Guest editors Patrick La Riviere, Rebecca Fahrig, and Norbert Pelc introduce the JMI Special Section Celebrating X-Ray Computed Tomography at 50.

Purpose: Coronary artery calcium (CAC) scoring with computed tomography (CT) has been proposed as a screening tool for coronary artery disease, but concerns remain regarding the radiation dose of CT CAC scoring. Photon counting detectors and iterative reconstruction (IR) are promising approaches for patient dose reduction, yet the preservation of CAC scores with IR has been questioned. The purpose of this study was to investigate the applicability of IR for quantification of CAC using a photon counting flat-detector.

Approach: We imaged a cardiac rod phantom with calcium hydroxyapatite (CaHA) inserts with different noise levels using an experimental photon counting flat-detector CT setup to simulate the clinical CAC scoring protocol. We applied filtered back projection (FBP) and two IR algorithms with different regularization strengths. We compared the air kerma values, image quality parameters [noise magnitude, noise power spectrum, modulation transfer function (MTF), and contrast-to-noise ratio], and CaHA quantification accuracy between FBP and IR.

Results: IR regularization strength influenced CAC scores significantly (p  <  0.05). The CAC volumes and scores between FBP and IRs were the most similar when the IR regularization strength was chosen to match the MTF of the FBP reconstruction.

Conclusion: When the regularization strength is selected to produce comparable spatial resolution with FBP, IR can yield comparable CAC scores and volumes with FBP. Nonetheless, at the lowest radiation dose setting, FBP produced more accurate CAC volumes and scores compared to IR, and no improved CAC scoring accuracy at low dose was demonstrated with the utilized IR methods.

Purpose: Cone-beam computed tomography (CBCT) is commonly used in the operating room to evaluate the placement of surgical implants in relation to critical anatomical structures. A particularly problematic setting, however, is the imaging of metallic implants, where strong artifacts can obscure visualization of both the implant and surrounding anatomy. Such artifacts are compounded when combined with low-dose imaging techniques such as sparse-view acquisition.

Approach: This work presents a dual convolutional neural network approach, one operating in the sinogram domain and one in the reconstructed image domain, that is specifically designed for the physics and setting of intraoperative CBCT to address the sources of beam hardening and sparse view sampling that contribute to metal artifacts. The networks were trained with images from cadaver scans with simulated metal hardware.

Results: The trained networks were tested on images of cadavers with surgically implanted metal hardware, and performance was compared with a method operating in the image domain alone. While both methods removed most image artifacts, superior performance was observed for the dual-convolutional neural network (CNN) approach in which beam-hardening and view sampling effects were addressed in both the sinogram and image domain.

Conclusion: The work demonstrates an innovative approach for eliminating metal and sparsity artifacts in CBCT using a dual-CNN framework which does not require a metal segmentation.

Purpose: We developed a deep learning method to reduce noise and beam-hardening artifact in virtual monoenergetic image (VMI) at low x-ray energy levels.

Approach: An encoder–decoder type convolutional neural network was implemented with customized inception modules and in-house-designed training loss (denoted as Incept-net), to directly estimate VMI from multi-energy CT images. Images of an abdomen-sized water phantom with varying insert materials were acquired from a research photon-counting-detector CT. The Incept-net was trained with image patches (64  ×  64  pixels) extracted from the phantom data, as well as synthesized, random-shaped numerical insert materials. The whole CT images (512  ×  512  pixels) with the remaining real insert materials that were unseen in network training were used for testing. Seven contrast-enhanced abdominal CT exams were used for preliminary evaluation of Incept-net generalizability over anatomical background. Mean absolute percentage error (MAPE) was used to evaluate CT number accuracy.

Results: Compared to commercial VMI software, Incept-net largely suppressed beam-hardening artifact and reduced noise (53%) in phantom study. Incept-net presented comparable CT number accuracy at higher-density (P-value [0.0625, 0.999]) and improved it at lower-density inserts (P-value  =  0.0313) with overall MAPE: Incept-net [2.9%, 4.6%]; commercial-VMI [6.7%, 10.9%]. In patient images, Incept-net suppressed beam-hardening artifact and reduced noise (up to 50%, P-value  =  0.0156).

Conclusion: In this preliminary study, Incept-net presented the potential to improve low-energy VMI quality.

Purpose: To characterize variability in image quality and radiation dose across a large cohort of computed tomography (CT) examinations and identify the scan factors with the highest influence on the observed variabilities.

Approach: This retrospective institutional-review-board-exempt investigation was performed on 87,629 chest and abdomen-pelvis CT scans acquired for 97 facilities from 2018 to 2019. Images were assessed in terms of noise, resolution, and dose metrics (global noise, frequency in which modulation transfer function is at 0.50, and volumetric CT dose index, respectively). The results were fit to linear mixed-effects models to quantify the variabilities as affected by scan parameters and settings and patient characteristics. A list of factors, ranked by t-value with p  <  0.05, was ascertained for each of the six mixed effects models. A type III p-value test was used to assess the influence of facility.

Results: Across different facilities, image quality and dose were significantly different (p  <  0.05), with little correlation between their mean magnitudes and consistency (Pearson’s correlation coefficient  <  0.34). Scanner model, slice thickness, recon field-of-view and kernel, mAs, kVp, patient size, and centering were the most influential factors. The two body regions exhibited similar rankings of these factors for noise (Spearman’s correlation coefficient  =  0.76) and dose (Spearman’s correlation coefficient  =  0.86) but not for resolution (Spearman’s correlation coefficient  =  0.52).

Conclusions: Clinical CT scans can vary in image quality and dose with broad implications for diagnostic utility and radiation burden. Average scan quality was not correlated with interpatient scan-quality consistency. For a given facility, this variability can be quite large, with magnitude differences across facilities. The knowledge of the most influential factors per body region may be used to better manage these variabilities within and across facilities.

Purpose: To investigate the influence of radiographic contrast agent on the accuracy of the photon counts arising from the emission of gamma rays of radionuclides in single-photon emission computed tomography (SPECT), when dual-energy x-ray CT (DXCT) is employed for providing object/energy-specific attenuation coefficient correction in SPECT.

Approach: Computer simulation was performed for three transmission CT approaches, namely, the conventional (single kVp, unimodal spectrum) x-ray CT, DXCT (single kVp, bimodal spectrum) with basis material decomposition (BMD), and DXCT with BMD followed by basis material coefficients transformation (BMT), to study the effects of these approaches on the accuracy of the photon counts from the SPECT image of a thorax-like phantom.

Results: All three CT approaches revealed that the error in the counts was both photon energy and iodine concentration-dependent. Differences in the trending increase/decrease in the errors with the respective increase in iodine concentration and photon energy were observed among the three CT approaches. Of the three, the BMT/SPECT approach resulted in the smallest error in the concentration of radionuclides measured, especially in the contrast agent-filled region, and the optimal level depended on the iodine concentration and photon energy.

Conclusion: With a judicious choice of the basis materials and photon energy, it may be possible to take advantage of the benefits of the BMT method to mitigate the accuracy problem in DXCT for quantitative SPECT imaging.

Purpose: To demonstrate the utility of high-resolution micro-computed tomography (μCT) for determining ground-truth size and shape properties of calcium grains for evaluation of detection performance in breast CT (bCT).

Approach: Calcium carbonate grains (∼200  μm) were suspended in 1% agar solution to emulate microcalcifications (μCalcs) within a fibroglandular tissue background. Ground-truth imaging was performed on a commercial μCT scanner and was used for assessing calcium-grain size and shape, and for generating μCalc signal profiles. Calcium grains were placed within a realistic breast-shaped phantom and imaged on a prototype bCT system at 3- and 6-mGy mean glandular dose (MGD) levels, and the non-prewhitening detectability was assessed. Additionally, the μCT-derived signal profiles were used in conjunction with the bCT system characterization (MTF and NPS) to obtain predictions of bCT detectability.

Results: Estimated detectability of the calcium grains on the bCT system ranged from 2.5 to 10.6 for 3 mGy and from 3.8 to 15.3 for 6 mGy with large fractions of the grains meeting the Rose criterion for visibility. Segmentation of μCT images based on morphological operations produced accurate results in terms of segmentation boundaries and segmented region size. A regression model linking bCT detectability to μCalc parameters indicated significant effects of μCalc size and vertical position within the breast phantom. Detectability using μCT-derived detection templates and bCT statistical properties (MTF and NPS) were in good correspondence with those measured directly from bCT (R²  >  0.88).

Conclusions: Parameters derived from μCT ground-truth data were shown to produce useful characterizations of detectability when compared to estimates derived directly from bCT. Signal profiles derived from μCT imaging can be used in conjunction with measured or hypothesized statistical properties to evaluate the performance of a system, or system component, that may not currently be available.

Purpose: Breast cancer is the most common cancer in women in developing and developed countries and is responsible for 15% of women’s cancer deaths worldwide. Conventional absorption-based breast imaging techniques lack sufficient contrast for comprehensive diagnosis. Propagation-based phase-contrast computed tomography (PB-CT) is a developing technique that exploits a more contrast-sensitive property of x-rays: x-ray refraction. X-ray absorption, refraction, and contrast-to-noise in the corresponding images depend on the x-ray energy used, for the same/fixed radiation dose. The aim of this paper is to explore the relationship between x-ray energy and radiological image quality in PB-CT imaging.

Approach: Thirty-nine mastectomy samples were scanned at the imaging and medical beamline at the Australian Synchrotron. Samples were scanned at various x-ray energies of 26, 28, 30, 32, 34, and 60 keV using a Hamamatsu Flat Panel detector at the same object-to-detector distance of 6 m and mean glandular dose of 4 mGy. A total of 132 image sets were produced for analysis. Seven observers rated PB-CT images against absorption-based CT (AB-CT) images of the same samples on a five-point scale. A visual grading characteristics (VGC) study was used to determine the difference in image quality.

Results: PB-CT images produced at 28, 30, 32, and 34 keV x-ray energies demonstrated statistically significant higher image quality than reference AB-CT images. The optimum x-ray energy, 30 keV, displayed the largest area under the curve   (  AUC_VGC  )   of 0.754 (p  =  0.009). This was followed by 32 keV (AUC_VGC  =  0.731, p  ≤  0.001), 34 keV (AUC_VGC  =  0.723, p  ≤  0.001), and 28 keV (AUC_VGC  =  0.654, p  =  0.015).

Conclusions: An optimum energy range (around 30 keV) in the PB-CT technique allows for higher image quality at a dose comparable to conventional mammographic techniques. This results in improved radiological image quality compared with conventional techniques, which may ultimately lead to higher diagnostic efficacy and a reduction in breast cancer mortalities.

Purpose: We provide a review of the key computed tomography (CT) technologies developed since the late 1980s and offer an overview of one of the future technologies under development. The focus of this review is mainly on the hardware and system development. The topics on the historical event linked to the early days of CT development and other innovations that contributed to the CT development, such as advanced image reconstruction techniques, are covered by companion papers in this special issue.

Approach: The review is divided into five major sections, each linked to a key innovation in CT: helical spiral data acquisition, multi-slice CT, wide-cone CT, dual-source CT, and spectral CT. Given the limited scope of this review, only one of the future technologies, photon-counting CT, is discussed in detail. Whenever possible, both theory of operation and clinical examples are provided.

Results: Theoretical analyses, phantom results, and clinical examples clearly demonstrate the efficacy and clinical relevancy of five historical technology developments and one future technology in CT. These technologies have improved and will continue to improve CT performance in terms of isotropic volume coverage, improved temporal resolution, and material differentiation and characterization capabilities.

Conclusions: Over the past 30 years, technological developments of CT have contributed to the success of CT in many clinical applications such as trauma, oncology, cardiac imaging, and stroke. Advanced clinical applications have and will continue to demand more advanced technology development.

As we arrive at the 50th anniversary of the first computed tomography (CT) scan of a live patient, we take this opportunity to revisit the history of early CT development. It is not an exaggeration to say that the invention of CT may represent the greatest revolution in medical imaging since the discovery of x-rays. We cover events over a period of about two decades that started with the realization that accurate cross-sectional soft-tissue detail is possible and could be a significant advance. We describe in some detail the development of the first CT system and then the rapid technical advances during the following years that included the entry of many companies into the field and the circumstances that led many of those entrants to exit the field. Rather than focusing on the specific technical details (which can be found elsewhere), we include stories and events in the hope that broader lessons can be learned. As the first x-ray-based digital imaging modality, CT brought into common use an exceptional tool that benefits countless patients every day. It also introduced dramatic changes to biomedical imaging as a field that continues to influence progress to this day.

Computed tomography was one of the first imaging modalities to require a computerized solution of an inverse problem to produce a useful image from the data acquired by the sensor hardware. The computerized solutions, which are known as image reconstruction algorithms, have thus been a critical component of every CT scanner ever sold. We review the history of commercially deployed CT reconstruction algorithms and consider the forces that led, at various points, both to innovation and to convergence around certain broadly useful algorithms. The forces include the emergence of new hardware capabilities, competitive pressures, the availability of computational power, and regulatory considerations. We consider four major historical periods and turning points. The original EMI scanner was developed with an iterative reconstruction algorithm, but an explosion of innovation coupled with rediscovery of an older literature led to the development of alternative algorithms throughout the early 1970s. Most CT vendors quickly converged on the use of the filtered back-projection (FBP) algorithm, albeit layered with a variety of proprietary corrections in both projection data and image domains to improve image quality. Innovations such as helical scanning and multi-row detectors were both enabled by and drove the development of additional applications of FBP in the 1990s and 2000s. Finally, the last two decades have seen a return of iterative reconstruction and the introduction of artificial intelligence approaches that benefit from increased computational power to reduce radiation dose and improve image quality.

Purpose: It has been 50 years since computed tomography was introduced to the worldwide neurologic medical and surgical community. In that time, tremendous advances in computer software and hardware, as well as creative changes in computerized tomographic (CT) hardware and tube technology, have dramatically improved the temporal and spatial resolution of CT. In this paper, I address what I feel are some of the most important impacts of CT in the field of clinical neuroscience over the last 50 years, as well as potential applications of CT that are on the horizon.

Approach: I have recounted from literature, colleagues, and personal recollection the historical impact of CT on neuroradiology practice and what appear to be near-term future applications.

Conclusions: Therapeutic applications beyond diagnosis, such as image-guided procedures, radiation, and surgical planning, and development of the field of theranostics have emerged and further increased the need for faster and more precise CT imaging. The integration of machine learning into the acquisition chain and radiologist tool kit has great implications for standardization, analysis, and diagnosis worldwide.

Purpose: Developing, validating, and evaluating a method for measuring noise texture directly from patient liver CT images (i.e., in vivo).

Approach: The method identifies target regions within patient scans that are least likely to have major contribution of patient anatomy, detrends them locally, and measures noise power spectrum (NPS) there using a previously phantom-validated technique targeting perceptual noise–non-anatomical fluctuations in the image that may interfere with the detection of focal lesions. Method development and validation used scanner-specific CT simulations of computational, anthropomorphic phantom (XCAT phantom, three phases of contrast-enhancement) with known ground truth of the NPS. Simulations were based on a clinical scanner (Definition Flash, Siemens) and clinically relevant settings (tube voltage of 120 kV at three dose levels). Images were reconstructed with filtered backprojection (kernel: B31, B41, and B50) and Sinogram Affirmed Iterative Reconstruction (kernel: I31, I41, and I50) using a manufacturer-specific reconstruction software (ReconCT, Siemens). All NPS measurements were made in the liver. Ground-truth NPS were taken as the sum of (1) a measurement in parenchymal regions of anatomy-subtracted (i.e., noise only) scans, and (2) a measurement in the same region of noise-free (pre-noise-insertion) images. To assess in vivo NPS performance, correlation of NPS average frequency (f_avg), was reported. Sensitivity of accuracy [root-mean-square-error (RMSE)] to number of pixels included in measurement was conducted via bootstrapped pixel-dropout. Sensitivity of NPS to dose and reconstruction kernel was assessed to confirm that ground truth NPS similarities were maintained in patient-specific measurements.

Results: Pearson and Spearman correlation coefficients 0.97 and 0.96 for f_avg indicated good correlation. Results suggested accurate NPS measurements (within 5% total RMSE) could be acquired with ∼10⁶  pixels.

Conclusions: Relationships of similar NPS due to reconstruction kernel and dose were preserved between gold standard and observed in vivo estimations. The NPS estimation method was further deployed on clinical cases to demonstrate the feasibility of clinical analysis.

Purpose: The invention of carbon nanotube (CNT) x-ray source arrays has enabled the development of novel imaging systems, including stationary tomosynthesis and stationary computed tomography (CT) with fast data acquisition, mechanically robust structures, and reduced image blur from source–detector motion. In this work, we report the results of simulation studies of potential system configurations for a stationary head CT (s-HCT) using linear CNT x-ray sources and detector arrays.

Approach: We explored s-HCT configurations that utilize one, two, and three linear CNT source arrays. Simulations were implemented using three digital phantoms with both CPU and GPU computing. Sinogram coverage was used for qualitative evaluation of the CT projection collection efficiency for each configuration. A modified low-contrast Shepp–Logan (SL) phantom was implemented for image quality assessment using quantitative metrics. Different iterative reconstruction (IR) methods were compared with both qualitative and quantitative assessments.

Results: Sinogram coverage of s-HCT configurations was sensitive to the number of CNT source arrays and geometry. The simulations suggest that a s-HCT configuration with three planes gives near complete sinogram coverage. Such a configuration enables accurate reconstruction of the low-contrast SL phantom and considerably diminished artifacts caused by the system geometry.

Conclusions: An optimized s-HCT system configuration with three linear CNT x-ray source arrays is feasible. IR algorithms can diminish artifacts caused by sparse and asymmetrical scans. The proposed s-HCT system configuration is currently under construction.

Research into conebeam CT concepts began as soon as the first clinical single-slice CT scanner was conceived. Early implementations of conebeam CT in the 1980s focused on high-contrast applications where concurrent high resolution (<200  μm), for visualization of small contrast-filled vessels, bones, or teeth, was an imaging requirement that could not be met by the contemporaneous CT scanners. However, the use of nonlinear imagers, e.g., x-ray image intensifiers, limited the clinical utility of the earliest diagnostic conebeam CT systems. The development of consumer-electronics large-area displays provided a technical foundation that was leveraged in the 1990s to first produce large-area digital x-ray detectors for use in radiography and then compact flat panels suitable for high-resolution and high-frame-rate conebeam CT. In this review, we show the concurrent evolution of digital flat panel (DFP) technology and clinical conebeam CT. We give a brief summary of conebeam CT reconstruction, followed by a brief review of the correction approaches for DFP-specific artifacts. The historical development and current status of flat-panel conebeam CT in four clinical areas—breast, fixed C-arm, image-guided radiation therapy, and extremity/head—is presented. Advances in DFP technology over the past two decades have led to improved visualization of high-contrast, high-resolution clinical tasks, and image quality now approaches the soft-tissue contrast resolution that is the standard in clinical CT. Future technical developments in DFPs will enable an even broader range of clinical applications; research in the arena of flat-panel CT shows no signs of slowing down.

Purpose: Proton radiography may guide proton therapy cancer treatments with beam’s-eye-view anatomical images and a proton-based estimation of proton stopping power. However, without contrast enhancement, proton radiography will not be able to distinguish tumor from tissue. To provide this contrast, functionalized, high-Z nanoparticles that specifically target a tumor could be injected into a patient before imaging. We conducted this study to understand the ability of gold, as a high-Z, biologically compatible tracer, to differentiate tumors from surrounding tissue.

Approach: Acrylic and gold phantoms simulate a tumor tagged with gold nanoparticles (AuNPs). Calculations correlate a given thickness of gold to levels of tumor AuNP uptake reported in the literature. An identity, ×3, and ×7 proton magnifying lens acquired lens-refocused proton radiographs at the 800-MeV LANSCE proton beam. The effects of gold in the phantoms, in terms of percent density change, were observed as changes in measured transmission. Variable areal densities of acrylic modeled the thickness of the human body.

Results: A 1-μm-thick gold strip was discernible within 1 cm of acrylic, an areal density change of 0.2%. Behind 20 cm of acrylic, a 40-μm gold strip was visible. A 1-cm-diameter tumor tagged with 1  ×  10⁵ 50-nm AuNPs per cell has an amount of contrast agent embedded within it that is equivalent to a 65-μm thickness of gold, an areal density change of 0.63% in a tissue thickness of 20 cm, which is expected to be visible in a typical proton radiograph.

Conclusions: We indicate that AuNP-enhanced proton radiography might be a feasible technology to provide image-guidance to proton therapy, potentially reducing off-target effects and sparing nearby tissue. These data can be used to develop treatment plans and clinical applications can be derived from the simulations.

Purpose: Magnetic resonance imaging has been recently used to examine the abnormalities of the placenta during pregnancy. Segmentation of the placenta and uterine cavity allows quantitative measures and further analyses of the organs. The objective of this study is to develop a segmentation method with minimal user interaction.

Approach: We developed a fully convolutional neural network (CNN) for simultaneous segmentation of the uterine cavity and placenta in three dimensions (3D) while a minimal operator interaction was incorporated for training and testing of the network. The user interaction guided the network to localize the placenta more accurately. In the experiments, we trained two CNNs, one using 70 normal training cases and the other using 129 training cases including normal cases as well as cases with suspected placenta accreta spectrum (PAS). We evaluated the performance of the segmentation algorithms on two test sets: one with 20 normal cases and the other with 50 images from both normal women and women with suspected PAS.

Results: For the normal test data, the average Dice similarity coefficient (DSC) was 92% and 82% for the uterine cavity and placenta, respectively. For the combination of normal and abnormal cases, the DSC was 88% and 83% for the uterine cavity and placenta, respectively. The 3D segmentation algorithm estimated the volume of the normal and abnormal uterine cavity and placenta with average volume estimation errors of 4% and 9%, respectively.

Conclusions: The deep learning-based segmentation method provides a useful tool for volume estimation and analysis of the placenta and uterus cavity in human placental imaging.

Purpose: Handling low-quality and few-feature medical images is a challenging task in automatic panorama mosaicking. Current mosaicking methods for disordered input images are based on feature point matching, whereas in this case intensity-based registration achieves better performance than feature-point registration methods. We propose a mosaicking method that enables the use of mutual information (MI) registration for mosaicking randomly ordered input images with insufficient features.

Approach: Dimensionality reduction is used to map disordered input images into a low dimensional space. Based on the low dimensional representation, the image global correspondence can be recognized efficiently. For adjacent image pairs, we optimize the MI metric for registration. The panorama is then created after image blending. We demonstrate our method on relatively lower-cost handheld devices that acquire images from the retina in vivo, kidney ex vivo, and bladder phantom, all of which contain sparse features.

Results: Our method is compared with three baselines: AutoStitch, “dimension reduction + SIFT,” and “MI-Only.” Our method compared to the first two feature-point based methods exhibits 1.25 (ex vivo microscope dataset) to two times (in vivo retina dataset) rate of mosaic completion, and MI-Only has the lowest complete rate among three datasets. When comparing the subsequent complete mosaics, our target registration errors can be 2.2 and 3.8 times reduced when using the microscopy and bladder phantom datasets.

Conclusions: Using dimensional reduction increases the success rate of detecting adjacent images, which makes MI-based registration feasible and narrows the search range of MI optimization. To the best of our knowledge, this is the first mosaicking method that allows automatic stitching of disordered images with intensity-based alignment, which provides more robust and accurate results when there are insufficient features for classic mosaicking methods.

Purpose: We introduce and evaluate deep learning methods for weakly supervised segmentation of tumor lesions in whole-body fluorodeoxyglucose-positron emission tomography (FDG-PET) based solely on binary global labels (“tumor” versus “no tumor”).

Approach: We propose a three-step approach based on (i) a deep learning framework for image classification, (ii) subsequent generation of class activation maps (CAMs) using different CAM methods (CAM, GradCAM, GradCAM++, ScoreCAM), and (iii) final tumor segmentation based on the aforementioned CAMs. A VGG-based classification neural network was trained to distinguish between PET image slices with and without FDG-avid tumor lesions. Subsequently, the CAMs of this network were used to identify the tumor regions within images. This proposed framework was applied to FDG-PET/CT data of 453 oncological patients with available manually generated ground-truth segmentations. Quantitative segmentation performance was assessed for the different CAM approaches and compared with the manual ground truth segmentation and with supervised segmentation methods. In addition, further biomarkers (MTV and TLG) were extracted from the segmentation masks.

Results: A weakly supervised segmentation of tumor lesions was feasible with satisfactory performance [best median Dice score 0.47, interquartile range (IQR) 0.35] compared with a fully supervised U-Net model (median Dice score 0.72, IQR 0.36) and a simple threshold based segmentation (Dice score 0.29, IQR 0.28). CAM, GradCAM++, and ScoreCAM yielded similar results. However, GradCAM led to inferior results (median Dice score: 0.12, IQR 0.21) and was likely to ignore multiple instances within a given slice. CAM, GradCAM++, and ScoreCAM yielded accurate estimates of metabolic tumor volume (MTV) and tumor lesion glycolysis. Again, worse results were observed for GradCAM.

Conclusions: This work demonstrated the feasibility of weakly supervised segmentation of tumor lesions and accurate estimation of derived metrics such as MTV and tumor lesion glycolysis.

Purpose: To differentiate oncocytoma and chromophobe renal cell carcinoma (RCC) using radiomics features computed from spherical samples of image regions of interest, “radiomic biopsies” (RBs).

Approach: In a retrospective cohort study of 102 CT cases [68 males (67%), 34 females (33%); mean age ± SD, 63  ±  12  years], we pathology-confirmed 42 oncocytomas (41%) and 60 chromophobes (59%). A board-certified radiologist performed two RB rounds. From each RB round, we computed radiomics features and compared the performance of a random forest and AdaBoost binary classifier trained from the features. To control for overfitting, we performed 10 rounds of 70% to 30% train-test splits with feature-selection, cross-validation, and hyperparameter-optimization on each split. We evaluated the performance with test ROC AUC. We tested models on data from the other RB round and compared with the same round testing with the DeLong test. We clustered important features for each round and measured a bootstrapped adjusted Rand index agreement.

Results: Our best classifiers achieved an average AUC of 0.71  ±  0.024. We found no evidence of an effect for RB round (p  =  1). We also found no evidence for a decrease in model performance when tested on the other RB round (p  =  0.85). Feature clustering produced seven clusters in each RB round with high agreement (Rand index  =  0.981  ±  0.002, p  <  0.00001).

Conclusions: A consistent radiomic signature can be derived from RBs and could help distinguish oncocytoma and chromophobe RCC.

Purpose: In conventional diagnosis, the visual inspection of the malaria parasite Plasmodium falciparum in infected red blood cells under a microscope, is done manually by pathologists, which is both laborious and error-prone. Recent studies on automating this process have been conducted using artificial intelligence and feature selection of positional and morphological features from blood smear cell images using convolutional neural network (CNN). However, most deep CNN models do not perform well as per the expectation on small datasets.

Approach: In this context, we propose a comprehensive computer-aided diagnosis scheme for automating the detection of malaria parasites in thin blood smear images using deep CNN, where transfer learning is used for optimizing the feature selection process. As an extra layer of security, layer embeddings are extracted from the intermediate convolutional layers using the feature matrix to cross-check the selection of features in the intermediate layers. The proposal includes the utilization of the ResNet 152 model integrated with the Deep Greedy Network for training, which produces an enhanced quality of prediction.

Results: The performance of the proposed hybrid model has been evaluated concerning the evaluation metrics such as accuracy, precision, recall, specificity, and F1-score, which has been further compared with the pre-existing deep learning algorithms.

Conclusions: The comparative analysis of the results reported based on the accuracy metrics demonstrates promising outcomes concerning the other models. Lastly, the embedding extraction from the intermediate hidden layers and their visual analysis also provides an opportunity for manual verification of the performance of the trained model.

Purpose: To develop an imaging-based 3D catheter navigation system for transbronchial procedures including biopsy and tumor ablation using a single-plane C-arm x-ray system. The proposed system provides time-resolved catheter shape and position as well as motion compensated 3D airway roadmaps.

Approach: A continuous-sweep limited angle (CLA) imaging mode where the C-arm continuously rotates back and forth within a limited angular range while acquiring x-ray images was used for device tracking. The catheter reconstruction was performed using a sliding window of the most recent x-ray images, which captures information on device shape and position versus time. The catheter was reconstructed using a model-based approach and was displayed together with the 3D airway roadmap extracted from a pre-navigational cone-beam CT (CBCT). The roadmap was updated in regular intervals using deformable registration to tomosynthesis reconstructions based on the CLA images. The approach was evaluated in a porcine study (three animals) and compared to a gold standard CBCT reconstruction of the device.

Results: The average 3D root mean squared distance between CLA and CBCT reconstruction of the catheter centerline was 1  ±  0.5  mm for a stationary catheter and 2.9  ±  1.1  mm for a catheter moving at ∼1  cm  /  s. The average tip localization error was 1.3  ±  0.7  mm and 2.7  ±  1.8  mm, respectively.

Conclusions: The results indicate catheter navigation based on the proposed single plane C-arm imaging technique is feasible with reconstruction errors similar to the diameter of a typical ablation catheter.

Purpose: One major challenge facing simultaneous positron emission tomography (PET)/ magnetic resonance imaging (MRI) is PET attenuation correction (AC) measurement and evaluation of its accuracy. There is a crucial need for the evaluation of current and emergent PET AC methodologies in terms of absolute quantitative accuracy in the reconstructed PET images.

Approach: To address this need, we developed and evaluated a lesion insertion tool for PET/MRI that will facilitate this evaluation process. This tool was developed for the Biograph mMR and evaluated using phantom and patient data. Contrast recovery coefficients (CRC) from the NEMA IEC phantom of synthesized lesions were compared to measurements. In addition, SUV biases of lesions inserted in human brain and pelvis images were assessed from PET images reconstructed with MRI-based AC (MRAC) and CT-based AC (CTAC).

Results: For cross-comparison PET/MRI scanners AC evaluation, we demonstrated that the developed lesion insertion tool can be harmonized with the GE-SIGNA lesion insertion tool. About <3  %   CRC curves difference between simulation and measurement was achieved. An average of 1.6% between harmonized simulated CRC curves obtained with mMR and SIGNA lesion insertion tools was achieved. A range of −5  %   to 12% MRAC to CTAC SUV bias was respectively achieved in the vicinity and inside bone tissues in patient images in two anatomical regions, the brain, and pelvis.

Conclusions: A lesion insertion tool was developed for the Biograph mMR PET/MRI scanner and harmonized with the SIGNA PET/MRI lesion insertion tool. These tools will allow for an accurate evaluation of different PET/MRI AC approaches and permit exploration of subtle attenuation correction differences across systems.

Purpose: Assessing cardiotoxicity as a result of breast cancer therapeutics is increasingly important as breast cancer diagnoses are trending younger and overall survival is increasing. With evidence showing that prevention of cardiotoxicity plays a significant role in increasing overall survival, there is an unmet need for accurate non-invasive methods to assess cardiac injury due to cancer therapies. Current clinical methods are too coarse and emerging research methods have not yet achieved clinical implementation.

Approach: As a proof of concept, we examine myocardial elasticity imaging in the setting of premenopausal women diagnosed with hormone receptor positive (HR-positive) breast cancer undergoing severe estrogen depletion, as cardiovascular injury from early estrogen depletion is well-established. We evaluate the ability of our model-based cardiac elasticity imaging analysis method to indicate subclinical cancer therapy-related cardiac decline by examining differences in the change in cardiac elasticity over time in two cohorts of premenopausal women either undergoing severe estrogen depletion for HR-positive breast cancer or triple negative breast cancer patients as comparators.

Results: Our method was capable of producing functional mechanical elasticity maps of the left ventricle (LV). Using these elasticity maps, we show significant differences in cardiac mechanical elasticity in the HR-positive breast cancer cohort compared to the comparator cohort.

Conclusions: We present our methodology to assess the mechanical stiffness of the LV by interrogating cardiac magnetic resonance images within a computational biomechanical model. Our preliminary study suggests the potential of this method for examining cardiac tissue mechanical stiffness properties as an early indicator of cardiac decline.

Purpose: Whole slide imaging (WSI) scanners produce tissue slide images with a large field of view and a high resolution for pathologists to use in diagnoses. Color performance tests of these color imaging devices are necessary and can use stained tissue slides if the color truth is established using a hyperspectral imaging microscopy system (HIMS). The purpose of this study was to estimate the reproducibility uncertainty of CIELAB coordinates for a reference tissue slide measured by both the HIMS and a WSI scanner.

Approach: We compared the color performances of the WSI scanner to those of the reference established by the HIMS using the International Commission on Illumination (Commission Internationale de l’Éclairage, or CIE) 1976 ΔEab* color difference with the just noticeable color difference (JNCD, ΔEab*≤2), and the results from the overlap of the CIELAB coordinates’ uncertainty within the error bar, with a coverage factor k  =  2. The reported uncertainty results from measurements and image registration uncertainties.

Results: For the blank area common to the HIMS and the WSI average images, the color agreement was higher using the JNCD condition versus the CIELAB uncertainty overlap criterion (82% and 20% of the pixels in the images, respectively). This difference is explained by the fact that numerous pixels have CIELAB coordinates near one another but corresponding to CIELAB uncertainty values small enough not to overlap. In the colored area of the images, the JNCD condition was met for 0.19% of the pixels in the images, compared with 4.3% for the CIELAB uncertainty overlap criterion.

Conclusions: The distribution of uncertainties on the CIELAB coordinates was broader for the HIMS compared with the WSI scanner. The WSI scanner had a systemic error in the color reproduction, which pointed to a potential inadequate color calibration of this device.