Ivana Išgum

Multi-Atlas-Based Segmentation With Local Decision Fusion—Application to Cardiac and Aortic Segmentation in CT Scans

A novel atlas-based segmentation approach based on the combination of multiple registrations is presented. Multiple atlases are registered to a target image. To obtain a segmentation of the target, labels of the atlas images are propagated to it. The propagated labels are combined by spatially varying decision fusion weights. These weights are derived from local assessment of the registration success. Furthermore, an atlas selection procedure is proposed that is equivalent to sequential forward selection from statistical pattern recognition theory. The proposed method is compared to three existing atlas-based segmentation approaches, namely (1) single atlas-based segmentation, (2) average-shape atlas-based segmentation, and (3) multi-atlas-based segmentation with averaging as decision fusion. These methods were tested on the segmentation of the heart and the aorta in computed tomography scans of the thorax. The results show that the proposed method outperforms other methods and yields results very close to those of an independent human observer. Moreover, the additional atlas selection step led to a faster segmentation at a comparable performance.

Automatic Segmentation of MR Brain Images With a Convolutional Neural Network

Automatic segmentation in MR brain images is important for quantitative analysis in large-scale studies with images acquired at all ages. This paper presents a method for the automatic segmentation of MR brain images into a number of tissue classes using a convolutional neural network. To ensure that the method obtains accurate segmentation details as well as spatial consistency, the network uses multiple patch sizes and multiple convolution kernel sizes to acquire multi-scale information about each voxel. The method is not dependent on explicit features, but learns to recognise the information that is important for the classification based on training data. The method requires a single anatomical MR image only. The segmentation method is applied to five different data sets: coronal T2-weighted images of preterm infants acquired at 30 weeks postmenstrual age (PMA) and 40 weeks PMA, axial T2-weighted images of preterm infants acquired at 40 weeks PMA, axial T1-weighted images of ageing adults acquired at an average age of 70 years, and T1-weighted images of young adults acquired at an average age of 23 years. The method obtained the following average Dice coefficients over all segmented tissue classes for each data set, respectively: 0.87, 0.82, 0.84, 0.86, and 0.91. The results demonstrate that the method obtains accurate segmentations in all five sets, and hence demonstrates its robustness to differences in age and acquisition protocol.Automatic segmentation in MR brain images is important for quantitative analysis in large-scale studies with images acquired at all ages. This paper presents a method for the automatic segmentation of MR brain images into a number of tissue classes using a convolutional neural network. To ensure that the method obtains accurate segmentation details as well as spatial consistency, the network uses multiple patch sizes and multiple convolution kernel sizes to acquire multi-scale information about each voxel. The method is not dependent on explicit features, but learns to recognise the information that is important for the classification based on training data. The method requires a single anatomical MR image only. The segmentation method is applied to five different data sets: coronal T2-weighted images of preterm infants acquired at 30 weeks postmenstrual age (PMA) and 40 weeks PMA, axial T2-weighted images of preterm infants acquired at 40 weeks PMA, axial T1-weighted images of ageing adults acquired at an average age of 70 years, and T1-weighted images of young adults acquired at an average age of 23 years. The method obtained the following average Dice coefficients over all segmented tissue classes for each data set, respectively: 0.87, 0.82, 0.84, 0.86, and 0.91. The results demonstrate that the method obtains accurate segmentations in all five sets, and hence demonstrates its robustness to differences in age and acquisition protocol.

Adaptive local multi-atlas segmentation: Application to the heart and the caudate nucleus

Atlas-based segmentation is a powerful generic technique for automatic delineation of structures in volumetric images. Several studies have shown that multi-atlas segmentation methods outperform schemes that use only a single atlas, but running multiple registrations on volumetric data is time-consuming. Moreover, for many scans or regions within scans, a large number of atlases may not be required to achieve good segmentation performance and may even deteriorate the results. It would therefore be worthwhile to include the decision which and how many atlases to use for a particular target scan in the segmentation process. To this end, we propose two generally applicable multi-atlas segmentation methods, adaptive multi-atlas segmentation (AMAS) and adaptive local multi-atlas segmentation (ALMAS). AMAS automatically selects the most appropriate atlases for a target image and automatically stops registering atlases when no further improvement is expected. ALMAS takes this concept one step further by locally deciding how many and which atlases are needed to segment a target image. The methods employ a computationally cheap atlas selection strategy, an automatic stopping criterion, and a technique to locally inspect registration results and determine how much improvement can be expected from further registrations. AMAS and ALMAS were applied to segmentation of the heart in computed tomography scans of the chest and compared to a conventional multi-atlas method (MAS). The results show that ALMAS achieves the same performance as MAS at a much lower computational cost. When the available segmentation time is fixed, both AMAS and ALMAS perform significantly better than MAS. In addition, AMAS was applied to an online segmentation challenge for delineation of the caudate nucleus in brain MRI scans where it achieved the best score of all results submitted to date.

CT-quantified emphysema in male heavy smokers: association with lung function decline

Background Emphysema and small airway disease both contribute to chronic obstructive pulmonary disease (COPD), a disease characterised by accelerated decline in lung function. The association between the extent of emphysema in male current and former smokers and lung function decline was investigated. Methods Current and former heavy smokers participating in a lung cancer screening trial were recruited to the study and all underwent CT. Spirometry was performed at baseline and at 3-year follow-up. The 15th percentile (Perc15) was used to assess the severity of emphysema. Results 2085 men of mean age 59.8 years participated in the study. Mean (SD) baseline Perc15 was −934.9 (19.5) HU. A lower Perc15 value correlated with a lower forced expiratory volume in 1 s (FEV1) at baseline (r=0.12, p<0.001). Linear mixed model analysis showed that a lower Perc15 was significantly related to a greater decline in FEV1 after follow-up (p<0.001). Participants without baseline airway obstruction who developed it after follow-up had significantly lower mean (SD) Perc15 values at baseline than those who did not develop obstruction (−934.2 (17.1) HU vs −930.2 (19.7) HU, p<0.001). Conclusion Greater baseline severity of CT-detected emphysema is related to lower baseline lung function and greater rates of lung function decline, even in those without airway obstruction. CT-detected emphysema aids in identifying non-obstructed male smokers who will develop airflow obstruction.

Comparing coronary artery calcium and thoracic aorta calcium for prediction of all-cause mortality and cardiovascular events on low-dose non-gated computed tomography in a high-risk population of heavy smokers.

BACKGROUND Coronary artery calcium (CAC) and thoracic aorta calcium (TAC) can be detected simultaneously on low-dose, non-gated computed tomography (CT) scans. CAC has been shown to predict cardiovascular (CVD) and coronary (CHD) events. A comparable association between TAC and CVD events has yet to be established, but TAC could be a more reproducible alternative to CAC in low-dose, non-gated CT. This study compared CAC and TAC as independent predictors of all-cause mortality and cardiovascular events in a population of heavy smokers using low-dose, non-gated CT. METHODS Within the NELSON study, a population-based lung cancer screening trial, the CT screen group consisted of 7557 heavy smokers aged 50-75 years. Using a case-cohort study design, CAC and TAC scores were calculated in a total of 958 asymptomatic subjects who were followed up for all-cause death, and CVD, CHD and non-cardiac events (stroke, aortic aneurysm, peripheral arterial occlusive disease). We used Cox proportional-hazard regression to compute hazard ratios (HRs) with adjustment for traditional cardiovascular risk factors. RESULTS A close association between the prevalence of TAC and increasing levels of CAC was established (p<0.001). Increasing CAC and TAC risk categories were associated with all-cause mortality (p for trend=0.01 and 0.001, respectively) and CVD events (p for trend <0.001 and 0.03, respectively). Compared with the lowest quartile (reference category), multivariate-adjusted HRs across categories of CAC were higher (all-cause mortality, HR: 9.13 for highest quartile; CVD events, HR: 4.46 for highest quartile) than of TAC scores (HR: 5.45 and HR: 2.25, respectively). However, TAC is associated with non-coronary events (HR: 4.69 for highest quartile, p for trend=0.01) and CAC was not (HR: 3.06 for highest quartile, p for trend=0.40). CONCLUSIONS CAC was found to be a stronger predictor than TAC of all-cause mortality and CVD events in a high-risk population of heavy smokers scored on low-dose, non-gated CT. TAC, however, is stronger associated with non-cardiac events than CAC and could prove to be a preferred marker for these events.

Identification of chronic obstructive pulmonary disease in lung cancer screening computed tomographic scans.

CONTEXT Smoking is a major risk factor for both cancer and chronic obstructive pulmonary disease (COPD). Computed tomography (CT)-based lung cancer screening may provide an opportunity to detect additional individuals with COPD at an early stage. OBJECTIVE To determine whether low-dose lung cancer screening CT scans can be used to identify participants with COPD. DESIGN, SETTING, AND PATIENTS Single-center prospective cross-sectional study within an ongoing lung cancer screening trial. Prebronchodilator pulmonary function testing with inspiratory and expiratory CT on the same day was obtained from 1140 male participants between July 2007 and September 2008. Computed tomographic emphysema was defined as percentage of voxels less than -950 Hounsfield units (HU), and CT air trapping was defined as the expiratory:inspiratory ratio of mean lung density. Chronic obstructive pulmonary disease was defined as the ratio of forced expiratory volume in the first second to forced vital capacity (FEV(1)/FVC) of less than 70%. Logistic regression was used to develop a diagnostic prediction model for airflow limitation. MAIN OUTCOME MEASURES Diagnostic accuracy of COPD diagnosis using pulmonary function tests as the reference standard. RESULTS Four hundred thirty-seven participants (38%) had COPD according to lung function testing. A diagnostic model with CT emphysema, CT air trapping, body mass index, pack-years, and smoking status corrected for overoptimism (internal validation) yielded an area under the receiver operating characteristic curve of 0.83 (95% CI, 0.81-0.86). Using the point of optimal accuracy, the model identified 274 participants with COPD with 85 false-positives, a sensitivity of 63% (95% CI, 58%-67%), specificity of 88% (95% CI, 85%-90%), positive predictive value of 76% (95% CI, 72%-81%); and negative predictive value of 79% (95% CI, 76%-82%). The diagnostic model showed an area under the receiver operating characteristic curve of 0.87 (95% CI, 0.86-0.88) for participants with symptoms and 0.78 (95% CI, 0.76-0.80) for those without symptoms. CONCLUSION Among men who are current and former heavy smokers, low-dose inspiratory and expiratory CT scans obtained for lung cancer screening can identify participants with COPD, with a sensitivity of 63% and a specificity of 88%.

Coronary artery calcium can predict all-cause mortality and cardiovascular events on low-dose ct screening for lung cancer

OBJECTIVE Performing coronary artery calcium (CAC) screening as part of low-dose CT lung cancer screening has been proposed as an efficient strategy to detect people with high cardiovascular risk and improve outcomes of primary prevention. This study aims to investigate whether CAC measured on low-dose CT in a population of former and current heavy smokers is an independent predictor of all-cause mortality and cardiac events. SUBJECTS AND METHODS We used a case-cohort study and included 958 subjects 50 years old or older within the screen group of a randomized controlled lung cancer screening trial. We used Cox proportional-hazard models to compute hazard ratios (HRs) adjusted for traditional cardiovascular risk factors to predict all-cause mortality and cardiovascular events. RESULTS During a median follow-up of 21.5 months, 56 deaths and 127 cardiovascular events occurred. Compared with a CAC score of 0, multivariate-adjusted HRs for all-cause mortality for CAC scores of 1-100, 101-1000, and more than 1000 were 3.00 (95% CI, 0.61-14.93), 6.13 (95% CI, 1.35-27.77), and 10.93 (95% CI, 2.36-50.60), respectively. Multivariate-adjusted HRs for coronary events were 1.38 (95% CI, 0.39-4.90), 3.04 (95% CI, 0.95-9.73), and 7.77 (95% CI, 2.44-24.75), respectively. CONCLUSION This study shows that CAC scoring as part of low-dose CT lung cancer screening can be used as an independent predictor of all-cause mortality and cardiovascular events.

Detection of coronary calcifications from computed tomography scans for automated risk assessment of coronary artery disease

A fully automated method for coronary calcification detection from non-contrast-enhanced, ECG-gated multi-slice computed tomography (CT) data is presented. Candidates for coronary calcifications are extracted by thresholding and component labeling. These candidates include coronary calcifications, calcifications in the aorta and in the heart, and other high-density structures such as noise and bone. A dedicated set of 64 features is calculated for each candidate object. They characterize the objects spatial position relative to the heart and the aorta, for which an automatic segmentation scheme was developed, its size and shape, and its appearance, which is described by a set of approximated Gaussian derivatives for which an efficient computational scheme is presented. Three classification strategies were designed. The first one tested direct classification without feature selection. The second approach also utilized direct classification, but with feature selection. Finally, the third scheme employed two-stage classification. In a computationally inexpensive first stage, the most easily recognizable false positives were discarded. The second stage discriminated between more difficult to separate coronary calcium and other candidates. Performance of linear, quadratic, nearest neighbor, and support vector machine classifiers was compared. The method was tested on 76 scans containing 275 calcifications in the coronary arteries and 335 calcifications in the heart and aorta. The best performance was obtained employing a two-stage classification system with a k-nearest neighbor (k-NN) classifier and a feature selection scheme. The method detected 73.8% of coronary calcifications at the expense of on average 0.1 false positives per scan. A calcium score was computed for each scan and subjects were assigned one of four risk categories based on this score. The method assigned the correct risk category to 93.4% of all scans.

Brain tissue volumes in preterm infants: prematurity, perinatal risk factors and neurodevelopmental outcome: A systematic review

Objective: To evaluate the clinical value of neonatal brain tissue segmentation in preterm infants according to the literature. Methods: A structured literature search was undertaken in MEDLINE/Pubmed. This included all publications on volumetric brain tissue assessment in preterm infants at term-equivalent age (TEA) compared to brain tissue volumes of term-born infants, related to perinatal risk factors or related to neurodevelopmental outcome. Results: Sixteen prospective cohort studies, described in 30 articles, fulfilled the criteria. Preterm infants displayed total and regional brain tissue alterations compared to healthy, term-born controls. These alterations seemed more prominent with decreasing gestational age. White matter injury, intraventricular haemorrhage, postnatal corticosteroid therapy, intra-uterine growth retardation and chronic lung disease were frequently associated with volume changes. Associations between volume alterations at TEA and neurodevelopmental outcome in early childhood were shown in a few studies. Conclusions: Preterm birth is associated with brain tissue volume alterations that become more pronounced in the presence of perinatal risk factors and white matter injury. Moreover, associations between volumetric alterations as early as TEA and long-term neurodevelopmental impairments are scarce.

Automatic Coronary Calcium Scoring in Low-Dose Chest Computed Tomography

The calcium burden as estimated from non-ECG-synchronized computed tomography (CT) exams acquired in screening of heavy smokers has been shown to be a strong predictor of cardiovascular events. We present a method for automatic coronary calcium scoring with low-dose, non-contrast-enhanced, non-ECG-synchronized chest CT. First, a probabilistic coronary calcium map was created using multi-atlas segmentation. This map assigned an a priori probability for the presence of coronary calcifications at every location in a scan. Subsequently, a statistical pattern recognition system was designed to identify coronary calcifications by texture, size, and spatial features; the spatial features were computed using the coronary calcium map. The detected calcifications were quantified in terms of volume and Agatston score. The best results were obtained by merging the results of three different supervised classification systems, namely direct classification with a nearest neighbor classifier, and two-stage classification with nearest neighbor and support vector machine classifiers. We used a total of 231 test scans containing 45 674mm of coronary calcifications. The presented method detected on average 157/198mm (sensitivity 79.2%) of coronary calcium volume with on average 4 mm false positive volume. Calcium scoring can be performed automatically in low-dose, noncontrast enhanced, non-ECG-synchronized chest CT in screening of heavy smokers to identify subjects who might benefit from preventive treatment.