lungman for chest optimization studies.

The H-scan analysis of ultrasound images is a matched-filter approach derived from analysis of scattering from incident pulses in the form of Gaussian-weighted Hermite polynomial functions. This framework is applied in a preliminary study of thyroid lesions to examine the H-scan outputs for three categories: normal thyroid, benign lesions, and cancerous lesions within a total group size of 46 patients. In addition, phantoms comprised of spherical scatterers are analyzed to establish independent reference values for comparison. The results demonstrate a small but significant difference in some measures of the H-scan channel outputs between the different groups.

The accuracy of Monte Carlo (MC) simulations in estimating the computed tomography radiation dose is highly dependent on the proprietary x-ray source information. To address this, this study develops a method to precisely estimate the x-ray spectrum and bowtie (BT) filter thickness of the x-ray source based on physical measurements and calculations. The static x-ray source of the CT localizer radiograph was assessed to measure the total filtration at the isocenter for the x-ray spectrum characterization and the BT profile (air-kerma values as a function of fan angle). With these values, the utilized BT filter in the localizer radiograph was assessed by integrating the measured air kerma in a full 360-deg cycle. The consistency observed between the integrated BT filter profiles and the directly measured profiles pointed to the similarity in the utilized BT filter in terms of thickness and material between the static and rotating x-ray geometries. Subsequently, the measured air kerma was used to calculate the BT filter thickness and was verified using MC simulations by comparing the calculated and measured air-kerma values, where a very good agreement was observed. This would allow a more accurate computed tomography simulation and facilitate the estimation of the dose delivered to the patients.

The detector pixel size can be a severe limitation in x-ray imaging of fine details in the human body. We demonstrate a method of using spectral x-ray measurements to image the spatial distribution of the linear attenuation coefficient on a length scale smaller than one pixel, based on the fact that interfaces parallel to the x-ray beam have a unique spectral response, which distinguishes them from homogeneous materials. We evaluate the method in a simulation study by simulating projection imaging of the border of an iodine insert with 200  mg  /  ml I in a soft tissue phantom. The results show that the projected iodine profile can be recovered with an RMS resolution of 5% to 34% of the pixel size, using an ideal energy-resolving detector. We also validate this method in an experimental study by imaging an iodine insert in a polyethylene phantom using a photon-counting silicon-strip detector. The results show that abrupt and gradual transitions can be distinguished based on the transmitted x-ray spectrum, in good agreement with simulations. The demonstrated method may potentially be used for improving visualization of blood vessel boundaries, e.g., in acute stroke care.

Fixed pattern noise due to nonuniform amplifier gains and scintillator sensitivity should be alleviated in radiography imaging to acquire low-noise x-ray images from detectors. Here, the noise property of the detector is usually evaluated observing the noise power spectrum (NPS). A gain-correction scheme, in which uniformly illuminated images are averaged to design a gain map, can be applied to alleviate the fixed pattern noise problem. The normalized NPS (NNPS) of the gain-corrected image decreases as the number of images for the average increases and converges to an infimum, which can be achieved if the fixed pattern noise is completely removed. If we know the NNPS infimum of the detector, then we can determine the performance of the gain-corrected images compared with the achievable lower bound. We first construct an image-formation model considering the nonuniform gain and then consider two measurement methods based on subtraction and division to estimate the NNPS infimum of the detector. In order to obtain a high-precision NNPS infimum estimate, we consider a time-averaging method. For several flat-panel radiography detectors, we constructed the NNPS infimum measurements and compared them with NNPS values of the gain-corrected images. We observed that the NNPS values of the gain-corrected images approached the NNPS infimum as the number of images for the average increased.

Treatment decision for coronary artery disease (CAD) is based on both morphological and functional information. Image fusion of coronary computed tomography angiography (CCTA) and three-dimensional echocardiography (3DE) could combine morphology and function into a single image to facilitate diagnosis. Three semiautomatic feature-based methods for CCTA/3DE registration were implemented and applied on CAD patients. Methods were verified and compared using landmarks manually identified by a cardiologist. All methods were found feasible for CCTA/3DE fusion.

We propose an automated segmentation method to detect, segment, and quantify hyperreflective foci (HFs) in three-dimensional (3-D) spectral domain optical coherence tomography (SD-OCT). The algorithm is divided into three stages: preprocessing, layer segmentation, and HF segmentation. In this paper, a supervised classifier (random forest) was used to produce the set of boundary probabilities in which an optimal graph search method was then applied to identify and produce the layer segmentation using the Sobel edge algorithm. An automated grow-cut algorithm was applied to segment the HFs. The proposed algorithm was tested on 20 3-D SD-OCT volumes from 20 patients diagnosed with proliferative diabetic retinopathy (PDR) and diabetic macular edema (DME). The average dice similarity coefficient and correlation coefficient ( r) are 62.30%, 96.90% for PDR, and 63.80%, 97.50% for DME, respectively. The proposed algorithm can provide clinicians with accurate quantitative information, such as the size and volume of the HFs. This can assist in clinical diagnosis, treatment, disease monitoring, and progression.

Segmentation of pulmonary lobes in inspiration and expiration chest CT scan pairs is an important prerequisite for lobe-based quantitative disease assessment. Conventional methods process each CT scan independently, resulting typically in lower segmentation performance at expiration compared to inspiration. To address this issue, we present an approach, which utilizes CT scans at both respiratory states. It consists of two main parts: a base method that processes a single CT scan and an extended method that utilizes the segmentation result obtained on the inspiration scan as a subject-specific prior for segmentation of the expiration scan. We evaluated the methods on a diverse set of 40 CT scan pairs. In addition, we compare the performance of our method to a registration-based approach. On inspiration scans, the base method achieved an average distance error of 0.59, 0.64, and 0.91 mm for the left oblique, right oblique, and right horizontal fissures, respectively, when compared with expert-based reference tracings. On expiration scans, the base method’s errors were 1.54, 3.24, and 3.34 mm, respectively. In comparison, utilizing proposed subject-specific priors for segmentation of expiration scans allowed decreasing average distance errors to 0.82, 0.79, and 1.04 mm, which represents a significant improvement ( p&lt;0.05) compared with all other methods investigated.

To establish the clinical validity and accuracy of automatic thresholding and manual trimming (ATMT) by comparing the method with the conventional contouring method for in vivo cardiac volume measurements. CMR was performed on 40 subjects (30 patients and 10 controls) using steady-state free precession cine sequences with slices oriented in the short-axis and acquired contiguously from base to apex. Left ventricular (LV) volumes, end-diastolic volume, end-systolic volume, and stroke volume (SV) were obtained with ATMT and with the conventional contouring method. Additionally, SV was measured independently using CMR phase velocity mapping (PVM) of the aorta for validation. Three methods of calculating SV were compared by applying Bland–Altman analysis. The Bland–Altman standard deviation of variation (SD) and offset bias for LV SV for the three sets of data were: ATMT-PVM (7.65, −0.82), ATMT-contours (7.85, −5.78), and contour-PVM (11.01, 4.97), respectively. Equating the observed range to the error contribution of each approach, the error magnitude of ATMT:PVM:contours was in the ratio 1:2.4:2.5. Use of ATMT for measuring ventricular volumes accommodates trabeculae and papillary structures more intuitively than contemporary contouring methods. This results in lower variation when analyzing cardiac structure and function and consequently improved accuracy in assessing chamber volumes.

The diffusion tensor model is nonspecific in regions where micrometer structural patterns are inconsistent at the millimeter scale (i.e., brain regions with pathways that cross, bend, branch, fan, etc.). Numerous models have been proposed to represent crossing fibers and complex intravoxel structure from in vivo diffusion weighted magnetic resonance imaging (e.g., high angular resolution diffusion imaging—HARDI). Here, we present an empirical comparison of two HARDI approaches—persistent angular structure MRI (PAS-MRI) and Q-ball—using a newly acquired reproducibility dataset. Briefly, a single subject was scanned 11 times with 96 diffusion weighted directions and 10 reference volumes for each of two b values (1000 and 3000  s  /  mm2 for a total of 2144 volumes). Empirical reproducibility of intravoxel fiber fractions (number/strength of peaks), angular orientation, and fractional anisotropy was compared with metrics from a traditional tensor analysis approach, focusing on b values of 1000 and 3000  s  /  mm2. PAS-MRI is shown to be more reproducible than Q-ball and offers advantages at low b values. However, there are substantial and biologically meaningful differences between the intravoxel structures estimated both in terms of analysis method as well as by b value. The two methods suggest a fundamentally different microarchitecture of the human brain; therefore, it is premature to perform meta-analysis or combine results across HARDI studies using a different analysis model or acquisition sequences.

Optic disc (OD) appearance in fundus images is one of the clinical indicators considered in the assessment of retinal diseases such as glaucoma. The cup-to-disc ratio (CDR) is the most common clinical measurement used to characterize glaucoma. However, the CDR only evaluates the relative sizes of the cup and the OD via their diameters. We propose to construct an atlas-based shape descriptor (ASD) to statistically characterize the geometric deformations of the OD shape and of the blood vessels’ configuration inside the OD region. A local representation of the OD region is proposed to construct a well-defined statistical atlas using nonlinear registration and statistical analysis of deformation fields. The shape descriptor is defined as being composed of several statistical measures from the atlas. Analysis of the average model and its principal modes of deformation are performed on a healthy population. The components of the ASD show a significant difference between pathological and healthy ODs. We show that the ASD is able to characterize healthy and glaucomatous OD regions. The deviation map extracted from the atlas can be used to assist clinicians in an early detection of deformation abnormalities in the OD region.

We present a general framework for automatic segmentation of fetal brain structures in ultrasound images inspired by recent advances in machine learning. The approach is based on a region descriptor that characterizes the shape and local intensity context of different neurological structures without explicit models. To validate our framework, we present experiments to segment two fetal brain structures of clinical importance that have quite different ultrasonic appearances—the corpus callosum (CC) and the choroid plexus (CP). Results demonstrate that our approach achieves high region segmentation accuracy (dice coefficient: 0.81  %    ±  0.06 CC, 0.76  %    ±  0.08 CP) relative to human delineation, whereas the derived automated biometry measurement deviations are within human intra/interobserver variations. The use of our proposed method may help to standardize intracranial anatomy measurements for both the routine examination and the detection of congenital conditions in the future.

A method is presented to automatically track and segment pelvic organs on dynamic magnetic resonance imaging (MRI) followed by multiple-object trajectory classification to improve understanding of pelvic organ prolapse (POP). POP is a major health problem in women where pelvic floor organs fall from their normal position and bulge into the vagina. Dynamic MRI is presently used to analyze the organs’ movements, providing complementary support for clinical examination. However, there is currently no automated or quantitative approach to measure the movement of the pelvic organs and their correlation with the severity of prolapse. In the proposed method, organs are first tracked and segmented using particle filters and k-means clustering with prior information. Then, the trajectories of the pelvic organs are modeled using a coupled switched hidden Markov model to classify the severity of POP. Results demonstrate that the presented method can automatically track and segment pelvic organs with a Dice similarity index above 78% and Hausdorff distance of <5.2  mm for 94 tested cases while demonstrating correlation between organ movement and POP. This work aims to enable automatic tracking and analysis of multiple deformable structures from images to improve understanding of medical disorders.

Patients suffering from ischemic stroke develop varying degrees of pial arterial supply (PAS), which can affect patient response to reperfusion therapy and risk of hemorrhage. Since vessel segmentation may be an important part in identifying PAS, we present a fuzzy c-means (FCM) clustering method to segment major vessels in x-ray angiograms. Our approach consists of semiautomatic region of interest (ROI) delineation, separation of major vessels from capillary blush and/or background noise through FCM clustering, and identification of the major vessel category. This method was applied to a database of x-ray angiograms of 24 patients acquired at various frame rates. The ground truth for performance evaluation was the designation by an expert radiologist selecting image pixels as being vessel or nonvessel. From receiver operating characteristic (ROC) analysis, area under the ROC curve (AUC) was the performance metric in the task of distinguishing between major vessels and blush or background. When clustering data into three categories and performing FCM segmentation on each ROI separately, the AUC was 0.89 for the entire database and ≥0.87 for all examined frame-rates. In conclusion, our method showed promising performance in identifying major vessels and is anticipated to become an integral part of automatic quantification of PAS.

Current computer-aided detection (CADe) systems for contrast-enhanced breast MRI rely on both spatial information obtained from the early-phase and temporal information obtained from the late-phase of the contrast enhancement. However, late-phase information might not be available in a screening setting, such as in abbreviated MRI protocols, where acquisition is limited to early-phase scans. We used deep learning to develop a CADe system that exploits the spatial information obtained from the early-phase scans. This system uses three-dimensional (3-D) morphological information in the candidate locations and the symmetry information arising from the enhancement differences of the two breasts. We compared the proposed system to a previously developed system, which uses the full dynamic breast MRI protocol. For training and testing, we used 385 MRI scans, containing 161 malignant lesions. Performance was measured by averaging the sensitivity values between 1/8—eight false positives. In our experiments, the proposed system obtained a significantly ( p=0.008) higher average sensitivity ( 0.6429±0.0537) compared with that of the previous CADe system ( 0.5325±0.0547). In conclusion, we developed a CADe system that is able to exploit the spatial information obtained from the early-phase scans and can be used in screening programs where abbreviated MRI protocols are used.

Deep learning methods have been shown to improve breast cancer diagnostic and prognostic decisions based on selected slices of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). However, incorporation of volumetric and temporal components into DCE-MRIs has not been well studied. We propose maximum intensity projection (MIP) images of subtraction MRI as a way to simultaneously include four-dimensional (4-D) images into lesion classification using convolutional neural networks (CNN). The study was performed on a dataset of 690 cases. Regions of interest were selected around each lesion on three MRI presentations: (i) the MIP image generated on the second postcontrast subtraction MRI, (ii) the central slice of the second postcontrast MRI, and (iii) the central slice of the second postcontrast subtraction MRI. CNN features were extracted from the ROIs using pretrained VGGNet. The features were utilized in the training of three support vector machine classifiers to characterize lesions as malignant or benign. Classifier performances were evaluated with fivefold cross-validation and compared based on area under the ROC curve (AUC). The approach using MIPs [AUC=0.88(se=0.01)] outperformed that using central-slices of either second postcontrast MRIs [0.80(se=0.02)] or second postcontrast subtraction MRIs [AUC=0.84(se=0.02)], at statistically significant levels.

We study the performance of a computer-aided detection (CAD) system for lung nodules in computed tomography (CT) as a function of slice thickness. In addition, we propose and compare three different training methodologies for utilizing nonhomogeneous thickness training data (i.e., composed of cases with different slice thicknesses). These methods are (1) aggregate training using the entire suite of data at their native thickness, (2) homogeneous subset training that uses only the subset of training data that matches each testing case, and (3) resampling all training and testing cases to a common thickness. We believe this study has important implications for how CT is acquired, processed, and stored. We make use of 192 CT cases acquired at a thickness of 1.25 mm and 283 cases at 2.5 mm. These data are from the publicly available Lung Nodule Analysis 2016 dataset. In our study, CAD performance at 2.5 mm is comparable with that at 1.25 mm and is much better than at higher thicknesses. Also, resampling all training and testing cases to 2.5 mm provides the best performance among the three training methods compared in terms of accuracy, memory consumption, and computational time.

We proposed the neutrosophic approach for segmenting breast lesions in breast computed tomography (bCT) images. The neutrosophic set considers the nature and properties of neutrality (or indeterminacy). We considered the image noise as an indeterminate component while treating the breast lesion and other breast areas as true and false components. We iteratively smoothed and contrast-enhanced the image to reduce the noise level of the true set. We then applied one existing algorithm for bCT images, the RGI segmentation, on the resulting noise-reduced image to segment the breast lesions. We compared the segmentation performance of the proposed method (named as NS-RGI) to that of the regular RGI segmentation. We used 122 breast lesions (44 benign and 78 malignant) of 111 noncontrast enhanced bCT cases. We measured the segmentation performances of the NS-RGI and the RGI using the Dice coefficient. The average Dice values of the NS-RGI and RGI were 0.82 and 0.80, respectively, and their difference was statistically significant (p value  =  0.004). We conducted a subsequent feature analysis on the resulting segmentations. The classifier performance for the NS-RGI (AUC  =  0.80) improved over that of the RGI (AUC  =  0.69, p value  =  0.006).

Optical techniques such as fluorescence and diffuse reflectance spectroscopy are proven to have the potential to provide tissue discrimination during the development of malignancies and hence treated as potential tools for noninvasive optical biopsy in clinical diagnostics. Quantitative optical biopsy is challenging and hence the majority of the existing strategies are based on a qualitative assessment of the concerned tissue. Light–tissue interaction models as well as precise optical phantoms can greatly help in the former and here we present a pilot study to assess the optical properties of a multilayer tissue-specific optical phantom with the help of a database generated using multilayer-Monte Carlo (MCML) models. A set of optical models mimicking the properties of actual and diseased conditions of tissues associated with nonmelanoma skin cancer (NMSC) were devised and MCML simulations of fluorescence and diffuse reflectance were performed on these models to generate the spectral signature of identified biomarkers of NMSC such as hemoglobin, flavin adenine dinucleotide, and collagen. A model library was generated and with the extracted features from modeled spectra, classification of normal and NMSC conditions were tested using the K-nearest neighbor (KNN) classifier. Using an in-house assembled scan-based automated bimodal spectral imaging system with reflectance and fluorescence modalities of operation, a layered, thin, tissue equivalent phantom, fabricated with controlled optical properties mimicking normal and NMSC conditions were tested. The spectral signatures corresponding to the NMSC biomarkers were acquired from this phantom and extracted features from the spectra were tested using the KNN classifier and classification accuracy of 100% was achieved. For further quantitative analysis, the experimental and simulated spectra were compared with respect to the light intensity at the emission peak or absorption dips, spectral line width, and average intensity over a range of wavelength of interest and observed to be analogous within specified and systematic error limits. This methodology is expected to give a better quantitative approach for estimation of tissue properties by correlating the experimental and simulated data.

Imaging of Cherenkov light emission from patient tissue during fractionated radiotherapy has been shown to be a possible way to visualize beam delivery in real time. If this tool is advanced as a delivery verification methodology, then a sequence of image processing steps must be established to maximize accurate recovery of beam edges. This was analyzed and developed here, focusing on the noise characteristics and representative images from both phantoms and patients undergoing whole breast radiotherapy. The processing included temporally integrating video data into a single, composite summary image at each control point. Each image stack was also median filtered for denoising and ultimately thresholded into a binary image, and morphologic small hole removal was used. These processed images were used for day-to-day comparison computation, and either the Dice coefficient or the mean distance to conformity values can be used to analyze them. Systematic position shifts of the phantom up to 5 mm approached the observed variation values of the patient data. This processing algorithm can be used to analyze the variations seen in patients being treated concurrently with daily Cherenkov imaging to quantify the day-to-day disparities in delivery as a quality audit system for position/beam verification.

Deep brain stimulation (DBS) of the subthalamic nucleus (STN) reduces motor symptoms in most patients with Parkinson disease (PD), yet may produce untoward effects. Investigation of DBS effects requires accurate localization of the STN, which can be difficult to identify on magnetic resonance images collected with clinically available 3T scanners. The goal of this study is to develop a high-quality STN atlas that can be applied to standard 3T images. We created a high-definition STN atlas derived from seven older participants imaged at 7T. This atlas was nonlinearly registered to a standard template representing 56 patients with PD imaged at 3T. This process required development of methodology for nonlinear multimodal image registration. We demonstrate mm-scale STN localization accuracy by comparison of our 3T atlas with a publicly available 7T atlas. We also demonstrate less agreement with an earlier histological atlas. STN localization error in the 56 patients imaged at 3T was less than 1 mm on average. Our methodology enables accurate STN localization in individuals imaged at 3T. The STN atlas and underlying 3T average template in MNI space are freely available to the research community. The image registration methodology developed in the course of this work may be generally applicable to other datasets.

Biomechanical breast models have been employed for applications in image registration and diagnostic analysis, breast augmentation simulation, and for surgical and biopsy guidance. Accurate applications of stress–strain relationships of tissue within the breast can improve the accuracy of biomechanical models that attempt to simulate breast deformations. Reported stiffness values for adipose, glandular, and cancerous tissue types vary greatly. Variations in reported stiffness properties have been attributed to differences in testing methodologies and assumptions, measurement errors, and natural interpatient differences in tissue elasticity. Therefore, the ability to determine patient-specific in vivo breast tissue properties would be advantageous for these procedural applications. While some in vivo elastography methods are not quantitative and others do not measure material properties under deformation conditions that are appropriate to the application of concern, in this study, we developed an elasticity estimation method that is performed using deformations representative of supine therapeutic procedures. More specifically, reconstruction of mechanical properties appropriate for the standard-of-care supine lumpectomy was performed by iteratively fitting two anatomical images before and after deformations taking place in the supine breast configuration. The method proposed is workflow-friendly, quantitative, and uses a noncontact, gravity-induced deformation source.

Photoacoustic imaging (PAI) is an emerging biomedical imaging technology, which can potentially be used in the clinic to preoperatively measure melanoma thickness and guide biopsy depth and sample location. We recruited 27 patients with pigmented cutaneous lesions suspicious for melanoma to test the feasibility of a handheld linear-array photoacoustic probe in imaging lesion architecture and measuring tumor depth. The probe was assessed in terms of measurement accuracy, image quality, and ease of application. Photoacoustic scans included single wavelength, spectral unmixing, and three-dimensional (3-D) scans. The photoacoustically measured lesion thickness gave a high correlation with the histological thickness measured from resected surgical samples (r=0.99, P<0.001 for melanomas, r=0.98, P<0.001 for nevi). Thickness measurements were possible for 23 of 26 cases for nevi and all (6) cases for melanoma. Our results show that handheld, linear-array PAI is highly reliable in measuring cutaneous lesion thickness in vivo, and can potentially be used to inform biopsy procedure and improve patient management.

Positioning of an intraoperative C-arm to achieve clear visualization of a particular anatomical feature often involves repeated fluoroscopic views, which cost time and radiation exposure to both the patient and surgical staff. A system for virtual fluoroscopy (called FluoroSim) that could dramatically reduce time- and dose-spent “fluoro-hunting” by leveraging preoperative computed tomography (CT), encoded readout of C-arm gantry position, and automatic 3D–2D image registration has been developed. The method is consistent with existing surgical workflow and does not require additional tracking equipment. Real-time virtual fluoroscopy was achieved via mechanical encoding of the C-arm motion, C-arm geometric calibration, and patient registration using a single radiograph. The accuracy, time, and radiation dose associated with C-arm positioning were measured for FluoroSim in comparison with conventional methods. Five radiology technologists were tasked with acquiring six standard pelvic views pertinent to sacro-illiac, anterior–inferior iliac spine, and superior-ramus screw placement in an anthropomorphic pelvis phantom using conventional and FluoroSim approaches. The positioning accuracy, exposure time, number of exposures, and total time for each trial were recorded, and radiation dose was characterized in terms of entrance skin dose and in-room scatter. The geometric accuracy of FluoroSim was measured to be 1.6±1.1  mm. There was no significant difference (p<0.05) observed in the accuracy or total elapsed time for C-arm positioning. However, the total fluoroscopy time required to achieve the desired view decreased by 4.1 s (4.7±3.6  s for conventional, compared with 0.5±0.0  s for FluoroSim, p<0.05), and the total number of exposures reduced by 4.0 (6.4±4.8 for conventional, compared with 2.0±0.0 for FluoroSim, p<0.05). These reductions amounted to a 50% to 78% decrease in patient entrance skin dose and a 55% to 70% reduction in in-room scatter. FluoroSim was found to reduce the radiation exposure required in C-arm positioning without diminishing positioning time or accuracy, providing a potentially valuable tool to assist technologists and surgeons.

All medical image segmentation algorithms need to be validated and compared, yet no evaluation framework is widely accepted within the imaging community. None of the evaluation metrics that are popular in the literature are consistent in the way they rank segmentation results: they tend to be sensitive to one or another type of segmentation error (size, location, and shape) but no single metric covers all error types. We introduce a family of metrics, with hybrid characteristics. These metrics quantify the similarity or difference of segmented regions by considering their average overlap in fixed-size neighborhoods of points on the boundaries of those regions. Our metrics are more sensitive to combinations of segmentation error types than other metrics in the existing literature. We compare the metric performance on collections of segmentation results sourced from carefully compiled two-dimensional synthetic data and three-dimensional medical images. We show that our metrics: (1) penalize errors successfully, especially those around region boundaries; (2) give a low similarity score when existing metrics disagree, thus avoiding overly inflated scores; and (3) score segmentation results over a wider range of values. We analyze a representative metric from this family and the effect of its free parameter on error sensitivity and running time.

Digital breast tomosynthesis (DBT) acquires a series of projection images from different angles as an x-ray source rotates around the breast. Such imaging geometry lends DBT naturally to stereoscopic viewing as two projection images with a reasonable separation angle can easily form a stereo pair. This simulation study assessed the efficacy of stereo viewing of DBT projection images. Three-dimensional computational breast phantoms with realistically shaped synthetic lesions were scanned by three simulated DBT systems. The projection images were combined into a sequence of stereo pairs and presented to a stereomatching-based model observer for deciding lesion presence. Signal-to-noise ratio was estimated, and the detection performance with stack viewing of reconstructed slices was the benchmark. We have shown that: (1) stereo viewing of projection images may underperform stack viewing of reconstructed slices for current DBT geometries; (2) DBT geometries may impact the efficacy of the two viewing modes differently: narrow-arc and wide-arc geometries may be better for stereo viewing and stack viewing, respectively; (3) the efficacy of stereo viewing may be more robust than stack viewing to reductions in dose. While in principle stereo viewing is potentially effective for visualizing DBT data, effective stereo viewing may require specifically optimized DBT image acquisition.

Phantom-based initial performance assessment of a prototype three-dimensional (3-D) x-ray system and comparison of 3-D tomography with computed tomography (CT) were proposed. A 3-D image quality phantom was scanned with a prototype version of 3-D cone-beam CT imaging implemented on a twin robotic x-ray system using three trajectories (163 deg = table, 188 deg = upright, and 200 deg = side), six tube voltages (60, 70, 81, 90, 100, and 121 kV), and four detector doses (0.348, 0.696, 1.740, and 3.480  μGy  /  projection). CT was obtained with a clinical protocol. Spatial resolution (line pairs/cm) and soft-tissue-contrast resolution were assessed by two independent readers. Radiation dose was assessed. Descriptive and analysis of variance (ANOVA) (P  &lt;  0.05) were performed. With 3-D tomography, a maximum of 16 lp/cm was visible and best soft-tissue-contrast resolution was 2 mm at 30 Hounsfield units (HU) for 160 projections. With CT, 10 lp/cm was visible and soft-tissue-contrast resolution was 4 mm at 20 HU. The upright trajectory yielded significantly better spatial resolution and soft tissue contrast, and the side trajectory yielded significantly higher soft tissue contrast than the table trajectory (P  &lt;  0.05). Radiation dose was higher in 3-D tomography (45 to 704 mGycm) than CT (44 mGycm). Three-dimensional tomography renders overall equal or higher spatial resolution and comparable soft tissue contrast to CT for medium- and high-dose protocols in the side and upright trajectories, but with higher radiation doses.

Minimal invasive endoscopic treatment for upper urinary tract urothelial carcinoma (UUT-UC) is advocated in patients with low-risk disease and limited tumor volume. Diagnostic ureterorenoscopy combined with biopsy is the diagnostic standard. This study aims to evaluate two alternative diagnostic techniques for UUT-UC: optical coherence tomography (OCT) and endoluminal ultrasound (ELUS). Following nephroureterectomy, OCT, ELUS, and computed tomography (CT) were performed of the complete nephroureterectomy specimen. Visualization software (AMIRA^®) was used for reconstruction and coregistration of CT, OCT, and ELUS. Finally, CT was used to obtain exact probe localization. Coregistered OCT and ELUS datasets were compared with histology. Coregistration with three-dimensional CT makes exact data matching possible in this ex-vivo setting to compare histology with OCT and ELUS. In OCT images of normal-appearing renal pelvis and ureter, urothelium, lamina propria, and muscularis were visible. With ELUS, all anatomical layers of the ureter could be distinguished, besides the urothelial layer. ELUS identified suspect lesions, although exact staging and differentiation between noninvasive and invasive lesions were not possible. OCT provides high-resolution imaging of normal ureter and ureter lesions. ELUS, however, is of limited value as it cannot differentiate between noninvasive and invasive tumors.