Search

Accurate estimation of motion field in respiration-correlated 4DCT images, is a precondition for the analysis of patient-specific breathing dynamics and subsequent image-supported treatment planning. However, the lung motion estimation often suffers from the sliding motion. In this paper, a novel lung motion method based on the non-rigid registration of point clouds is proposed, and the tangent-plane distance is used to represent the distance term, which describes the difference between two point clouds. Local affine transformation model is used to express the non-rigid deformation of the lung motion. The final objective function is expressed in the Frobenius norm formation, and matrix optimization scheme is carried out to find out the optimal transformation parameters that minimize the objective function. A key advantage of our proposed method is that it alleviates the requirement that the source point cloud and the reference point cloud should be in one-to-one corresponding relationship, and the requirement is difficult to be satisfied in practical application. Furthermore, the proposed method takes the sliding motion of the lung into consideration and improves the registration accuracy by reducing the constraint of the motion along the tangent direction. Non-rigid registration experiments are carried out to validate the performance of the proposed method using popi-model data. The results demonstrate that the proposed method outperforms the traditional method with about 20% accuracy increase. [ABSTRACT FROM AUTHOR]

Glioma detection and segmentation is a challenging task for radiologists and clinicians. The research reported in this paper seeks to develop a better clinical decision support algorithm for clinicians diagnosis. This paper presents a probabilistic method for detection and segmentation between abnormal tissue regions and brain tumour (tumour core and edema) portions from Magnetic Resonance Imaging (MRI). A framework is constructed to learn structure of undirected graphical models that can represent the spatial relationships among variables and apply it to glioma segmentation. Compared with the pixel of image, the superpixel is more consistent with human visual cognition and contains less redundancy, thus, the superpixels are considered as the basic unit of structure learning and glioma segmentation scheme. ℓ1-regularization techniques are applied to learn the appropriate structure for modeling graphical models. Conditional Random Fields (CRF) are used to model the spatial interactions among image superpixel regions and their measurements. A number of features including statistics features, the combined features from the local binary pattern as well as gray level run length, curve features, and fractal features were extracted from each superpixel. The features are then passed by ℓ1-regularization to ensure a robust classification. The proposed method is compared with support vector machine and Fuzzy c-means to classify each superpixel into normal and abnormal tissue. The proposed system is tested for the presence of low grade as well as high grade glioma tumors on images collected from BRATS2013, BRATS2015 data set and Henan Provincial People’s Hospital (HNPPH) data set. The experiments performed provides similarity between segmented and truth image up to 91.5% by correlation method. [ABSTRACT FROM AUTHOR]

Purpose: Many studies have reported that carbohydrate antigen 153 (CA153) in breast secretions (BS) can discriminate breast cancer (BC) patients from healthy individuals, indicating CA153 in BS as a potential index for BC. This meta-analysis aimed to evaluate the actual diagnostic value of CA153 in BS. Methods: Related papers were obtained from Pubmed, Embase, Scopus, Ovid, Sciverse, the Cochrane library, Chinese Biomedical literature Database (CBM), Technology of Chongqing (VIP), Wan Fang Data, and Chinese National Knowledge Infrastructure (CNKI). Pooled sensitivity, specificity, and diagnostic odds ratio (DOR) of CA153 in BS for BC diagnosis were analyzed with the random effect model. SROC and the area under the curve (AUC) were applied to assess overall diagnostic efficiency. Results: This meta-analysis included five studies with a total of 329 BC patients and 381 healthy subjects. For CA153 in BS, the summary sensitivity, specificity, and DOR to diagnose BC were 0.63 (95% confidence interval (CI): 0.57∼0.68), 0.82 (95% CI: 0.78∼0.86), and 9.18 (95% CI: 4.22∼19.95), respectively. Furthermore, the AUC of BS CA153 in the diagnosis of BC was 0.8614. Conclusions: CA153 in BS is a valuable molecular marker in diagnosing BC and should be applied in standard clinical practices of BC screening upon confirmation of our findings in a larger prospective study. [ABSTRACT FROM AUTHOR]

Content-based image retrieval (CBIR) techniques have currently gained increasing popularity in the medical field because they can use numerous and valuable archived images to support clinical decisions. In this paper, we concentrate on developing a CBIR system for retrieving brain tumors in T1-weighted contrast-enhanced MRI images. Specifically, when the user roughly outlines the tumor region of a query image, brain tumor images in the database of the same pathological type are expected to be returned. We propose a novel feature extraction framework to improve the retrieval performance. The proposed framework consists of three steps. First, we augment the tumor region and use the augmented tumor region as the region of interest to incorporate informative contextual information. Second, the augmented tumor region is split into subregions by an adaptive spatial division method based on intensity orders; within each subregion, we extract raw image patches as local features. Third, we apply the Fisher kernel framework to aggregate the local features of each subregion into a respective single vector representation and concatenate these per-subregion vector representations to obtain an image-level signature. After feature extraction, a closed-form metric learning algorithm is applied to measure the similarity between the query image and database images. Extensive experiments are conducted on a large dataset of 3604 images with three types of brain tumors, namely, meningiomas, gliomas, and pituitary tumors. The mean average precision can reach 94.68%. Experimental results demonstrate the power of the proposed algorithm against some related state-of-the-art methods on the same dataset. [ABSTRACT FROM AUTHOR]

How can we predict the occurrence of acute kidney injury (AKI) in cancer patients based on machine learning with serum creatinine data? Given irregular and heterogeneous clinical data, how can we make the most of it for accurate AKI prediction? AKI is a common and significant complication in cancer patients, and correlates with substantial morbidity and mortality. Since no effective treatment for AKI still exists, it is important to take timely preventive measures. While several approaches have been proposed for predicting AKI, their scope and applicability are limited as they either assume regular data measured over a short hospital stay, or do not fully utilize heterogeneous data. In this paper, we provide an AKI prediction model with a greater applicability, which relaxes the constraints of existing approaches, and fully utilizes irregular and heterogeneous data for learning the model. In a cohort of 21,022 cancer patients who were registered into Korea Central Cancer Registry (KCCR) in Seoul National University Hospital between January 1, 2004 and December 31, 2013, our method achieves 0.7892 precision, 0.7506 recall, and 0.7576 F-measure in predicting whether a patient will develop AKI during the next 14 days. [ABSTRACT FROM AUTHOR]

As an important way to help express the preference relation between alternatives, distributed preference relation (DPR) can represent the preferred, non-preferred, indifferent, and uncertain degrees of one alternative over another simultaneously. DPR, however, is unavailable in some situations where a decision maker cannot provide the precise degrees of one alternative over another due to lack of knowledge, experience, and data. In this paper, to address this issue, we propose interval-valued DPR (IDPR) and present its properties of validity and normalization. Through constructing two optimization models, an IDPR matrix is transformed into a score matrix to facilitate the comparison between any two alternatives. The properties of the score matrix are analyzed. To guarantee the rationality of the comparisons between alternatives derived from the score matrix, the additive consistency of the score matrix is developed. In terms of these, IDPR is applied to model and solve multiple criteria group decision making (MCGDM) problem. Particularly, the relationship between the parameters for the consistency of the score matrix associated with each decision maker and those for the consistency of the score matrix associated with the group of decision makers is analyzed. A manager selection problem is investigated to demonstrate the application of IDPRs to MCGDM problems. [ABSTRACT FROM AUTHOR]

This paper presents a novel two-stage image segmentation method using an edge scaled energy functional based on local and global information for intensity inhomogeneous image segmentation. In the first stage, we integrate global intensity term with a geodesic edge term, which produces a preliminary rough segmentation result. Thereafter, by taking final contour of the first stage as initial contour, we begin second stage segmentation process by integrating local intensity term with geodesic edge term to get final segmentation result. Due to the suitable initialization from the first stage, the second stage precisely achieves desirable segmentation result for inhomogeneous image segmentation. Two stage segmentation technique not only increases the accuracy but also eliminates the problem of initial contour existed in traditional local segmentation methods. The energy function of the proposed method uses both global and local terms incorporated with compacted geodesic edge term in an additive fashion which uses image gradient information to delineate obscured boundaries of objects inside an image. A Gaussian kernel is adapted for the regularization of the level set function and to avoid an expensive re-initialization. The experiments were carried out on synthetic and real images. Quantitative validations were performed on Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) 2015 and PH2 skin lesion database. The visual and quantitative comparisons will demonstrate the efficiency of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper we address the problem of automated grading of invasive breast carcinoma through the encoding of histological images as VLAD (Vector of Locally Aggregated Descriptors) representations on the Grassmann manifold. The proposed method considers each image as a set of multidimensional spatially-evolving signals that can be efficiently modeled through a higher-order linear dynamical systems analysis. Subsequently, each H&E (Hematoxylin and Eosin) stained breast cancer histological image is represented as a cloud of points on the Grassmann manifold, while a vector representation approach is applied aiming to aggregate the Grassmannian points based on a locality criterion on the manifold. To evaluate the efficiency of the proposed methodology, two datasets with different characteristics were used. More specifically, we created a new medium-sized dataset consisting of 300 annotated images (collected from 21 patients) of grades 1, 2 and 3, while we also provide experimental results using a large dataset, namely BreaKHis, containing 7,909 breast cancer histological images, collected from 82 patients, of both benign and malignant cases. Experimental results have shown that the proposed method outperforms a number of state of the art approaches providing average classification rates of 95.8% and 91.38% with our dataset and the BreaKHis dataset, respectively. [ABSTRACT FROM AUTHOR]

Blood oxygenation mapping by the means of optical oximetry is of significant importance in clinical trials. This paper uses hyperspectral imaging technology to obtain in vivo images for blood oxygenation detection. The experiment involves dorsal skin fold window chamber preparation which was built on adult (8–10 weeks of age) female BALB/c nu/nu mice and in vivo image acquisition which was performed by hyperspectral imaging system. To get the accurate spatial and spectral information of targets, an automatic registration scheme is proposed. An adaptive feature detection method which combines the local threshold method and the level-set filter is presented to extract target vessels. A reliable feature matching algorithm with the correlative information inherent in hyperspectral images is used to kick out the outliers. Then, the registration images are used for blood oxygenation mapping. Registration evaluation results show that most of the false matches are removed and the smooth and concentrated spectra are obtained. This intensity invariant feature detection with outliers-removing feature matching proves to be effective in hyperspectral vessel image registration. Therefore, in vivo hyperspectral imaging system by the assistance of the proposed registration scheme provides a technique for blood oxygenation research. [ABSTRACT FROM AUTHOR]

Stain colour estimation is a prominent factor of the analysis pipeline in most of histology image processing algorithms. Providing a reliable and efficient stain colour deconvolution approach is fundamental for robust algorithm. In this paper, we propose a novel method for stain colour deconvolution of histology images. This approach statistically analyses the multi-resolutional representation of the image to separate the independent observations out of the correlated ones. We then estimate the stain mixing matrix using filtered uncorrelated data. We conducted an extensive set of experiments to compare the proposed method to the recent state of the art methods and demonstrate the robustness of this approach using three different datasets of scanned slides, prepared in different labs using different scanners. [ABSTRACT FROM AUTHOR]

In this paper, a total variation (TV) minimization strategy is proposed to overcome the problem of sparse spatial resolution and large amounts of noise in low dose positron emission tomography (PET) imaging reconstruction. Two types of objective function were established based on two statistical models of measured PET data, least-square (LS) TV for the Gaussian distribution and Poisson-TV for the Poisson distribution. To efficiently obtain high quality reconstructed images, the alternating direction method (ADM) is used to solve these objective functions. As compared with the iterative shrinkage/thresholding (IST) based algorithms, the proposed ADM can make full use of the TV constraint and its convergence rate is faster. The performance of the proposed approach is validated through comparisons with the expectation-maximization (EM) method using synthetic and experimental biological data. In the comparisons, the results of both LS-TV and Poisson-TV are taken into consideration to find which models are more suitable for PET imaging, in particular low-dose PET. To evaluate the results quantitatively, we computed bias, variance, and the contrast recovery coefficient (CRC) and drew profiles of the reconstructed images produced by the different methods. The results show that both Poisson-TV and LS-TV can provide a high visual quality at a low dose level. The bias and variance of the proposed LS-TV and Poisson-TV methods are 20% to 74% less at all counting levels than those of the EM method. Poisson-TV gives the best performance in terms of high-accuracy reconstruction with the lowest bias and variance as compared to the ground truth (14.3% less bias and 21.9% less variance). In contrast, LS-TV gives the best performance in terms of the high contrast of the reconstruction with the highest CRC. [ABSTRACT FROM AUTHOR]

Context: Magnetic resonance imaging (MRI) is currently the most accurate imaging modality to assess local prostate cancer stage. Despite a growing body of evidence, incorporation of MRI images into decision-making process concerning surgical template of radical prostatectomy, is complex and still poorly understood. Objective: We sought to determine the value of MRI in preoperative planning before radical prostatectomy. Materials and methods: Systematic search through electronic PubMed, EMBASE, and Cochrane databases from 2000 up to April 2018 was performed. Only studies that used preoperative MRI in decision-making process regarding extension of resection in patients with prostate cancer, in whom radical prostatectomy was an initial form of treatment were included into analysis. Their quality was scored by Risk Of Bias In Non-Randomized Studies of Interventions system. Meta-analysis was performed to calculate the weighted summary proportion under the fixed or random effects model as appropriate and pooled effects were depicted on forest plots. Results: The results showed that the preoperative MRI led to the modification of initial surgical template in one third of cases (35%). This occurred increasingly with the rising prostate cancer-risk category: 28%, 33%, 52% in low-, intermediate- and high-risk group, respectively. Modification of neurovascular bundle-sparing surgery based on MRI appeared to have no impact on the positive surgical margin rate. The decision based on MRI was correct on average in 77% of cases and differed across prostate cancer-risk categories: 63%, 75% and 91% in low-, intermediate- and high-risk group, accordingly. Conclusions: In summary, MRI has a considerable impact on the decision-making process regarding the extent of resection during radical prostatectomy. Adaptation of MRI images by operating surgeons has at worst no significant impact on surgical margin status, however its ability to decrease the positive surgical margin rates remains unconfirmed. [ABSTRACT FROM AUTHOR]

Lung 4D computed tomography (4D-CT) plays an important role in high-precision radiotherapy because it characterizes respiratory motion, which is crucial for accurate target definition. However, the manual segmentation of a lung tumor is a heavy workload for doctors because of the large number of lung 4D-CT data slices. Meanwhile, tumor segmentation is still a notoriously challenging problem in computer-aided diagnosis. In this paper, we propose a new method based on an improved graph cut algorithm with context information constraint to find a convenient and robust approach of lung 4D-CT tumor segmentation. We combine all phases of the lung 4D-CT into a global graph, and construct a global energy function accordingly. The sub-graph is first constructed for each phase. A context cost term is enforced to achieve segmentation results in every phase by adding a context constraint between neighboring phases. A global energy function is finally constructed by combining all cost terms. The optimization is achieved by solving a max-flow/min-cut problem, which leads to simultaneous and robust segmentation of the tumor in all the lung 4D-CT phases. The effectiveness of our approach is validated through experiments on 10 different lung 4D-CT cases. The comparison with the graph cut without context constraint, the level set method and the graph cut with star shape prior demonstrates that the proposed method obtains more accurate and robust segmentation results. [ABSTRACT FROM AUTHOR]

Introduction: Ovarian tumors are the most common diagnostic challenge for gynecologists and ultrasound examination has become the main technique for assessment of ovarian pathology and for preoperative distinction between malignant and benign ovarian tumors. However, ultrasonography is highly examiner-dependent and there may be an important variability between two different specialists when examining the same case. The objective of this work is the evaluation of different well-known Machine Learning (ML) systems to perform the automatic categorization of ovarian tumors from ultrasound images. Methods: We have used a real patient database whose input features have been extracted from 348 images, from the IOTA tumor images database, holding together with the class labels of the images. For each patient case and ultrasound image, its input features have been previously extracted using Fourier descriptors computed on the Region Of Interest (ROI). Then, four ML techniques are considered for performing the classification stage: K-Nearest Neighbors (KNN), Linear Discriminant (LD), Support Vector Machine (SVM) and Extreme Learning Machine (ELM). Results: According to our obtained results, the KNN classifier provides inaccurate predictions (less than 60% of accuracy) independently of the size of the local approximation, whereas the classifiers based on LD, SVM and ELM are robust in this biomedical classification (more than 85% of accuracy). Conclusions: ML methods can be efficiently used for developing the classification stage in computer-aided diagnosis systems of ovarian tumor from ultrasound images. These approaches are able to provide automatic classification with a high rate of accuracy. Future work should aim at enhancing the classifier design using ensemble techniques. Another ongoing work is to exploit different kind of features extracted from ultrasound images. [ABSTRACT FROM AUTHOR]

Objective: We performed a meta-analysis to evaluate the diagnostic value of positron emission tomography/computed tomography (PET/CT) in the detection of para-aortic lymph node metastasis in cervical cancer. Methods: We searched the PubMed, Embase, Web of Science, Cochrane Library, Chinese Biological Medicine (CBM), Chinese National Knowledge Infrastructure (CNKI), Wanfang and VIP databases in all languages from their inception to September 2018. Stat15.0 software was used to obtain pooled estimates of sensitivity (SEN), specificity (SPE), positive likelihood ratio (PLR), and negative likelihood ratio (NLR) as well as a summary receiver operating characteristic (SROC) curves. Deek‘s funnel plot was used to assess publication bias. QUADAS-2 was used to evaluate the quality of the studies. The protocol for this meta-analysis is registered in PROSPERO (CRD42019115330). Results: We obtained 14 studies, and the pooled estimates for sensitivity and specificity of PET/CT were 0.71 (95% confidence interval (CI) = 0.54–0.83) and 0.97 (95% CI = 0.93–0.98), respectively. Pooled PLR and NLR were 21.53 and 0.30, respectively. The diagnostic odds ratio (DOR) was70.59, and the area under the curve (AUC) was 0.95. Conclusion: PET/CT is an effective and important imaging method for the diagnosis of para-aortic lymph node metastasis in early cervical cancer. [ABSTRACT FROM AUTHOR]

Background: There are limited data about the role of endoscopic ultrasound-guided tissue acquisition (EUS-TA), by fine needle aspiration (EUS-FNA) or biopsy (EUS-FNB), in the evaluation of the adrenal glands (AG). The primary aim was to assess the diagnostic yield and safety. The secondary aims were the malignancy predictors, and to create a predictive model of malignancy. Methods: This was a retrospective nationwide study involving all Spanish hospitals experienced in EUS-TA of AGs. Inclusion period was from April-2003 to April-2016. Inclusion criteria: all consecutive cases that underwent EUS-TA of AGs. EUS and cytopathology findings were evaluated. Statistical analyses: diagnostic accuracy of echoendoscopist's suspicion using cytology by EUS-TA, as gold standard; multivariate logistic regression model to predict tumor malignancy. Results: A total of 204 EUS-TA of AGs were evaluated. Primary tumor locations were lung70%, others19%, and unknown11%. AG samples were adequate for cytological diagnosis in 91%, and confirmed malignancy in 60%. Diagnostic accuracy of the endosonographer's suspicion was 68%. The most common technique was: a 22-G (65%) and cytological needle (75%) with suction-syringe (66%). No serious adverse events were described. The variables most associated with malignancy were size>30mm (OR2.27; 95%CI, 1.16–4.05), heterogeneous echo-pattern (OR2.11; 95%CI, 1.1–3.9), variegated AG shape (OR2.46; 95%CI, 1–6.24), and endosonographer suspicion (OR17.46; 95%CI, 6.2–58.5). The best variables for a predictive multivariate logistic model of malignancy were age, sex, echo-pattern, and AG-shape. Conclusions: EUS-TA of the AGs is a safe, minimally invasive procedure, allowing an excellent diagnostic yield. These results suggest the possibility of developing a pre-EUS procedure predictive malignancy model. [ABSTRACT FROM AUTHOR]

Background: Prognostic models based on individual patient characteristics can improve treatment decisions and outcome in the future. In many (radiomic) studies, small size and heterogeneity of datasets is a challenge that often limits performance and potential clinical applicability of these models. The current study is example of a retrospective multi-centric study with challenges and caveats. To highlight common issues and emphasize potential pitfalls, we aimed for an extensive analysis of these multi-center pre-treatment datasets, with an additional 18F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) scan acquired during treatment. Methods: The dataset consisted of 138 stage II-IV non-small cell lung cancer (NSCLC) patients from four different cohorts acquired from three different institutes. The differences between the cohorts were compared in terms of clinical characteristics and using the so-called ‘cohort differences model’ approach. Moreover, the potential prognostic performances for overall survival of radiomic features extracted from CT or FDG-PET, or relative or absolute differences between the scans at the two time points, were assessed using the LASSO regression method. Furthermore, the performances of five different classifiers were evaluated for all image sets. Results: The individual cohorts substantially differed in terms of patient characteristics. Moreover, the cohort differences model indicated statistically significant differences between the cohorts. Neither LASSO nor any of the tested classifiers resulted in a clinical relevant prognostic model that could be validated on the available datasets. Conclusion: The results imply that the study might have been influenced by a limited sample size, heterogeneous patient characteristics, and inconsistent imaging parameters. No prognostic performance of FDG-PET or CT based radiomics models can be reported. This study highlights the necessity of extensive evaluations of cohorts and of validation datasets, especially in retrospective multi-centric datasets. [ABSTRACT FROM AUTHOR]

Objectives: The extent of peritumoral tumor cell infiltrations in glioblastoma contributes to poor prognosis. We aimed to assess additive prognostic value of Minkowski functionals in analyzing heterogeneity of peritumoral hyperintensity on T2WI in glioblastoma patients. Methods: Clinical data (age, sex, extent of surgical resection), O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status and pre-operative T2WI of 113 pathologically confirmed glioblastoma patients (from our institution, n = 61; from the Cancer Imaging Archive, n = 52) were retrospectively reviewed. The patients were randomly grouped into a training set (n = 80) and a test set (n = 33). Peritumoral T2 hyperintensity was manually segmented and Minkowski functionals—a texture analysis method capturing heterogeneity of MR images—were computed as a function of 11 grayscale thresholds. The Cox proportional hazards models were fitted with clinical variables, Minkowski functionals features as well as both combined. The risk prediction performances of the Minkowski functionals and combined models were validated on a separate test dataset. The sex-specific survival difference of the entire cohort was analyzed according to MGMT methylation status via Kaplan-Meier survival curves. Results: Thirty-three Minkowski features (11 area, 11 perimeter and 11 genus) for each patient were acquired giving a total of 3729 features. Cox regression models fitted with clinical data, Minkowski features, and both combined had incremental concordance indices of 0.577 (P = 0.02), 0.706 (P = 0.02) and 0.714 (P = 0.01) respectively. The prediction error rate of the combined model—having clinical and Minkowski features—was lower than that of Minkowski functionals model (0.135 and 0.161, respectively) when validated on a test dataset. No sex-specific survival difference was found according to MGMT methylation status (male, P = 0.2; female, P = 0.22). Conclusions: Minkowski functionals features computed from peritumoral hyperintensity can capture heterogeneity of glioblastoma on T2WI and have additive prognostic value in predicting survival, demonstrating their potential in complementing currently available prognostic parameters. [ABSTRACT FROM AUTHOR]

Purpose: Normalized emphysema score is a protocol-robust CT biomarker of mortality. We aimed to improve mortality prediction by including the emphysema score progression rate–its change over time–into the models. Method and materials: CT scans from 6000 National Lung Screening Trial CT arm participants were included. Of these, 1810 died (445 lung cancer-specific). The remaining 4190 survivors were sampled with replacement up to 24432 to approximate the full cohort. Three overlapping subcohorts were formed which required participants to have images from specific screening rounds. Emphysema scores were obtained after resampling, normalization, and bullae cluster analysis of the original images. Base models contained solely the latest emphysema score. Progression models included emphysema score progression rate. Models were adjusted by including baseline age, sex, BMI, smoking status, smoking intensity, smoking duration, and previous COPD diagnosis. Cox proportional hazard models predicting all-cause and lung cancer mortality were compared by calculating the area under the curve per year follow-up. Results: In the subcohort of participants with baseline and first annual follow-up scans, the analysis was performed on 4940 participants (23227 after resampling). Area under the curve for all-cause mortality predictions of the base and progression models 6 years after baseline were 0.564 (0.564 to 0.565) and 0.569 (0.568 to 0.569) when unadjusted, and 0.704 (0.703 to 0.704) to 0.705 (0.704 to 0.705) when adjusted. The respective performances predicting lung cancer mortality were 0.638 (0.637 to 0.639) and 0.643 (0.642 to 0.644) when unadjusted, and 0.724 (0.723 to 0.725) and 0.725 (0.725 to 0.726) when adjusted. Conclusion: Including emphysema score progression rate into risk models shows no clinically relevant improvement in mortality risk prediction. This is because scan normalization does not adjust for an overall change in lung density. Adjusting for changes in smoking behavior is likely required to make this a clinically useful measure of emphysema progression. [ABSTRACT FROM AUTHOR]

Objective: The objective is to present a proof-of-concept of a semi-automatic method to reduce hippocampus segmentation time on magnetic resonance images (MRI). Materials and methods: FAst Segmentation Through SURface Fairing (FASTSURF) is based on a surface fairing technique which reconstructs the hippocampus from sparse delineations. To validate FASTSURF, simulations were performed in which sparse delineations extracted from full manual segmentations served as input. On three different datasets with different diagnostic groups, FASTSURF hippocampi were compared to the original segmentations using Jaccard overlap indices and percentage volume differences (PVD). In one data set for which back-to-back scans were available, unbiased estimates of overlap and PVD were obtained. Using longitudinal scans, we compared hippocampal atrophy rates measured by manual, FASTSURF and two automatic segmentations (FreeSurfer and FSL-FIRST). Results: With only seven input contours, FASTSURF yielded mean Jaccard indices ranging from 72(±4.3)% to 83(±2.6)% and PVDs ranging from 0.02(±2.40)% to 3.2(±3.40)% across the three datasets. Slightly poorer results were obtained for the unbiased analysis, but the performance was still considerably better than both tested automatic methods with only five contours. Conclusions: FASTSURF segmentations have high accuracy and require only a fraction of the delineation effort of fully manual segmentation. Atrophy rate quantification based on completely manual segmentation is well reproduced by FASTSURF. Therefore, FASTSURF is a promising tool to be implemented in clinical workflow, provided a future prospective validation confirms our findings. [ABSTRACT FROM AUTHOR]

The current research study is concerned with the automated differentiation between histopathological slides from colon tissues with respect to four classes (healthy tissue and cancerous of grades 1, 2 or 3) through an optimized ensemble of predictors. Six distinct classifiers with prediction accuracies ranging from 87% to 95% are considered for the task. The proposed method of combining them takes into account the probabilities of the individual classifiers for each sample to be assigned to any of the four classes, optimizes weights for each technique by differential evolution and attains an accuracy that is significantly better than the individual results. Moreover, a degree of confidence is defined that would allow the pathologists to separate the data into two distinct sets, one that is correctly classified with a high level of confidence and the rest that would need their further attention. The tandem is also validated on other benchmark data sets. The proposed methodology proves to be efficient in improving the classification accuracy of each algorithm taken separately and performs reasonably well on other data sets, even with default weights. In addition, by establishing a degree of confidence the method becomes more viable for use by actual practitioners. [ABSTRACT FROM AUTHOR]

Purpose: In dynamic contrast enhanced (DCE) MRI, separation of signal contributions from perfusion and leakage requires robust estimation of parameters in a pharmacokinetic model. We present and quantify the performance of a method to compute tissue hemodynamic parameters from DCE data using established pharmacokinetic models. Methods: We propose a Bayesian scheme to obtain perfusion metrics from DCE MRI data. Initial performance is assessed through digital phantoms of the extended Tofts model (ETM) and the two-compartment exchange model (2CXM), comparing the Bayesian scheme to the standard Levenberg-Marquardt (LM) algorithm. Digital phantoms are also invoked to identify limitations in the pharmacokinetic models related to measurement conditions. Using computed maps of the extra vascular volume (ve) from 19 glioma patients, we analyze differences in the number of un-physiological high-intensity ve values for both ETM and 2CXM, using a one-tailed paired t-test assuming un-equal variance. Results: The Bayesian parameter estimation scheme demonstrated superior performance over the LM technique in the digital phantom simulations. In addition, we identified limitations in parameter reliability in relation to scan duration for the 2CXM. DCE data for glioma and cervical cancer patients was analyzed with both algorithms and demonstrated improvement in image readability for the Bayesian method. The Bayesian method demonstrated significantly fewer non-physiological high-intensity ve values for the ETM (p<0.0001) and the 2CXM (p<0.0001). Conclusion: We have demonstrated substantial improvement of the perceptive quality of pharmacokinetic parameters from advanced compartment models using the Bayesian parameter estimation scheme as compared to the LM technique. [ABSTRACT FROM AUTHOR]

Purpose: This study aimed to evaluate the dosimetric accuracy of respiratory gated volumetric modulated arc therapy (VMAT) for lung stereotactic body radiation therapy (SBRT) under simulation conditions similar to the actual clinical situation using patient-specific lung phantoms and realistic target movements. Methods: Six heterogeneous lung phantoms were fabricated using a 3D-printer (3DISON, ROKIT, Seoul, Korea) to be dosimetrically equivalent to actual target regions of lung SBRT cases treated via gated VMAT. They were designed to move realistically via a motion device (QUASAR, Modus Medical Devices, Canada). Using the lung phantoms and a homogeneous phantom (model 500–3315, Modus Medical Devices), film dosimetry was performed with and without respiratory gating for VMAT delivery (TrueBeam STx; Varian Medical Systems, Palo Alto, CA, USA). The measured results were analyzed with the gamma passing rates (GPRs) of 2%/1 mm criteria, by comparing with the calculated dose via the AXB and AAA algorithms of the Eclipse Treatment Planning System (version 10.0.28; Varian Medical Systems). Results: GPRs were greater than the acceptance criteria 80% for all film measurements with the stationary and homogeneous phantoms in conventional QAs. Regardless of the heterogeneity of phantoms, there were no significant differences (p > 0.05) in GPRs obtained with and without target motions; the statistical significance (p = 0.031) was presented between both algorithms under the utilization of heterogeneous phantoms. Conclusions: Dosimetric verification with heterogeneous patient-specific lung phantoms could be successfully implemented as the evaluation method for gated VMAT delivery. In addition, it could be dosimetrically confirmed that the AXB algorithm improved the dose calculation accuracy under patient-specific simulations using 3D printed lung phantoms. [ABSTRACT FROM AUTHOR]

Introduction: 11C-acetate (ACE)-PET/CT is used for staging of high-risk prostate cancer. PET data is reconstructed with iterative algorithms, such as VUEPointHD ViP (VPHD) and VUEPoint HD Sharp IR (SharpIR), the latter with additional resolution recovery. It is expected that the resolution recovery algorithm should render more accurate maximum and mean standardized uptake values (SUVmax and SUVmean) and functional tumor volumes (FTV) than the ordinary OSEM. Performing quantitative analysis, choice of volume-of-interest delineation algorithm (SUV threshold) may influence FTV. Optimizing PET acquisition time is justified if image quality and quantitation do not deteriorate. The aim of this study is to identify the optimal reconstruction algorithm, SUV threshold and acquisition time for ACE-PET/CT. Methods: ACE-PET/CT data acquired with a General Electric Discovery 690 PET/CT from 16 consecutive high-risk prostate cancer patients was reconstructed with VPHD and SharpIR. Forty pelvic lymph nodes (LNs) and 14 prostate glands were delineated with 42% and estimated threshold. SUVmax, SUVmean, FTV and total lesion uptake were measured. Default acquisition time was four minutes per bed position. In a subset of lesions, acquisition times of one, two and four minutes were evaluated. Structural tumor volumes (STV) of the LNs were measured with CT for correlation with functional volumetric parameters. To validate SUV quantification under different conditions with SharpIR 42%, recovery coefficients (RCs) of SUVmean and FTV were calculated from a phantom with 18F-fluoro-deoxy-glucose (FDG)-filled volumes 0.1–9.2cm3 and signal-to-background (S/B) ratios 4.3–15.9. Results: With SharpIR, SUVmax and SUVmean were higher and FTV lower compared with VPHD, regardless of threshold method, in both prostates and LNs. Total lesion uptake determined with both threshold methods was lower with SharpIR compared with VPHD with both threshold methods, except in subgroup analysis of prostate targets where estimated threshold returned higher values. Longer acquisition times returned higher FTV for both threshold methods, regardless of reconstruction algorithm. The FTV difference was most pronounced with one minute’s acquisition per bed position, which also produced visually the highest noise. SUV parameters were unaffected by varying acquisition times. FTV with SharpIR 42% showed the best correspondence with STV. SharpIR 42% gave higher RCs of SUVmean and FTV with increasing phantom size and S/B-ratio, as expected. Conclusions: Delineation with SharpIR 42% seems to provide the most accurate combined information from SUVmax, SUVmean, FTV and total lesion uptake. Acquisition time may be shortened to two minutes per bed position with preserved image quality. [ABSTRACT FROM AUTHOR]

Background: Chest radiograph interpretation is critical for the detection of thoracic diseases, including tuberculosis and lung cancer, which affect millions of people worldwide each year. This time-consuming task typically requires expert radiologists to read the images, leading to fatigue-based diagnostic error and lack of diagnostic expertise in areas of the world where radiologists are not available. Recently, deep learning approaches have been able to achieve expert-level performance in medical image interpretation tasks, powered by large network architectures and fueled by the emergence of large labeled datasets. The purpose of this study is to investigate the performance of a deep learning algorithm on the detection of pathologies in chest radiographs compared with practicing radiologists.Methods and Findings: We developed CheXNeXt, a convolutional neural network to concurrently detect the presence of 14 different pathologies, including pneumonia, pleural effusion, pulmonary masses, and nodules in frontal-view chest radiographs. CheXNeXt was trained and internally validated on the ChestX-ray8 dataset, with a held-out validation set consisting of 420 images, sampled to contain at least 50 cases of each of the original pathology labels. On this validation set, the majority vote of a panel of 3 board-certified cardiothoracic specialist radiologists served as reference standard. We compared CheXNeXt's discriminative performance on the validation set to the performance of 9 radiologists using the area under the receiver operating characteristic curve (AUC). The radiologists included 6 board-certified radiologists (average experience 12 years, range 4-28 years) and 3 senior radiology residents, from 3 academic institutions. We found that CheXNeXt achieved radiologist-level performance on 11 pathologies and did not achieve radiologist-level performance on 3 pathologies. The radiologists achieved statistically significantly higher AUC performance on cardiomegaly, emphysema, and hiatal hernia, with AUCs of 0.888 (95% confidence interval [CI] 0.863-0.910), 0.911 (95% CI 0.866-0.947), and 0.985 (95% CI 0.974-0.991), respectively, whereas CheXNeXt's AUCs were 0.831 (95% CI 0.790-0.870), 0.704 (95% CI 0.567-0.833), and 0.851 (95% CI 0.785-0.909), respectively. CheXNeXt performed better than radiologists in detecting atelectasis, with an AUC of 0.862 (95% CI 0.825-0.895), statistically significantly higher than radiologists' AUC of 0.808 (95% CI 0.777-0.838); there were no statistically significant differences in AUCs for the other 10 pathologies. The average time to interpret the 420 images in the validation set was substantially longer for the radiologists (240 minutes) than for CheXNeXt (1.5 minutes). The main limitations of our study are that neither CheXNeXt nor the radiologists were permitted to use patient history or review prior examinations and that evaluation was limited to a dataset from a single institution.Conclusions: In this study, we developed and validated a deep learning algorithm that classified clinically important abnormalities in chest radiographs at a performance level comparable to practicing radiologists. Once tested prospectively in clinical settings, the algorithm could have the potential to expand patient access to chest radiograph diagnostics. [ABSTRACT FROM AUTHOR]

We aimed to describe the development of an inference model for computer-aided diagnosis of lung nodules that could provide valid reasoning for any inferences, thereby improving the interpretability and performance of the system. An automatic construction method was used that considered explanation adequacy and inference accuracy. In addition, we evaluated the usefulness of prior experts’ (radiologists’) knowledge while constructing the models. In total, 179 patients with lung nodules were included and divided into 79 and 100 cases for training and test data, respectively. F-measure and accuracy were used to assess explanation adequacy and inference accuracy, respectively. For F-measure, reasons were defined as proper subsets of Evidence that had a strong influence on the inference result. The inference models were automatically constructed using the Bayesian network and Markov chain Monte Carlo methods, selecting only those models that met the predefined criteria. During model constructions, we examined the effect of including radiologist’s knowledge in the initial Bayesian network models. Performance of the best models in terms of F-measure, accuracy, and evaluation metric were as follows: 0.411, 72.0%, and 0.566, respectively, with prior knowledge, and 0.274, 65.0%, and 0.462, respectively, without prior knowledge. The best models with prior knowledge were then subjectively and independently evaluated by two radiologists using a 5-point scale, with 5, 3, and 1 representing beneficial, appropriate, and detrimental, respectively. The average scores by the two radiologists were 3.97 and 3.76 for the test data, indicating that the proposed computer-aided diagnosis system was acceptable to them. In conclusion, the proposed method incorporating radiologists’ knowledge could help in eliminating radiologists’ distrust of computer-aided diagnosis and improving its performance. [ABSTRACT FROM AUTHOR]

Background: Four-dimensional computed tomography (4D-CT) ventilation is an emerging imaging modality. Functional avoidance of regions according to 4D-CT ventilation may reduce lung toxicity after radiation therapy. This study evaluated associations between 4D-CT ventilation-based dosimetric parameters and clinical outcomes. Methods: Pre-treatment 4D-CT data were used to retrospectively construct ventilation images for 40 thoracic cancer patients retrospectively. Fifteen patients were treated with conventional radiation therapy, 6 patients with hyperfractionated radiation therapy and 19 patients with stereotactic body radiation therapy (SBRT). Ventilation images were calculated from 4D-CT data using a deformable image registration and Jacobian-based algorithm. Each ventilation map was normalized by converting it to percentile images. Ventilation-based dosimetric parameters (Mean Dose, V5 [percent lung volume receiving ≥5 Gy], and V20 [percent lung volume receiving ≥20 Gy]) were calculated for highly and poorly ventilated regions. To test whether the ventilation-based dosimetric parameters could be used predict radiation pneumonitis of ≥Grade 2, the area under the curve (AUC) was determined from the receiver operating characteristic analysis. Results: For Mean Dose, poorly ventilated lung regions in the 0–30% range showed the highest AUC value (0.809; 95% confidence interval [CI], 0.663–0.955). For V20, poorly ventilated lung regions in the 0–20% range had the highest AUC value (0.774; 95% [CI], 0.598–0.915), and for V5, poorly ventilated lung regions in the 0–30% range had the highest AUC value (0.843; 95% [CI], 0.732–0.954). The highest AUC values for Mean Dose, V20, and V5 were obtained in poorly ventilated regions. There were significant differences in all dosimetric parameters between radiation pneumonitis of Grade 1 and Grade ≥2. Conclusions: Poorly ventilated lung regions identified on 4D-CT had higher AUC values than highly ventilated regions, suggesting that functional planning based on poorly ventilated regions may reduce the risk of lung toxicity in radiation therapy. [ABSTRACT FROM AUTHOR]

In this work, a computer-aided tool for detection was developed to segment breast masses from clinical ultrasound (US) scans. The underlying Multi U-net algorithm is based on convolutional neural networks. Under the Mayo Clinic Institutional Review Board protocol, a prospective study of the automatic segmentation of suspicious breast masses was performed. The cohort consisted of 258 female patients who were clinically identified with suspicious breast masses and underwent clinical US scan and breast biopsy. The computer-aided detection tool effectively segmented the breast masses, achieving a mean Dice coefficient of 0.82, a true positive fraction (TPF) of 0.84, and a false positive fraction (FPF) of 0.01. By avoiding positioning of an initial seed, the algorithm is able to segment images in real time (13–55 ms per image), and can have potential clinical applications. The algorithm is at par with a conventional seeded algorithm, which had a mean Dice coefficient of 0.84 and performs significantly better (P< 0.0001) than the original U-net algorithm. [ABSTRACT FROM AUTHOR]

Background: There are two main types of strategies to identify target population for lung cancer screening: 1) strategies based on age and cumulative smoking criteria, 2) risk prediction models allowing the calculation of an individual risk. The objective of this study was to compare different strategies to identify the proportion of the Spanish population at high risk of developing lung cancer, susceptible to be included in a lung cancer screening programme. Methods: Cross-sectional study. We used the data of the Spanish National Interview Health Survey (ENSE) of 2011–2012 (21,006 individuals) to estimate the proportion of participants at high risk of developing lung cancer. This estimation was performed using the U.S. national lung screening trial (NLST) criteria and a 6-year prediction model (PLCOm2012), both independently and in combination. Results: The prevalence of individuals at high risk of developing lung cancer according to the NLST criteria was 4.9% (7.9% for men, 2.4% for women). Among the 1,034 subjects who met the NLST criteria, 533 (427 men and 106 women) had a 6-year lung cancer risk ≥2.0%. The combination of these two selection strategies showed that 2.5% of the Spanish population had a high risk of developing lung cancer. However, this selection process did not take into account different groups of subjects <75 years old having an individual risk of lung cancer ≥2%, such as heavy smokers <55 years old who were long-time former smokers, and ever-smokers having smoked <30 pack-years with other risk factors. Conclusions: Further research is needed to determine which selection strategy achieves a higher benefit/harm ratio and to assess other prevention strategies for individuals with elevated risk for lung cancer but who do not meet the screening eligibility criteria. [ABSTRACT FROM AUTHOR]

Surgery for brain cancer is a major problem in neurosurgery. The diffuse infiltration into the surrounding normal brain by these tumors makes their accurate identification by the naked eye difficult. Since surgery is the common treatment for brain cancer, an accurate radical resection of the tumor leads to improved survival rates for patients. However, the identification of the tumor boundaries during surgery is challenging. Hyperspectral imaging is a non-contact, non-ionizing and non-invasive technique suitable for medical diagnosis. This study presents the development of a novel classification method taking into account the spatial and spectral characteristics of the hyperspectral images to help neurosurgeons to accurately determine the tumor boundaries in surgical-time during the resection, avoiding excessive excision of normal tissue or unintentionally leaving residual tumor. The algorithm proposed in this study to approach an efficient solution consists of a hybrid framework that combines both supervised and unsupervised machine learning methods. Firstly, a supervised pixel-wise classification using a Support Vector Machine classifier is performed. The generated classification map is spatially homogenized using a one-band representation of the HS cube, employing the Fixed Reference t-Stochastic Neighbors Embedding dimensional reduction algorithm, and performing a K-Nearest Neighbors filtering. The information generated by the supervised stage is combined with a segmentation map obtained via unsupervised clustering employing a Hierarchical K-Means algorithm. The fusion is performed using a majority voting approach that associates each cluster with a certain class. To evaluate the proposed approach, five hyperspectral images of surface of the brain affected by glioblastoma tumor in vivo from five different patients have been used. The final classification maps obtained have been analyzed and validated by specialists. These preliminary results are promising, obtaining an accurate delineation of the tumor area. [ABSTRACT FROM AUTHOR]

Background: Advances in bronchoscopy and CT-guided lung biopsy have improved the evaluation of small pulmonary lesions (PLs), leading to an increase in preoperative histological diagnosis. We aimed to evaluate the efficacy and safety of transbronchial lung biopsy using radial endobronchial ultrasound and virtual bronchoscopic navigation (TBLB-rEBUS&VBN) and CT-guided transthoracic needle biopsy (CT-TNB) for tissue diagnosis of small PLs. Methods: A systematic search was performed in five electronic databases, including MEDLINE, EMBASE, Cochrane Library Central Register of Controlled Trials, Web of Science, and Scopus, for relevant studies in May 2016; the selected articles were assessed using meta-analysis. The articles were limited to those published after 2000 that studied small PLs ≤ 3 cm in diameter. Results: From 7345 records, 9 articles on the bronchoscopic (BR) approach and 15 articles on the percutaneous (PC) approach were selected. The pooled diagnostic yield was 75% (95% confidence interval [CI], 69–80) using the BR approach and 93% (95% CI, 90–96) using the PC approach. For PLs ≤ 2 cm, the PC approach (pooled diagnostic yield: 92%, 95% CI: 88–95) was superior to the BR approach (66%, 95% CI: 55–76). However, for PLs > 2 cm but ≤ 3 cm, the diagnostic yield using the BR approach was improved to 81% (95% CI, 75–85). Complications of pneumothorax and hemorrhage were rare with the BR approach but common with the PC approach. Conclusions: CT-TNB was superior to TBLB-rEBUS&VBN for the evaluation of small PLs. However, for lesions greater than 2 cm, the BR approach may be considered considering its diagnostic yield of over 80% and the low risk of procedure-related complications. [ABSTRACT FROM AUTHOR]

Purpose: Determine the characteristics of percutaneous core biopsies that are adequate for a next generation sequencing (NGS) genomic panel. Materials and methods: All patients undergoing percutaneous core biopsies in interventional radiology (IR) with samples evaluated for a 46-gene NGS panel during 1-year were included in this retrospective study. Patient and procedure variables were collected. An imaging-based likelihood of adequacy score incorporating targeting and sampling factors was assigned to each biopsied lesion. Univariate and multivariate logistic regression was performed. Results: 153 patients were included (58.2% female, average age 59.5 years). The most common malignancy was lung cancer (40.5%), most common biopsied site was lung (36%), and average size of biopsied lesions was 3.8 cm (+/- 2.7). Adequacy for NGS was 69.9%. Univariate analysis showed higher likelihood of adequacy score (p = 0.004), primary malignancy type (p = 0.03), and absence of prior systemic therapy (p = 0.018) were associated with adequacy for NGS. Multivariate analysis showed higher adequacy for lesions with likelihood of adequacy scored 3 (high) versus lesions scored 1 (low) (OR, 7.82; p = 0.002). Melanoma lesions had higher adequacy for NGS versus breast cancer lesions (OR 9.5; p = 0.01). Absence of prior systemic therapy (OR, 6.1; p = 0.02) and systemic therapy 3 months before biopsy yielded greater adequacy for NGS. Lesions <3 cm had greater adequacy for NGS than larger lesions (OR 2.72, p = 0.02). Conclusion: As targeted therapy becomes standard for more cancers, percutaneous biopsy specimens adequate for NGS genomic testing will be needed. An imaging-based likelihood of adequacy score assigned by IR physicians and other pre-procedure variables can help predict the likelihood of biopsy adequacy for NGS. [ABSTRACT FROM AUTHOR]

Accurate and automatic brain metastases target delineation is a key step for efficient and effective stereotactic radiosurgery (SRS) treatment planning. In this work, we developed a deep learning convolutional neural network (CNN) algorithm for segmenting brain metastases on contrast-enhanced T1-weighted magnetic resonance imaging (MRI) datasets. We integrated the CNN-based algorithm into an automatic brain metastases segmentation workflow and validated on both Multimodal Brain Tumor Image Segmentation challenge (BRATS) data and clinical patients' data. Validation on BRATS data yielded average DICE coefficients (DCs) of 0.75±0.07 in the tumor core and 0.81±0.04 in the enhancing tumor, which outperformed most techniques in the 2015 BRATS challenge. Segmentation results of patient cases showed an average of DCs 0.67±0.03 and achieved an area under the receiver operating characteristic curve of 0.98±0.01. The developed automatic segmentation strategy surpasses current benchmark levels and offers a promising tool for SRS treatment planning for multiple brain metastases. [ABSTRACT FROM AUTHOR]

The fast and accurate segmentation of lung nodule image sequences is the basis of subsequent processing and diagnostic analyses. However, previous research investigating nodule segmentation algorithms cannot entirely segment cavitary nodules, and the segmentation of juxta-vascular nodules is inaccurate and inefficient. To solve these problems, we propose a new method for the segmentation of lung nodule image sequences based on superpixels and density-based spatial clustering of applications with noise (DBSCAN). First, our method uses three-dimensional computed tomography image features of the average intensity projection combined with multi-scale dot enhancement for preprocessing. Hexagonal clustering and morphological optimized sequential linear iterative clustering (HMSLIC) for sequence image oversegmentation is then proposed to obtain superpixel blocks. The adaptive weight coefficient is then constructed to calculate the distance required between superpixels to achieve precise lung nodules positioning and to obtain the subsequent clustering starting block. Moreover, by fitting the distance and detecting the change in slope, an accurate clustering threshold is obtained. Thereafter, a fast DBSCAN superpixel sequence clustering algorithm, which is optimized by the strategy of only clustering the lung nodules and adaptive threshold, is then used to obtain lung nodule mask sequences. Finally, the lung nodule image sequences are obtained. The experimental results show that our method rapidly, completely and accurately segments various types of lung nodule image sequences. [ABSTRACT FROM AUTHOR]

Background: In our clinical practice we have observed that despite a high hepatic metastatic tumor burden, serum alkaline phosphatase (AP) levels are frequently normal in cases of metastatic neuroendocrine tumor (NET). Patients and methods: We retrospectively reviewed the records of patients with grade 1 and 2 NETs with liver metastases but without bone metastases seen at our institution in 2013. In total, 49 patients were included (22 female), with a median age of 60 years (range: 28 to 84 years). The primary tumors were located in the duodenum/pancreas (n = 29), small bowel (n = 17) or colon/rectum (n = 3); 10 cases were grade 1 and 39 grade 2. Hepatic involvement was bulky, with more than 10 lesions in 23 patients and a tumor burden above 10% of the liver volume in 26 patients. Results: Serum AP levels were elevated (≥ upper limit of normal (ULN)) in 16 patients. In multiparametric analysis, elevated serum AP levels were not associated with the primary site, grade, or number or volume of metastases. In multiparametric analysis, progression-free survival was only correlated with grade (p = 0.010) and AP level (p = 0.017). Conclusions: Serum AP levels are frequently normal in liver metastases from NET, even in the event of a major tumor burden, and the serum AP level can be of prognostic value. [ABSTRACT FROM AUTHOR]

Objective: The purpose of this study was to investigate the relationship between visual score of emphysema and homology-based emphysema quantification (HEQ) and evaluate whether visual score was accurately predicted by machine learning and HEQ. Materials and methods: A total of 115 anonymized computed tomography images from 39 patients were obtained from a public database. Emphysema quantification of these images was performed by measuring the percentage of low-attenuation lung area (LAA%). The following values related to HEQ were obtained: nb0 and nb1. LAA% and HEQ were calculated at various threshold levels ranging from −1000 HU to −700 HU. Spearman’s correlation coefficients between emphysema quantification and visual score were calculated at the various threshold levels. Visual score was predicted by machine learning and emphysema quantification (LAA% or HEQ). Random Forest was used as a machine learning algorithm, and accuracy of prediction was evaluated by leave-one-patient-out cross validation. The difference in the accuracy was assessed using McNemar’s test. Results: The correlation coefficients between emphysema quantification and visual score were as follows: LAA% (−950 HU), 0.567; LAA% (−910 HU), 0.654; LAA% (−875 HU), 0.704; nb0 (−950 HU), 0.552; nb0 (−910 HU), 0.629; nb0 (−875 HU), 0.473; nb1 (−950 HU), 0.149; nb1 (−910 HU), 0.519; and nb1 (−875 HU), 0.716. The accuracy of prediction was as follows: LAA%, 55.7% and HEQ, 66.1%. The difference in accuracy was statistically significant (p = 0.0290). Conclusion: LAA% and HEQ at −875 HU showed a stronger correlation with visual score than those at −910 or −950 HU. HEQ was more useful than LAA% for predicting visual score. [ABSTRACT FROM AUTHOR]

Background: Novel advanced MRI techniques are investigated in patients treated for head and neck tumors as conventional anatomical MRI is unreliable to differentiate tumor from treatment related imaging changes. Purpose: As the diagnostic accuracy of MRI techniques to detect tumor residual or recurrence during or after treatment is variable reported in the literature, we performed a systematic meta-analysis. Data sources: Pubmed, EMBASE and Web of Science were searched from their first record to September 23th 2014. Study selection: Studies reporting diagnostic accuracy of anatomical, ADC, perfusion or spectroscopy to identify tumor response confirmed by histology or follow-up in treated patients for head and neck tumors were selected by two authors independently. Data analysis: Two authors independently performed data extraction including true positives, false positives, true negatives, false negatives and general study characteristics. Meta-analysis was performed using bivariate random effect models when ≥5 studies per test were included. Data synthesis: We identified 16 relevant studies with anatomical MRI and ADC. No perfusion or spectroscopy studies were identified. Pooled analysis of anatomical MRI of the primary site (11 studies, N = 854) displayed a sensitivity of 84% (95%CI 72–92) and specificity of 82% (71–89). ADC of the primary site (6 studies, N = 287) showed a pooled sensitivity of 89% (74–96) and specificity of 86% (69–94). Limitations: Main limitation are the low, but comparable quality of the included studies and the variability between the studies. Conclusions: The higher diagnostic accuracy of ADC values over anatomical MRI for the primary tumor location emphases the relevance to include DWI with ADC for response evaluation of treated head and neck tumor patients. [ABSTRACT FROM AUTHOR]

A new methodology based on tensor algebra that uses a higher order singular value decomposition to perform three-dimensional voxel reconstruction from a series of temporal images obtained using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is proposed. Principal component analysis (PCA) is used to robustly extract the spatial and temporal image features and simultaneously de-noise the datasets. Tumour segmentation on enhanced scaled (ES) images performed using a fuzzy C-means (FCM) cluster algorithm is compared with that achieved using the proposed tensorial framework. The proposed algorithm explores the correlations between spatial and temporal features in the tumours. The multi-channel reconstruction enables improved breast tumour identification through enhanced de-noising and improved intensity consistency. The reconstructed tumours have clear and continuous boundaries; furthermore the reconstruction shows better voxel clustering in tumour regions of interest. A more homogenous intensity distribution is also observed, enabling improved image contrast between tumours and background, especially in places where fatty tissue is imaged. The fidelity of reconstruction is further evaluated on the basis of five new qualitative metrics. Results confirm the superiority of the tensorial approach. The proposed reconstruction metrics should also find future applications in the assessment of other reconstruction algorithms. [ABSTRACT FROM AUTHOR]

Evaluation of bone marrow involvement (BMI) is paramount in diffuse large B-cell lymphoma (DLBCL) for prognostic and therapeutic reasons. PET/CT scanner (PET) is now a routine examination for the staging of DLBCL with prognostic and therapeutic implications. This study evaluates the role of PET for detecting marrow involvement compared to bone marrow biopsy (BMB). This monocentric study included 54 patients diagnosed with DLBCL between 2009 and 2013 and who had FDG PET/CT in a pre-treatment setting. A correlation analysis of the detection of BMI by PET and BMB was performed. A prognostic evaluation of BMI by BMB and/or PET/CT and correlation with an overall 2-year survival were analyzed. PET was more sensitive for the detection of BMI than BMB (92.3% vs. 38.5%). It can be considered a discriminatory Pre-BMB test with a negative predictive value of 97.6%. In addition, BMI by PET had a prognostic value with strong correlation with progression-free survival (PFS) (HR = 3.81; p = 0.013) and overall survival (OS) (HR = 4.12; p = 0.03) while the BMB had not. PET shows superior performance to the BMB for the detection of marrow involvement in DLBCL. It may be considered as the first line examination of bone marrow instead of the biopsy. [ABSTRACT FROM AUTHOR]

Purpose: To develop and validate fully automated generation of VMAT plan-libraries for plan-of-the-day adaptive radiotherapy in locally-advanced cervical cancer. Material and Methods: Our framework for fully automated treatment plan generation (Erasmus-iCycle) was adapted to create dual-arc VMAT treatment plan libraries for cervical cancer patients. For each of 34 patients, automatically generated VMAT plans (autoVMAT) were compared to manually generated, clinically delivered 9-beam IMRT plans (CLINICAL), and to dual-arc VMAT plans generated manually by an expert planner (manVMAT). Furthermore, all plans were benchmarked against 20-beam equi-angular IMRT plans (autoIMRT). For all plans, a PTV coverage of 99.5% by at least 95% of the prescribed dose (46 Gy) had the highest planning priority, followed by minimization of V45Gy for small bowel (SB). Other OARs considered were bladder, rectum, and sigmoid. Results: All plans had a highly similar PTV coverage, within the clinical constraints (above). After plan normalizations for exactly equal median PTV doses in corresponding plans, all evaluated OAR parameters in autoVMAT plans were on average lower than in the CLINICAL plans with an average reduction in SB V45Gy of 34.6% (p<0.001). For 41/44 autoVMAT plans, SB V45Gy was lower than for manVMAT (p<0.001, average reduction 30.3%), while SB V15Gy increased by 2.3% (p = 0.011). AutoIMRT reduced SB V45Gy by another 2.7% compared to autoVMAT, while also resulting in a 9.0% reduction in SB V15Gy (p<0.001), but with a prolonged delivery time. Differences between manVMAT and autoVMAT in bladder, rectal and sigmoid doses were ≤ 1%. Improvements in SB dose delivery with autoVMAT instead of manVMAT were higher for empty bladder PTVs compared to full bladder PTVs, due to differences in concavity of the PTVs. Conclusions: Quality of automatically generated VMAT plans was superior to manually generated plans. Automatic VMAT plan generation for cervical cancer has been implemented in our clinical routine. Due to the achieved workload reduction, extension of plan libraries has become feasible. [ABSTRACT FROM AUTHOR]

Background: Using serum carbohydrate antigen 19–9 (CA 19–9) in discriminating between benign and malignant pancreatic disease remains controversial. We aim to evaluate the diagnostic value of serum CA 19–9 in predicting malignant pancreatic cystic lesions. Methods: Eligible studies were identified through searching MEDLINE and EMBASE prior to March 2016. Studies were assessed for quality using the Quality Assessment for Studies of Diagnostic Accuracy, 2nd version (QUADAS-2). Pooled sensitivity and specificity with 95% confidence interval (CI) were calculated using random-effects models. Summary receiver operator characteristic (SROC) curves and the area under curve (AUC) were performed. Results: A total of thirteen studies including 1437 patients were enrolled in this meta-analysis. The pooled sensitivity and specificity were 0.47(95% CI: 0.35–0.59), and 0.88(95% CI: 0.86–0.91), respectively, and the AUC was 0.87(95% CI, 0.84–0.90). Meta-regression analysis showed that sample size, region and reference standards were not the main sources of heterogeneity. Conclusions: Serum CA 19–9 has satisfying pooled specificity while poor pooled sensitivity for discriminating benign from malignant PCNs. It deserves to be widely used as complementary to other clinical diagnostic methods. [ABSTRACT FROM AUTHOR]

Prostate cancer (PCa) is the most common non-cutaneous cancer in male subjects and the second leading cause of cancer-related death in developed countries. The necessity of a non-invasive technique for the diagnosis of PCa in early stage has grown through years. Proton magnetic resonance spectroscopy (1H-MRS) and proton magnetic resonance spectroscopy imaging (1H-MRSI) are advanced magnetic resonance techniques that can mark the presence of metabolites such as citrate, choline, creatine and polyamines in a selected voxel, or in an array of voxels (in MRSI) inside prostatic tissue. Abundance or lack of these metabolites can discriminate between pathological and healthy tissue. Although the use of magnetic resonance spectroscopy (MRS) is well established in brain and liver with dedicated software for spectral analysis, quantification of metabolites in prostate can be very difficult to achieve, due to poor signal to noise ratio and strong J-coupling of the citrate. The aim of this work is to develop a software prototype for automatic quantification of citrate, choline and creatine in prostate. Its core is an original fitting routine that makes use of a fixed step gradient descent minimization algorithm (FSGD) and MRS simulations developed with the GAMMA libraries in C++. The accurate simulation of the citrate spin systems allows to predict the correct J-modulation under different NMR sequences and under different coupling parameters. The accuracy of the quantifications was tested on measurements performed on a Philips Ingenia 3T scanner using homemade phantoms. Some acquisitions in healthy volunteers have been also carried out to test the software performance in vivo. [ABSTRACT FROM AUTHOR]

Breast screening is an important tool for the early detection of breast cancers. However, tumours are typically present in less than 1% of mammograms. This low prevalence could cause radiologists to detect fewer tumours than they otherwise would, an issue known as the prevalence effect. The aim of our study was to investigate a novel breast screening protocol, designed to decrease the number of tumours missed by radiologists, without increasing their workload. We ran two laboratory-based experiments to assess the degree to which the novel protocol, called the catch trial (CT) protocol, resulted in greater sensitivity (d’) than the double screener protocol (DS), currently utilised in Australia. In our first experiment we found evidence that the CT protocol resulted in a criterion shift relative to the DS protocol but the evidence that sensitivity was greater in the CT protocol relative to the DS protocol was less clear. A second experiment, using more realistic stimuli that were more representative of actual tumours, also failed to find convincing evidence that sensitivity was greater in the CT protocol than in the DS protocol. This experiment instead found that both the hit rate and the false alarm rate increased in the CT protocol relative to the DS protocol. So while there was again evidence that the CT protocol induced a criterion shift, the sensitivity appeared to be approximately the same in both protocols. Our results suggest the CT protocol is unlikely to result in an improvement in sensitivity over the DS protocol, so we cannot recommend that it be trialled in a clinical setting. [ABSTRACT FROM AUTHOR]

Background: Glioma remains a diagnostic challenge because of its variable clinical presentation and a lack of reliable screening tools. Long noncoding RNAs (lncRNAs) regulate gene function in a wide range of pathophysiological processes and are therefore emerging biomarkers for prostate cancer, hepatic cancer, and other tumor diseases. However, the effective use of lncRNAs as biomarkers for the diagnosis of glioma remains unproven. Methods: This study included 42 glioma patients and 10 healthy controls. lncRNA and mRNA microarray chips were used to identify dysregulated lncRNAs in tumor tissue and tumor-adjacent normal tissue, and SYBR Green–based miRNA quantitative real-time reverse transcription polymerase chain reactions were used to validate upregulated lncRNAs. A receiver operating characteristic curve analysis was conducted to evaluate the diagnostic accuracy of the lncRNA identified as the candidate biomarker. Results: miR210HG levels were significantly higher in tumor tissue than in tumor-adjacent normal tissue in participating glioma patients. Serum miR210HG levels were also significantly higher in glioma patients than in healthy controls. The receiver operating characteristic curve showed that serum miR210HG was a specific diagnostic predictor of acute pulmonary embolism with an area under the curve of 0.8323 (95% confidence interval, 0.7347 to 0.9299, p < 0.001). Conclusion: Our findings indicate that miR210HG could be an important biomarker for the diagnosis of glioma, and, as such, large-scale investigations are urgently needed to pave the way from basic research to clinical use. [ABSTRACT FROM AUTHOR]

Objective: The study was designed to evaluate diffusion-weighted magnetic resonance imaging (DWI) vs. PET-CT of the thorax in the determination of gross tumor volume (GTV) in radiotherapy planning of non-small-cell lung cancer (NSCLC). Materials and Methods: Eligible patients with NSCLC who were supposed to receive definitive radio(chemo)therapy were prospectively recruited. For MRI, a respiratory gated T2-weighted sequence in axial orientation and non-gated DWI (b = 0, 800, 1,400 and apparent diffusion coefficient map [ADC]) were acquired on a 1.5 Tesla scanner. Primary tumors were delineated on FDG-PET/CT (stGTV) and DWI images (dwGTV). The definition of stGTV was based on the CT and visually adapted to the FDG-PET component if indicated (e.g., in atelectasis). For DWI, dwGTV was visually determined and adjusted for anatomical plausibility on T2w sequences. Beside a statistical comparison of stGTV and dwGTB, spatial agreement was determined with the “Hausdorff-Distance” (HD) and the “Dice Similarity Coefficient” (DSC). Results: Fifteen patients (one patient with two synchronous NSCLC) were evaluated. For 16 primary tumors with UICC stages I (n = 4), II (n = 3), IIIA (n = 2) and IIIB (n = 7) mean values for dwGTV were significantly larger than those of stGTV (76.6 ± 84.5 ml vs. 66.6 ± 75.2 ml, p<0.01). The correlation of stGTV and dwGTV was highly significant (r = 0.995, p<0.001). Yet, some considerable volume deviations between these two methods were observed (median 27.5%, range 0.4–52.1%). An acceptable agreement between dwGTV and stGTV regarding the spatial extent of primary tumors was found (average HD: 2.25 ± 0.7 mm; DC 0.68 ± 0.09). Conclusion: The overall level of agreement between PET-CT and MRI based GTV definition is acceptable. Tumor volumes may differ considerably in single cases. DWI-derived GTVs are significantly, yet modestly, larger than their PET-CT based counterparts. Prospective studies to assess the safety and efficacy of DWI-based radiotherapy planning in NSCLC are warranted. [ABSTRACT FROM AUTHOR]

Background: Image texture analysis is a noninvasive technique for quantifying intratumoral heterogeneity, with derived texture features reported to be closely related to the treatment outcome of tumors. Gastric cancer is one of the most common tumors and the third leading cause of cancer-related deaths worldwide. Although trastuzumab is associated with a survival gain among patients with human epidermal growth factor receptor 2 (HER2)-positive advanced gastric cancer, optimal patient selection is challenging. The purpose of this study was to determine whether CT texture features of HER2-positive gastric cancer were related to the survival rate after trastuzumab treatment. Methods and Findings: Patients diagnosed with HER2-positive advanced gastric cancer from February 2007 to August 2014 were retrospectively selected. Using in-house built software, histogram features (kurtosis and skewness) and gray-level co-occurrence matrices (GLCM) features (angular second moment [ASM], contrast, entropy, variance, and correlation) were derived from the CT images of HER2-positive advanced gastric cancer in 26 patients. All the patients were followed up for more than 6 months, with no confirmed deaths. The patients were dichotomized into a good and poor survival group based on cutoff points of overall survival of 12 months. A receiver-operating characteristics (ROC) analysis was performed to test the ability of each texture parameter to identify the good survival group. Kaplan–Meier curves for patients above and below each threshold were constructed. Using a threshold of >265.8480 for contrast, >488.3150 for variance, and ≤0.1319×10−3. for correlation, all of the area under the ROC curves showed fair accuracy (>0.7). Kaplan–Meier analysis showed statistically significant survival difference between two groups according to optimal cutoff values of contrast, variance, correlation and ASM. However, as this study had a small number of patients, a further study with a larger population will be needed to validate the results. Conclusions: Heterogeneous texture features on CT images were associated with better survival in patients with HER2-positive advanced gastric cancer who received trastuzumab-based treatment. Therefore, texture analysis shows potential to be a clinically useful imaging biomarker providing additional prognostic information for patient selection. [ABSTRACT FROM AUTHOR]

Purpose: The value of adding induction chemotherapy to chemoradiotherapy in locoregionally advanced nasopharyngeal carcinoma (LA-NPC) remains controversial, yet high-risk patients with LA-NPC have poor outcomes after chemoradiotherapy. We aimed to assess the survival benefits of induction chemotherapy in stage IVa-b NPC. Patients and Methods: A total of 602 patients with stage IVa-b NPC treated with intensity-modulated radiation therapy (IMRT) and concurrent chemotherapy with or without induction chemotherapy were retrospectively analyzed. Overall survival (OS), locoregional relapse-free survival (LRFS), distant metastasis-free survival (DMFS) and progression-free survival (PFS) were evaluated using the Kaplan-Meier method, log-rank test and Cox regression analysis. Results: In univariate analysis, 5-year OS was 83.2% for induction chemotherapy plus concurrent chemotherapy and 74.8% for concurrent chemotherapy alone, corresponding to an absolute risk reduction of 8.4% (P = 0.022). Compared to concurrent chemotherapy alone, addition of induction chemotherapy improved 5-year DMFS (83.2% vs. 74.4%, P = 0.018) but not 5-year LRFS (83.7% vs. 83.0%, P = 0.848) or PFS (71.9% vs. 66.0%, P = 0.12). Age, T category, N category, chemotherapy strategy and clinical stage were associated with 5-year OS (P = 0.017, P = 0.031, P = 0.007, P = 0.022, P = 0.001, respectively). In multivariate analysis, induction chemotherapy plus concurrent chemotherapy was an independent favorable prognostic factor for OS (HR, 0.62; 95% CI, 0.43–0.90, P = 0.012) and DMFS (HR, 0.57; 95% CI, 0.38–0.83, P = 0.004). In subgroup analysis, induction chemotherapy significantly improved 5-year DMFS in stage IVa (86.8% vs. 77.3%, P = 0.008), but provided no significant benefit in stage IVb. Conclusions: In patients with stage IVa-b NPC treated with IMRT, addition of induction chemotherapy to concurrent chemotherapy significantly improved 5-year OS and 5-year DMFS. This study provides a basis for selection of high risk patients in future clinical therapeutic trials. [ABSTRACT FROM AUTHOR]

Objectives: To evaluate the performance of MRI for diagnosis of breast cancer in non-calcified equivocal breast findings. Materials and Methods: We performed a systematic review and meta-analysis of peer-reviewed studies in PubMed from 01/01/1986 until 06/15/2015. Eligible were studies applying dynamic contrast-enhanced breast MRI as an adjunct to conventional imaging (mammography, ultrasound) to clarify equivocal findings without microcalcifications. Reference standard for MRI findings had to be established by histopathological sampling or imaging follow-up of at least 12 months. Number of true or false positives and negatives and other characteristics were extracted, and possible bias was determined using the QUADAS-2 applet. Statistical analyses included data pooling and heterogeneity testing. Results: Fourteen out of 514 studies comprising 2,316 lesions met our inclusion criteria. Pooled diagnostic parameters were: sensitivity (99%, 95%-CI: 93–100%), specificity (89%, 95%-CI: 85–92%), PPV (56%, 95%-CI: 42–70%) and NPV (100%, 95%-CI: 99–100%). These estimates displayed significant heterogeneity (P<0.001). Conclusions: Breast MRI demonstrates an excellent diagnostic performance in case of non-calcified equivocal breast findings detected in conventional imaging. However, considering the substantial heterogeneity with regard to prevalence of malignancy, problem solving criteria need to be better defined. [ABSTRACT FROM AUTHOR]

Purpose: This study aimed to investigate the variability of textural features (TF) as a function of acquisition and reconstruction parameters within the context of multi-centric trials. Methods: The robustness of 15 selected TFs were studied as a function of the number of iterations, the post-filtering level, input data noise, the reconstruction algorithm and the matrix size. A combination of several reconstruction and acquisition settings was devised to mimic multi-centric conditions. We retrospectively studied data from 26 patients enrolled in a diagnostic study that aimed to evaluate the performance of PET/CT 68Ga-DOTANOC in gastro-entero-pancreatic neuroendocrine tumors. Forty-one tumors were extracted and served as the database. The coefficient of variation (COV) or the absolute deviation (for the noise study) was derived and compared statistically with SUVmax and SUVmean results. Results: The majority of investigated TFs can be used in a multi-centric context when each parameter is considered individually. The impact of voxel size and noise in the input data were predominant as only 4 TFs presented a high/intermediate robustness against SUV-based metrics (Entropy, Homogeneity, RP and ZP). When combining several reconstruction settings to mimic multi-centric conditions, most of the investigated TFs were robust enough against SUVmax except Correlation, Contrast, LGRE, LGZE and LZLGE. Conclusion: Considering previously published results on either reproducibility or sensitivity against delineation approach and our findings, it is feasible to consider Homogeneity, Entropy, Dissimilarity, HGRE, HGZE and ZP as relevant for being used in multi-centric trials. [ABSTRACT FROM AUTHOR]