Evaluation of perivascular fat attenuation with coronary CT angiography in cardiac transplantation patients: an imaging biomarker candidate for prediction of cardiac mortality and re-transplantation.

OBJECTIVES: In cardiac transplant recipients, non-invasive allograft surveillance for identifying patients at risk for graft failure remains challenging. The fat attenuation index (FAI) of the perivascular adipose tissue in coronary computed tomography angiography (CCTA) predicts outcomes in coronary artery disease in non-transplanted hearts; however, it has not been evaluated in cardiac transplant patients. METHODS: We followed 39 cardiac transplant patients with two or more CCTAs obtained between 2010 and 2021. We performed FAI measurements around the proximal 4 cm segments of the left anterior descending (LAD), right coronary artery (RCA), and left circumflex artery (LCx) using a previously validated methodology. The FAI was analyzed at a threshold of - 30 to - 190 Hounsfield units. RESULTS: FAI measurements were completed in 113 CCTAs, obtained on two same-vendor CT models. Within each CCTA, the FAI values between coronary vessels were strongly correlated (RCA and LAD R = 0.67 (p < 0.0001), RCA and LCx R = 0.58 (p < 0.0001), LAD and LCx R = 0.67 (p < 0.0001)). The FAIs of each coronary vessel between the patient's first and last CCTA completed at 120 kV were also correlated (RCA R = 0.73 (p < 0.0001), LAD R = 0.81 (p < 0.0001), LCx R = 0.55 (p = 0.0069). Finally, a high mean FAI value of all three coronary vessels at baseline (mean ≥ - 71 HU) was predictive of cardiac mortality or re-transplantation, however, not predictive of all cause-mortality. CONCLUSION: High baseline FAI values may identify a higher-risk cardiac transplant population; thus, FAI may support the implementation of CCTA in post-transplant surveillance. KEY POINT: • Perivascular fat attenuation measured with coronary CT is feasible in cardiac transplant patients and may predict cardiac mortality or need for re-transplantation.

Efficacy and prognostic value of delta radiomics on dual-energy computed tomography for gastric cancer with neoadjuvant chemotherapy: a preliminary study.

BACKGROUND: A non-invasive tool for tumor regression grade (TRG) evaluation is urgently needed for gastric cancer (GC) treated with neoadjuvant chemotherapy (NAC). PURPOSE: To develop and validate a radiomics signature (RS) to evaluate TRG for locally advanced GC after NAC and assess its prognostic value. MATERIAL AND METHODS: A total of 103 patients with GC treated with NAC were retrospectively recruited from April 2018 to December 2019 and were randomly allocated into a training cohort (n = 69) and a validation cohort (n = 34). Delineation was performed on both mixed and iodine-uptake images based on dual-energy computed tomography (DECT). A total of 4094 radiomics features were extracted from the pre-NAC, post-NAC, and delta feature sets. Spearman correlation and the least absolute shrinkage and selection operator were used for dimensionality reduction. Multivariable logistic regression was used for TRG evaluation and generated the optimal RS. Kaplan-Meier survival analysis with the log-rank test was implemented in an independent cohort of 40 patients to validate the prognostic value of the optimal RS. RESULTS: Three, five, and six radiomics features were finally selected for the pre-NAC, post-NAC, and delta feature sets. The delta model demonstrated the best performance in assessing TRG in both the training and the validation cohorts (AUCs=0.91 and 0.76, respectively; P>0.1). The optimal RS from the delta model showed a significant capability to predict survival in the independent cohort (P<0.05). CONCLUSION: Delta radiomics based on DECT images serves as a potential biomarker for TRG evaluation and shows prognostic value for patients with GC treated with NAC.

Performance of a Deep Neural Network Algorithm Based on a Small Medical Image Dataset: Incremental Impact of 3D-to-2D Reformation Combined with Novel Data Augmentation, Photometric Conversion, or Transfer Learning.

Collecting and curating large medical-image datasets for deep neural network (DNN) algorithm development is typically difficult and resource-intensive. While transfer learning (TL) decreases reliance on large data collections, current TL implementations are tailored to two-dimensional (2D) datasets, limiting applicability to volumetric imaging (e.g., computed tomography). Targeting performance enhancement of a DNN algorithm based on a small image dataset, we assessed incremental impact of 3D-to-2D projection methods, one supporting novel data augmentation (DA); photometric grayscale-to-color conversion (GCC); and/or TL on training of an algorithm from a small coronary computed tomography angiography (CCTA) dataset (200 examinations, 50% with atherosclerosis and 50% atherosclerosis-free) producing 245 diseased and 1127 normal coronary arteries/branches. Volumetric CCTA data was converted to a 2D format creating both an Aggregate Projection View (APV) and a Mosaic Projection View (MPV), supporting DA per vessel; both grayscale and color-mapped versions of each view were also obtained. Training was performed both without and with TL, and algorithm performance of all permutations was compared using area under the receiver operating characteristics curve. Without TL, APV performance was 0.74 and 0.87 on grayscale and color images, respectively, compared to 0.90 and 0.87 for MPV. With TL, APV performance was 0.78 and 0.88 on grayscale and color images, respectively, compared with 0.93 and 0.91 for MPV. In conclusion, TL enhances performance of a DNN algorithm from a small volumetric dataset after proposed 3D-to-2D reformatting, but additive gain is achieved with application of either GCC to APV or the proposed novel MPV technique for DA.

Automatic detection of lytic and blastic thoracolumbar spine metastases on computed tomography.

OBJECTIVE: To evaluate a computer-aided detection (CADe) system for lytic and blastic spinal metastases on computed tomography (CT). METHODS: We retrospectively evaluated the CADe system on 20 consecutive patients with 42 lytic and on 30 consecutive patients with 172 blastic metastases. The CADe system was trained using CT images of 114 subjects with 102 lytic and 308 blastic spinal metastases. Lesions were annotated by experienced radiologists. Detected benign lesions were considered false-positive findings. Detector sensitivity and the number of false-positive findings were calculated as the criteria for detector performance, and free-response receiver operating characteristic (FROC) analysis was conducted. Detailed analysis of false-positive and false-negative findings was performed. RESULTS: Algorithm runtime is 3 ± 0.5 min per patient. The system achieves a sensitivity of 83 % at 3.5 false positives per patient on average for blastic metastases and a sensitivity of 88 % at 3.7 false positives for lytic metastases. False positives appeared predominantly in the area of degenerative changes in the case of the blastic metastasis detector and in osteoporotic areas in the case of the lytic metastasis detector. CONCLUSION: The CADe system reliably detects thoracolumbar spine metastases in real time. An additional study is planned to evaluate how the bone lesion CADe system improves radiologists' accuracy and efficiency in a clinical setting. KEY POINTS: • Computer-aided detection (CADe) of bone metastases has been developed for spinal CT. • The CADe system exhibits high sensitivity with a tolerable false-positive rate. • Analysis of false-positive detection may further improve the system. • CADe may reduce the number of missed spinal metastases at CT interpretation.

Computed Tomography-based Radiomics Evaluation of Postoperative Local Recurrence of Pancreatic Ductal Adenocarcinoma.

OBJECTIVE: To develop and validate an effective model for identifying patients with postoperative local disease recurrence of pancreatic ductal adenocarcinoma (PDAC). METHODS: A total of 153 patients who had undergone surgical resection of PDAC with regular postoperative follow-up were consecutively enrolled and randomly divided into training (n = 108) and validation (n = 45) cohorts. The postoperative soft-tissue biopsy results or clinical follow-up results served as the reference diagnostic criteria. Radiomics analysis of the postoperative soft-tissue was performed on a commercially available prototype software using portal vein phase image. Three models were built to characterize postoperative soft tissue: computed tomography (CT)-based radiomics, clinicoradiological, and their combination. The area under the receiver operating characteristic curves (AUC) was used to evaluate the differential diagnostic performance. A nomogram was used to select the final model with best performance. One radiologist's diagnostic choices that were made with and without the nomogram's assistance were evaluated. RESULTS: A seven-feature-combined radiomics signature was constructed as a predictor of postoperative local recurrence. The nomogram model combining the radiomics signature with postoperative CA 19-9 elevation showed the best performance (training cohort, AUC = 0.791 [95%CI: 0.707, 0.876]; validation cohort, AUC = 0.742 [95%CI: 0.590, 0.894]). In the validation cohort, the AUC for differential diagnosis was significantly improved for the combined model relative to that for postoperative CA 19-9 elevation (AUC = 0.742 vs. 0.533, p < 0.001). The calibration curve and decision curve analysis demonstrated the clinical usefulness of the proposed nomogram. The diagnostic performance of the radiologist was not significantly improve by using the proposed nomogram (AUC = 0.742 vs. 0.670, p = 0.17). CONCLUSION: The combined model using CT radiomic features and CA 19-9 elevation effectively characterized postoperative soft tissue and potentially may improve treatment strategies and facilitate personalized treatment for PDAC after surgical resection.

How scan parameter choice affects deep learning-based coronary artery disease assessment from computed tomography.

Recently, algorithms capable of assessing the severity of Coronary Artery Disease (CAD) in form of the Coronary Artery Disease-Reporting and Data System (CAD-RADS) grade from Coronary Computed Tomography Angiography (CCTA) scans using Deep Learning (DL) were proposed. Before considering to apply these algorithms in clinical practice, their robustness regarding different commonly used Computed Tomography (CT)-specific image formation parameters-including denoising strength, slab combination, and reconstruction kernel-needs to be evaluated. For this study, we reconstructed a data set of 500 patient CCTA scans under seven image formation parameter configurations. We select one default configuration and evaluate how varying individual parameters impacts the performance and stability of a typical algorithm for automated CAD assessment from CCTA. This algorithm consists of multiple preprocessing and a DL prediction step. We evaluate the influence of the parameter changes on the entire pipeline and additionally on only the DL step by propagating the centerline extraction results of the default configuration to all others. We consider the standard deviation of the CAD severity prediction grade difference between the default and variation configurations to assess the stability w.r.t. parameter changes. For the full pipeline we observe slight instability (± 0.226 CAD-RADS) for all variations. Predictions are more stable with centerlines propagated from the default to the variation configurations (± 0.122 CAD-RADS), especially for differing denoising strengths (± 0.046 CAD-RADS). However, stacking slabs with sharp boundaries instead of mixing slabs in overlapping regions (called true stack ± 0.313 CAD-RADS) and increasing the sharpness of the reconstruction kernel (± 0.150 CAD-RADS) leads to unstable predictions. Regarding the clinically relevant tasks of excluding CAD (called rule-out; AUC default 0.957, min 0.937) and excluding obstructive CAD (called hold-out; AUC default 0.971, min 0.964) the performance remains on a high level for all variations. Concluding, an influence of reconstruction parameters on the predictions is observed. Especially, scans reconstructed with the true stack parameter need to be treated with caution when using a DL-based method. Also, reconstruction kernels which are underrepresented in the training data increase the prediction uncertainty.

Human AI Teaming for Coronary CT Angiography Assessment: Impact on Imaging Workflow and Diagnostic Accuracy.

As the number of coronary computed tomography angiography (CTA) examinations is expected to increase, technologies to optimize the imaging workflow are of great interest. The aim of this study was to investigate the potential of artificial intelligence (AI) to improve clinical workflow and diagnostic accuracy in high-volume cardiac imaging centers. A total of 120 patients (79 men; 62.4 (55.0-72.7) years; 26.7 (24.9-30.3) kg/m2) undergoing coronary CTA were randomly assigned to a standard or an AI-based (human AI) coronary analysis group. Severity of coronary artery disease was graded according to CAD-RADS. Initial reports were reviewed and changes were classified. Both groups were similar with regard to age, sex, body mass index, heart rate, Agatston score, and CAD-RADS. The time for coronary CTA assessment (142.5 (106.5-215.0) s vs. 195.0 (146.0-265.5) s; p < 0.002) and the total reporting time (274.0 (208.0-377.0) s vs. 350 (264.0-445.5) s; p < 0.02) were lower in the human AI than in the standard group. The number of cases with no, minor, or CAD-RADS relevant changes did not differ significantly between groups (52, 7, 1 vs. 50, 8, 2; p = 0.80). AI-based analysis significantly improves clinical workflow, even in a specialized high-volume setting, by reducing CTA analysis and overall reporting time without compromising diagnostic accuracy.

Evaluation of dual-energy CT derived radiomics signatures in predicting outcomes in patients with advanced gastric cancer after neoadjuvant chemotherapy.

BACKGROUND: To investigate the prognostic value of dual-energy CT (DECT) based radiomics to predict disease-free survival (DFS) and overall survival (OS) for patients with advanced gastric cancer (AGC) after neoadjuvant chemotherapy (NAC). METHODS: From January 2014 to December 2018, a total of 156 AGC patients were enrolled and randomly allocated into a training cohort and a testing cohort at a ratio of 2:1. Volume of interest of primary tumor was delineated on eight image series. Four feature sets derived from pre-NAC and delta radiomics were generated for each survival arm. Random survival forest was used for generating the optimal radiomics signature (RS). Statistical metrics for model evaluation included Harrell's concordance index (C-index) and the average cumulative/dynamic AUC throughout follow-up. A clinical model and a combined Rad-clinical model were built for comparison. RESULTS: The pre-IU (derived from iodine uptake images before NAC) RS performed best for DFS and OS in the testing cohort (C-indices, 0.784 and 0.698; the average cumulative/dynamic AUCs, 0.80 and 0.77). When compared with the clinical model, the radiomics model had significantly higher C-index to predict DFS in the testing cohort (0.784 vs. 0.635, p < 0.001), but no statistical difference was found for OS (0.698 vs. 0.680, p = 0.473). The combined Rad-clinical models showed improved performance in the testing cohort, with C-indices of 0.810 and 0.710 for DFS and OS, respectively. CONCLUSION: DECT-derived radiomics serves as a promising non-invasive biomarker to predict survival for AGC patients after NAC, providing an opportunity for transforming proper treatment.

Automated Dual-energy Computed Tomography-based Extracellular Volume Estimation for Myocardial Characterization in Patients With Ischemic and Nonischemic Cardiomyopathy.

OBJECTIVES: We aimed to validate and test a prototype algorithm for automated dual-energy computed tomography (DECT)-based myocardial extracellular volume (ECV) assessment in patients with various cardiomyopathies. METHODS: This retrospective study included healthy subjects (n=9; 61±10 y) and patients with cardiomyopathy (n=109, including a validation cohort n=60; 68±9 y; and a test cohort n=49; 69±11 y), who had previously undergone cardiac DECT. Myocardial ECV was calculated using a prototype-based fully automated algorithm and compared with manual assessment. Receiver-operating characteristic analysis was performed to test the algorithm's ability to distinguish healthy subjects and patients with cardiomyopathy. RESULTS: The fully automated method led to a significant reduction of postprocessing time compared with manual assessment (2.2±0.4 min and 9.4±0.7 min, respectively, P <0.001). There was no significant difference in ECV between the automated and manual methods ( P =0.088). The automated method showed moderate correlation and agreement with the manual technique ( r =0.68, intraclass correlation coefficient=0.66). ECV was significantly higher in patients with cardiomyopathy compared with healthy subjects, regardless of the method used ( P <0.001). In the test cohort, the automated method yielded an area under the curve of 0.98 for identifying patients with cardiomyopathies. CONCLUSION: Automated ECV estimation based on DECT showed moderate agreement with the manual method and matched with previously reported ECV values for healthy volunteers and patients with cardiomyopathy. The automatically derived ECV demonstrated an excellent diagnostic performance to discriminate between healthy and diseased myocardium, suggesting that it could be an effective initial screening tool while significantly reducing the time of assessment.

CT-Based Radiomics Showing Generalization to Predict Tumor Regression Grade for Advanced Gastric Cancer Treated With Neoadjuvant Chemotherapy.

Objective: The aim of this study was to develop and validate a radiomics model to predict treatment response in patients with advanced gastric cancer (AGC) sensitive to neoadjuvant therapies and verify its generalization among different regimens, including neoadjuvant chemotherapy (NAC) and molecular targeted therapy. Materials and Methods: A total of 373 patients with AGC receiving neoadjuvant therapies were enrolled from five cohorts. Four cohorts of patients received different regimens of NAC, including three retrospective cohorts (training cohort and internal and external validation cohorts) and a prospective Dragon III cohort (NCT03636893). Another prospective SOXA (apatinib in combination with S-1 and oxaliplatin) cohort received neoadjuvant molecular targeted therapy (ChiCTR-OPC-16010061). All patients underwent computed tomography before treatment, and thereafter, tumor regression grade (TRG) was assessed. The primary tumor was delineated, and 2,452 radiomics features were extracted for each patient. Mutual information and random forest were used for dimensionality reduction and modeling. The performance of the radiomics model to predict TRG under different neoadjuvant therapies was evaluated. Results: There were 28 radiomics features selected. The radiomics model showed generalization to predict TRG for AGC patients across different NAC regimens, with areas under the curve (AUCs) (95% interval confidence) of 0.82 (0.76~0.90), 0.77 (0.63~0.91), 0.78 (0.66~0.89), and 0.72 (0.66~0.89) in the four cohorts, with no statistical difference observed (all p > 0.05). However, the radiomics model showed poor predictive value on the SOXA cohort [AUC, 0.50 (0.27~0.73)], which was significantly worse than that in the training cohort (p = 0.010). Conclusion: Radiomics is generalizable to predict TRG for AGC patients receiving NAC treatments, which is beneficial to transform appropriate treatment, especially for those insensitive to NAC.

Dual-Energy Computed Tomography-Based Radiomics to Predict Peritoneal Metastasis in Gastric Cancer.

OBJECTIVE: To develop and validate a dual-energy computed tomography (DECT) derived radiomics model to predict peritoneal metastasis (PM) in patients with gastric cancer (GC). METHODS: This retrospective study recruited 239 GC (non-PM = 174, PM = 65) patients with histopathological confirmation for peritoneal status from January 2015 to December 2019. All patients were randomly divided into a training cohort (n = 160) and a testing cohort (n = 79). Standardized iodine-uptake (IU) images and 120-kV-equivalent mixed images (simulating conventional CT images) from portal-venous and delayed phases were used for analysis. Two regions of interest (ROIs) including the peritoneal area and the primary tumor were independently delineated. Subsequently, 1691 and 1226 radiomics features were extracted from the peritoneal area and the primary tumor from IU and mixed images on each phase. Boruta and Spearman correlation analysis were used for feature selection. Three radiomics models were established, including the R_IU model for IU images, the R_MIX model for mixed images and the combined radiomics model (the R_comb model). Random forest was used to tune the optimal radiomics model. The performance of the clinical model and human experts to assess PM was also recorded. RESULTS: Fourteen and three radiomics features with low redundancy and high importance were extracted from the IU and mixed images, respectively. The R_IU model showed significantly better performance to predict PM than the R_MIX model in the training cohort (AUC, 0.981 vs. 0.917, p = 0.034). No improvement was observed in the R_comb model (AUC = 0.967). The R_IU model was the optimal radiomics model which showed no overfitting in the testing cohort (AUC = 0.967, p = 0.528). The R_IU model demonstrated significantly higher predictive value on peritoneal status than the clinical model and human experts in the testing cohort (AUC, 0.785, p = 0.005; AUC, 0.732, p <0.001, respectively). CONCLUSION: DECT derived radiomics could serve as a non-invasive and easy-to-use biomarker to preoperatively predict PM for GC, providing opportunity for those patients to tailor appropriate treatment.

Publisher Correction: The Technome - A Predictive Internal Calibration Approach for Quantitative Imaging Biomarker Research.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

The Technome - A Predictive Internal Calibration Approach for Quantitative Imaging Biomarker Research.

The goal of radiomics is to convert medical images into a minable data space by extraction of quantitative imaging features for clinically relevant analyses, e.g. survival time prediction of a patient. One problem of radiomics from computed tomography is the impact of technical variation such as reconstruction kernel variation within a study. Additionally, what is often neglected is the impact of inter-patient technical variation, resulting from patient characteristics, even when scan and reconstruction parameters are constant. In our approach, measurements within 3D regions-of-interests (ROI) are calibrated by further ROIs such as air, adipose tissue, liver, etc. that are used as control regions (CR). Our goal is to derive general rules for an automated internal calibration that enhance prediction, based on the analysed features and a set of CRs. We define qualification criteria motivated by status-quo radiomics stability analysis techniques to only collect information from the CRs which is relevant given a respective task. These criteria are used in an optimisation to automatically derive a suitable internal calibration for prediction tasks based on the CRs. Our calibration enhanced the performance for centrilobular emphysema prediction in a COPD study and prediction of patients' one-year-survival in an oncological study.

A User Interface for Optimizing Radiologist Engagement in Image Data Curation for Artificial Intelligence.

PURPOSE: To delineate image data curation needs and describe a locally designed graphical user interface (GUI) to aid radiologists in image annotation for artificial intelligence (AI) applications in medical imaging. MATERIALS AND METHODS: GUI components support image analysis toolboxes, picture archiving and communication system integration, third-party applications, processing of scripting languages, and integration of deep learning libraries. For clinical AI applications, GUI components included two-dimensional segmentation and classification; three-dimensional segmentation and quantification; and three-dimensional segmentation, quantification, and classification. To assess radiologist engagement and performance efficiency associated with GUI-related capabilities, image annotation rate (studies per day) and speed (minutes per case) were evaluated in two clinical scenarios of varying complexity: hip fracture detection and coronary atherosclerotic plaque demarcation and stenosis grading. RESULTS: For hip fracture, 1050 radiographs were annotated over 7 days (150 studies per day; median speed: 10 seconds per study [interquartile range, 3-21 seconds per study]). A total of 294 coronary CT angiographic studies with 1843 arteries and branches were annotated for atherosclerotic plaque over 23 days (15.2 studies [80.1 vessels] per day; median speed: 6.08 minutes per study [interquartile range, 2.8-10.6 minutes per study] and 73 seconds per vessel [interquartile range, 20.9-155 seconds per vessel]). CONCLUSION: GUI-component compatibility with common image analysis tools facilitates radiologist engagement in image data curation, including image annotation, supporting AI application development and evolution for medical imaging. When complemented by other GUI elements, a continuous integrated workflow supporting formation of an agile deep neural network life cycle results.Supplemental material is available for this article.© RSNA, 2019.

Estimating a patient surface model for optimizing the medical scanning workflow.

In this paper, we present the idea of equipping a tomographic medical scanner with a range imaging device (e.g. a 3D camera) to improve the current scanning workflow. A novel technical approach is proposed to robustly estimate patient surface geometry by a single snapshot from the camera. Leveraging the information of the patient surface geometry can provide significant clinical benefits, including automation of the scan, motion compensation for better image quality, sanity check of patient movement, augmented reality for guidance, patient specific dose optimization, and more. Our approach overcomes the technical difficulties resulting from suboptimal camera placement due to practical considerations. Experimental results on more than 30 patients from a real CT scanner demonstrate the robustness of our approach.

Fast and robust 3D vertebra segmentation using statistical shape models.

We propose a top-down fully automatic 3D vertebra segmentation algorithm using global shape-related as well as local appearance-related prior information. The former is brought into the system by a global statistical shape model built from annotated training data, i.e., annotated CT volumes. The latter is handled by a machine learning-based component, i.e., a boundary detector, providing a strong discriminative model for vertebra surface appearance by making use of local context-encoding features. This boundary detector, which is essentially a probabilistic boosting-tree classifier, is also learnt from annotated training data. Contextual information is taken into account by representing vertebra surface candidate voxels with high-dimensional vectors of 3D steerable features derived from the observed volume intensities. Our system does not only consider the body of the individual vertebrae but also the spinal processes. Before segmentation, the image parts depicting individual vertebrae are spatially normalized with respect to their bounding box information in terms of translation, orientation, and scale leading to more accurate results. We evaluate segmentation accuracy on 7 CT volumes each depicting 22 vertebrae. The results indicate a symmetric point-to-mesh surface error of 1.37 ± 0.37 mm, which matches the current state-of-the-art.

Image-based Co-Registration of Angiography and Intravascular Ultrasound Images.

In image-guided cardiac interventions, X-ray imaging and intravascular ultrasound (IVUS) imaging are two often used modalities. Interventional X-ray images, including angiography and fluoroscopy, are used to assess the lumen of the coronary arteries and to monitor devices in real time. IVUS provides rich intravascular information, such as vessel wall composition, plaque, and stent expansions, but lacks spatial orientations. Since the two imaging modalities are complementary to each other, it is highly desirable to co-register the two modalities to provide a comprehensive picture of the coronaries for interventional cardiologists. In this paper, we present a solution for co-registering 2-D angiography and IVUS through image-based device tracking. The presented framework includes learning-based vessel detection and device detections, model-based tracking, and geodesic distance-based registration. The system first interactively detects the coronary branch under investigation in a reference angiography image. During the pullback of the IVUS transducers, the system acquires both ECG-triggered fluoroscopy and IVUS images, and automatically tracks the position of the medical devices in fluoroscopy. The localization of tracked IVUS transducers and guiding catheter tips is used to associate an IVUS imaging plane to a corresponding location on the vessel branch under investigation. The presented image-based solution can be conveniently integrated into existing cardiology workflow. The system is validated with a set of clinical cases, and achieves good accuracy and robustness.

Spine detection in CT and MR using iterated marginal space learning.

Examinations of the spinal column with both, Magnetic Resonance (MR) imaging and Computed Tomography (CT), often require a precise three-dimensional positioning, angulation and labeling of the spinal disks and the vertebrae. A fully automatic and robust approach is a prerequisite for an automated scan alignment as well as for the segmentation and analysis of spinal disks and vertebral bodies in Computer Aided Diagnosis (CAD) applications. In this article, we present a novel method that combines Marginal Space Learning (MSL), a recently introduced concept for efficient discriminative object detection, with a generative anatomical network that incorporates relative pose information for the detection of multiple objects. It is used to simultaneously detect and label the spinal disks. While a novel iterative version of MSL is used to quickly generate candidate detections comprising position, orientation, and scale of the disks with high sensitivity, the anatomical network selects the most likely candidates using a learned prior on the individual nine dimensional transformation spaces. Finally, we propose an optional case-adaptive segmentation approach that allows to segment the spinal disks and vertebrae in MR and CT respectively. Since the proposed approaches are learning-based, they can be trained for MR or CT alike. Experimental results based on 42 MR and 30 CT volumes show that our system not only achieves superior accuracy but also is among the fastest systems of its kind in the literature. On the MR data set the spinal disks of a whole spine are detected in 11.5s on average with 98.6% sensitivity and 0.073 false positive detections per volume. On the CT data a comparable sensitivity of 98.0% with 0.267 false positives is achieved. Detected disks are localized with an average position error of 2.4 mm/3.2 mm and angular error of 3.9°/4.5° in MR/CT, which is close to the employed hypothesis resolution of 2.1 mm and 3.3°.

Data-driven breast decompression and lesion mapping from digital breast tomosynthesis.

Digital breast tomosynthesis (DBT) emerges as a new 3D modality for breast cancer screening and diagnosis. Like in conventional 2D mammography the breast is scanned in a compressed state. For orientation during surgical planning, e.g., during presurgical ultrasound-guided anchor-wire marking, as well as for improving communication between radiologists and surgeons it is desirable to estimate an uncompressed model of the acquired breast along with a spatial mapping that allows localizing lesions marked in DBT in the uncompressed model. We therefore propose a method for 3D breast decompression and associated lesion mapping from 3D DBT data. The method is entirely data-driven and employs machine learning methods to predict the shape of the uncompressed breast from a DBT input volume. For this purpose a shape space has been constructed from manually annotated uncompressed breast surfaces and shape parameters are predicted by multiple multi-variate Random Forest regression. By exploiting point correspondences between the compressed and uncompressed breasts, lesions identified in DBT can be mapped to approximately corresponding locations in the uncompressed breast model. To this end, a thin-plate spline mapping is employed. Our method features a novel completely data-driven approach to breast shape prediction that does not necessitate prior knowledge about biomechanical properties and parameters of the breast tissue. Instead, a particular deformation behavior (decompression) is learned from annotated shape pairs, compressed and uncompressed, which are obtained from DBT and magnetic resonance image volumes, respectively. On average, shape prediction takes 26s and achieves a surface distance of 15.80 +/- 4.70 mm. The mean localization error for lesion mapping is 22.48 +/- 8.67 mm.