Perfusion vector-a new method to quantify myocardial perfusion scintigraphy images: a simulation study with validation in patients.

BACKGROUND: The interpretation of myocardial perfusion scintigraphy (MPS) largely relies on visual assessment by the physician of the localization and extent of a perfusion defect. The aim of this study was to introduce the concept of the perfusion vector as a new objective quantitative method for further assisting the visual interpretation and to test the concept using simulated MPS images as well as patients. METHODS: The perfusion vector is based on calculating the difference between the anatomical centroid and the perfusion center of gravity of the left ventricle. Simulated MPS images were obtained using the SIMIND Monte Carlo program together with XCAT phantom. Four different-sized anterior and four lateral defects were simulated, and perfusion vector components x-, y-, and z-axes were calculated. For the patient study, 40 normal and 80 abnormal studies were included. Perfusion vectors were compared between normal and abnormal (apical, inferior, anterior, and lateral ischemia or infarction) studies and also correlated to the defect size. RESULTS: For simulated anterior defects, the stress perfusion vector component on the y-axis (anterior-inferior direction) increased in proportion to the defect size. For the simulated lateral defects, the stress perfusion vector component on the x-axis (septal-lateral direction) decreased in proportion to the defect size. When comparing normal and abnormal patients, there was a statistically significant difference for the stress perfusion vector on the x-axis for apical and lateral defects; on the y-axis for apical, inferior, and lateral defects; and on the z-axis (basal-apical direction) for apical, anterior, and lateral defects. A significant difference was shown for the difference vector magnitude (stress/rest) between normal and ischemic patients (p = 0.001) but not for patients with infarction. The correlation between the defect size and stress vector magnitude was also found to be significant (p < 0.001). CONCLUSIONS: The concept of the perfusion vector introduced in this study is shown to have potential in assisting the visual interpretation in MPS studies. Further studies are needed to validate the concept in patients.

Area of ischemia assessed by physicians and software packages from myocardial perfusion scintigrams.

BACKGROUND: The European Society of Cardiology recommends that patients with >10% area of ischemia should receive revascularization. We investigated inter-observer variability for the extent of ischemic defects reported by different physicians and by different software tools, and if inter-observer variability was reduced when the physicians were provided with a computerized suggestion of the defects. METHODS: Twenty-five myocardial perfusion single photon emission computed tomography (SPECT) patients who were regarded as ischemic according to the final report were included. Eleven physicians in nuclear medicine delineated the extent of the ischemic defects. After at least two weeks, they delineated the defects again, and were this time provided a suggestion of the defect delineation by EXINI HeartTM (EXINI). Summed difference scores and ischemic extent values were obtained from four software programs. RESULTS: The median extent values obtained from the 11 physicians varied between 8% and 34%, and between 9% and 16% for the software programs. For all 25 patients, mean extent obtained from EXINI was 17.0% (± standard deviation (SD) 14.6%). Mean extent for physicians was 22.6% (± 15.6%) for the first delineation and 19.1% (± 14.9%) for the evaluation where they were provided computerized suggestion. Intra-class correlation (ICC) increased from 0.56 (95% confidence interval (CI) 0.41-0.72) to 0.81 (95% CI 0.71-0.90) between the first and the second delineation, and SD between physicians were 7.8 (first) and 5.9 (second delineation). CONCLUSIONS: There was large variability in the estimated ischemic defect size obtained both from different physicians and from different software packages. When the physicians were provided with a suggested delineation, the inter-observer variability decreased significantly.

Progression of bone metastases in patients with prostate cancer - automated detection of new lesions and calculation of bone scan index.

BACKGROUND: The objective of this study was firstly to develop and evaluate an automated method for the detection of new lesions and changes in bone scan index (BSI) in serial bone scans and secondly to evaluate the prognostic value of the method in a group of patients receiving chemotherapy. METHODS: The automated method for detection of new lesions was evaluated in a group of 266 patients using the classifications by three experienced bone scan readers as a gold standard. The prognostic value of the method was assessed in a group of 31 metastatic hormone-refractory prostate cancer patients who were receiving docetaxel. Cox proportional hazards were used to investigate the association between percentage change in BSI, number of new lesions and overall survival. Kaplan-Meier estimates of the survival function were used to indicate a significant difference between patients with an increase/decrease in BSI or those with two or more new lesions or less than two new lesions. RESULTS: The automated method detected progression defined as two or more new lesions with a sensitivity of 93% and a specificity of 87%. In the treatment group, both BSI changes and the number of new metastases were significantly associated with survival. Two-year survival for patients with increasing and decreasing BSI from baseline to follow-up scans were 18% and 57% (p = 0.03), respectively. Two-year survival for patients fulfilling and not fulfilling the criterion of two or more new lesions was 35% and 38% (n.s.), respectively. CONCLUSIONS: An automated method can be used to calculate the number of new lesions and changes in BSI in serial bone scans. These imaging biomarkers contained prognostic information in a small group of patients with prostate cancer receiving chemotherapy.

Normal stress databases in myocardial perfusion scintigraphy--how many subjects do you need?

BACKGROUND: Commercial normal stress databases in myocardial perfusion scintigraphy (MPS) commonly consist of 30-40 individuals. The aim of the study was to determine how many subjects are needed. METHODS: Four normal stress databases were developed using patients who underwent 99mTc MPS: non-corrected images (NC) for male, NC for female, attenuation-corrected images (AC) for male and AC for female subjects. 126 male and 205 female subjects were included. The normal database was created by alternatingly computing the mean of all normal subjects and normalizing the subjects with respect to this mean, until convergence. Coefficients of variation (CV) were created for increasing number of included patients in the four different normal stress databases. RESULTS: Normal stress databases with < 35 subjects had a high CV. Mean CV -2 standard deviations (SD) decreased with 28% between two and five included subjects, 71% between two and 35 subjects and 83% between two and 100 included subjects for NC man. CONCLUSIONS: We conclude that the commonly used 30-40 individuals for making a normal stress database might not be enough due to the high CV. We propose that normal stress databases should consist of more than 30-40 individuals, preferably more than 50 individuals, both for NC and AC studies.

Automated segmentation of the skeleton in whole-body bone scans: influence of difference in atlas.

AIM: Automated segmentation of the skeleton is the first step for quantitative analysis and computer-aided diagnosis (CAD) of whole-body bone scans. The purpose of this study was to examine the influence of differences in skeletal atlas on the automated segmentation of skeletons in a Japanese patient group. METHODS: The study was based on a bone scan CAD system that included a skeletal atlas obtained using 10 normal bone scans from European patients and 23 normal bone scans from Japanese patients. These were incorporated into the CAD system. The performance of the skeletal segmentation, based on either the European or the Japanese Atlas, was evaluated independently by three observers in a group of 50 randomly selected bone scans from Japanese patients. RESULTS: The skeletal segmentation was classified as correct in 41-44 of the 50 cases by the three observers using the Japanese atlas. The corresponding results were 15-18 of the 50 cases using the European atlas, and this difference was statistically significant (P<0.001). The anatomical areas most commonly classified as not correct were the skull, cervical vertebrae, and ribs. CONCLUSION: Automated segmentation of the skeleton in a Japanese patient group was more successful when the CAD system based on a Japanese atlas was used than when the corresponding system based on a European atlas was used. The results of this study indicate that it is of value to use a skeletal atlas based on normal Japanese bone scans in a CAD system for Japanese patients.

Computer-aided diagnosis system for bone scintigrams from Japanese patients: importance of training database.

AIM: Computer-aided diagnosis (CAD) software for bone scintigrams have recently been introduced as a clinical quality assurance tool. The purpose of this study was to compare the diagnostic accuracy of two CAD systems, one based on a European and one on a Japanese training database, in a group of bone scans from Japanese patients. METHOD: The two CAD software are trained to interpret bone scans using training databases consisting of bone scans with the desired interpretation, metastatic disease or not. One software was trained using 795 bone scans from European patients and the other with 904 bone scans from Japanese patients. The two CAD softwares were evaluated using the same group of 257 Japanese patients, who underwent bone scintigraphy because of suspected metastases of malignant tumors in 2009. The final diagnostic results made by clinicians were used as gold standard. RESULTS: The Japanese CAD software showed a higher specificity and accuracy compared to the European CAD software [81 vs. 57 % (p < 0.05) and 82 vs. 61 % (p < 0.05), respectively]. The sensitivity was 90 % for the Japanese CAD software and 83 % for the European CAD software (n.s). CONCLUSION: The CAD software trained with a Japanese database showed significantly higher performance than the corresponding CAD software trained with a European database for the analysis of bone scans from Japanese patients. These results could at least partly be caused by the physical differences between Japanese and European patients resulting in less influence of attenuation in Japanese patients and possible different judgement of count intensities of hot spots.

A novel automated platform for quantifying the extent of skeletal tumour involvement in prostate cancer patients using the Bone Scan Index.

BACKGROUND: There is little consensus on a standard approach to analysing bone scan images. The Bone Scan Index (BSI) is predictive of survival in patients with progressive prostate cancer (PCa), but the popularity of this metric is hampered by the tedium of the manual calculation. OBJECTIVE: Develop a fully automated method of quantifying the BSI and determining the clinical value of automated BSI measurements beyond conventional clinical and pathologic features. DESIGN, SETTING, AND PARTICIPANTS: We conditioned a computer-assisted diagnosis system identifying metastatic lesions on a bone scan to automatically compute BSI measurements. A training group of 795 bone scans was used in the conditioning process. Independent validation of the method used bone scans obtained ≤3 mo from diagnosis of 384 PCa cases in two large population-based cohorts. An experienced analyser (blinded to case identity, prior BSI, and outcome) scored the BSI measurements twice. We measured prediction of outcome using pretreatment Gleason score, clinical stage, and prostate-specific antigen with models that also incorporated either manual or automated BSI measurements. MEASUREMENTS: The agreement between methods was evaluated using Pearson's correlation coefficient. Discrimination between prognostic models was assessed using the concordance index (C-index). RESULTS AND LIMITATIONS: Manual and automated BSI measurements were strongly correlated (ρ=0.80), correlated more closely (ρ=0.93) when excluding cases with BSI scores≥10 (1.8%), and were independently associated with PCa death (p<0.0001 for each) when added to the prediction model. Predictive accuracy of the base model (C-index: 0.768; 95% confidence interval [CI], 0.702-0.837) increased to 0.794 (95% CI, 0.727-0.860) by adding manual BSI scoring, and increased to 0.825 (95% CI, 0.754-0.881) by adding automated BSI scoring to the base model. CONCLUSIONS: Automated BSI scoring, with its 100% reproducibility, reduces turnaround time, eliminates operator-dependent subjectivity, and provides important clinical information comparable to that of manual BSI scoring.

Small average differences in attenuation corrected images between men and women in myocardial perfusion scintigraphy: a novel normal stress database.

BACKGROUND: The American Society of Nuclear Cardiology and the Society of Nuclear Medicine state that incorporation of attenuation-corrected (AC) images in myocardial perfusion scintigraphy (MPS) will improve image quality, interpretive certainty, and diagnostic accuracy. However, commonly used software packages for MPS usually include normal stress databases for non-attenuation corrected (NC) images but not for attenuation-corrected (AC) images. The aim of the study was to develop and compare different normal stress databases for MPS in relation to NC vs. AC images, male vs. female gender, and presence vs. absence of obesity. The principal hypothesis was that differences in mean count values between men and women would be smaller with AC than NC images, thereby allowing for construction and use of gender-independent AC stress database. METHODS: Normal stress perfusion databases were developed with data from 126 male and 205 female patients with normal MPS. The following comparisons were performed for all patients and separately for normal weight vs. obese patients: men vs. women for AC; men vs. women for NC; AC vs. NC for men; and AC vs. NC for women. RESULTS: When comparing AC for men vs. women, only minor differences in mean count values were observed, and there were no differences for normal weight vs. obese patients. For all other analyses major differences were found, particularly for the inferior wall. CONCLUSIONS: The results support the hypothesis that it is possible to use not only gender independent but also weight independent AC stress databases.

Statistical regularization of deformation fields for atlas-based segmentation of bone scintigraphy images.

The construction and application of statistical models of deformations based on non-rigid image registration methods have gained recent popularity. This paper presents the application of such a model to restricting a general-purpose registration algorithm to anatomically plausible solutions. Specifically, the Morphon registration method is used for atlas-based segmentation of bone scintigraphy images. From a training set of 734 images, a model of characteristic deformation fields is built and used for regularizing the registration of 113 test images. Results show that around 300 training images and 30 principal modes are sufficient for building a useful model. The segmentation succeeded in 106 of 113 test images.

Sparse decomposition and modeling of anatomical shape variation.

Recent advances in statistics have spawned powerful methods for regression and data decomposition that promote sparsity, a property that facilitates interpretation of the results. Sparse models use a small subset of the available variables and may perform as well or better than their full counterparts if constructed carefully. In most medical applications, models are required to have both good statistical performance and a relevant clinical interpretation to be of value. Morphometry of the corpus callosum is one illustrative example. This paper presents a method for relating spatial features to clinical outcome data. A set of parsimonious variables is extracted using sparse principal component analysis, producing simple yet characteristic features. The relation of these variables with clinical data is then established using a regression model. The result may be visualized as patterns of anatomical variation related to clinical outcome. In the present application, landmark-based shape data of the corpus callosum is analyzed in relation to age, gender, and clinical tests of walking speed and verbal fluency. To put the data-driven sparse principal component method into perspective, we consider two alternative techniques, one where features are derived using a model-based wavelet approach, and one where the original variables are regressed directly on the outcome.

A path algorithm for the support vector domain description and its application to medical imaging.

The support vector domain description is a one-class classification method that estimates the distributional support of a data set. A flexible closed boundary function is used to separate trustworthy data on the inside from outliers on the outside. A single regularization parameter determines the shape of the boundary and the proportion of observations that are regarded as outliers. Picking an appropriate amount of regularization is crucial in most applications but is, for computational reasons, commonly limited to a small collection of parameter values. This paper presents an algorithm where the solutions for all possible values of the regularization parameter are computed at roughly the same computational complexity previously required to obtain a single solution. Such a collection of solutions is known as a regularization path. Knowledge of the entire regularization path not only aids model selection, but may also provide new information about a data set. We illustrate this potential of the method in two applications; one where we establish a sensible ordering among a set of corpora callosa outlines, and one where ischemic segments of the myocardium are detected in patients with acute myocardial infarction.

Hierarchical linear modeling (HLM) of longitudinal brain structural and cognitive changes in alcohol-dependent individuals during sobriety.

BACKGROUND: Hierarchical linear modeling (HLM) can reveal complex relationships between longitudinal outcome measures and their covariates under proper consideration of potentially unequal error variances. We demonstrate the application of HLM to the study of magnetic resonance imaging (MRI)-derived brain volume changes and cognitive changes in abstinent alcohol-dependent individuals as a function of smoking status, smoking severity, and drinking quantities. METHODS: Twenty non-smoking recovering alcoholics (nsALC) and 30 age-matched smoking recovering alcoholics (sALC) underwent quantitative MRI and cognitive assessments at 1 week, 1 month, and 7 months of sobriety. Eight non-smoking light drinking controls were studied at baseline and 7 months later. Brain and ventricle volumes at each time point were quantified using MRI masks, while the boundary shift integral method measured volume changes between time points. Using HLM, we modeled volumetric and cognitive outcome measures as a function of cigarette and alcohol use variables. RESULTS: Different hierarchical linear models with unique model structures are presented and discussed. The results show that smaller brain volumes at baseline predict faster brain volume gains, which were also related to greater smoking and drinking severities. Over 7 months of abstinence from alcohol, sALC compared to nsALC showed less improvements in visuospatial learning and memory despite larger brain volume gains and ventricular shrinkage. CONCLUSIONS: Different and unique hierarchical linear models allow assessments of the complex relationships among outcome measures of longitudinal data sets. These HLM applications suggest that chronic cigarette smoking modulates the temporal dynamics of brain structural and cognitive changes in alcoholics during prolonged sobriety.

The entire regularization path for the support vector domain description.

The support vector domain description is a one-class classification method that estimates the shape and extent of the distribution of a data set. This separates the data into outliers, outside the decision boundary, and inliers on the inside. The method bears close resemblance to the two-class support vector machine classifier. Recently, it was shown that the regularization path of the support vector machine is piecewise linear, and that the entire path can be computed efficiently. This paper shows that this property carries over to the support vector domain description. Using our results the solution to the one-class classification can be obtained for any amount of regularization with roughly the same computational complexity required to solve for a particularly value of the regularization parameter. The possibility of evaluating the results for any amount of regularization not only offers more accurate and reliable models, but also makes way for new applications. We illustrate the potential of the method by determining the order of inclusion in the model for a set of corpora callosa outlines.