Sophie Ribes

The retrospective binning method improves the consistency of phase binning in respiratory-gated PET/CT

This study assesses the accuracy of prospective phase-gated PET/CT data binning and presents a retrospective data binning method that improves image quality and consistency. Respiratory signals from 17 patients who underwent 4D PET/CT were analysed to evaluate the reproducibility of temporal triggers used for the standard phase-based gating method. Breathing signals were reprocessed to implement retrospective PET data binning. The mean and standard deviation of time lags between automatic triggers provided by the Real-time Position Management (RPM, Varian) gating device and inhalation peaks derived from respiratory curves were computed for each patient. The total number of respiratory cycles available for 4D PET/CT according to the binning mode (prospective versus retrospective) was compared. The maximum standardized uptake value (SUV(max)), biological tumour volume (BTV) and tumour trajectory measures were determined from the PET/CT images of five patients. Compared to retrospective binning (RB), prospective gating approach led to (i) a significant loss in breathing cycles (15%) and (ii) the inconsistency of data binning due to temporal dispersion of triggers (average 396 ms). Consequently, tumour characterization could be impacted. In retrospective mode, SUV(max) was up to 27% higher, where no significant difference appeared in BTV. In addition, prospective mode gave an inconsistent spatial location of the tumour throughout the bins. Improved consistency with breathing patterns and greater motion amplitude of the tumour centroid were observed with retrospective mode. The detection of the tumour motion and trajectory was improved also for small temporal dispersion of triggers. This study shows that the binning mode could have a significant impact on 4D PET images. The consistency of triggers with breathing signals should be checked before clinical use of gated PET/CT images, and our RB method improves 4D PET/CT image quantification.

A new respiratory gating device to improve 4D PET/CT

PURPOSEnRespiratory motion creates artifacts in positon emission tomography with computed tomography (PET/CT) images especially for lung tumors, and can alter diagnosis. To account for motion effects, respiratory gating techniques have been developed. However, the lack of measures strongly correlated with tumor motion limits their accuracy. The authors developed a real-time pneumotachograph device (SPI) allowing to sort PET and CT images depending on lung volumes.nnnMETHODSnThe performance of this innovative respiratory tracking system was characterized and compared to a standard system. Our experimental setup consisted in a movable platform and a thorax phantom with six fillable spheres simulating lung tumors. The accuracy of SPI to detect inhalation peaks was also determined on volunteers. A comparison with the real-time position management (RPM) device, that relies on abdominal height measurement, was then investigated.nnnRESULTSnExperiments showed a high accuracy of the measured signal compared to the input signal (R = 0.88 to 0.99), and of the detection of the inhalation peaks (error of 0.1 +/- 5.8 ms) necessary for prospective binning mode. Activity recovery coefficient was improved (until +39%) and the smearing effect was reduced (until 2.74 times lower) with SPI compared to ungated PET/CT acquisition. The spatial distribution of activity in spheres was similar for 4D PET gated with SPI and RPM. Significant improvement of the binning stability and matching between PET and CT were highlighted for irregular breathing patterns with SPI.nnnCONCLUSIONSnSPI is an innovative device that provides better binning performance than the current gating device on phantom experiments. Future works will focus on patients where the authors expect a significant improvement of specificity and sensitivity of PET/CT examinations.

Automatic segmentation of breast MR images through a Markov random field statistical model.

An algorithm dedicated to automatic segmentation of breast magnetic resonance images is presented in this paper. Our approach is based on a pipeline that includes a denoising step and statistical segmentation. The noise removal preprocessing relies on an anisotropic diffusion scheme, whereas the statistical segmentation is conducted through a Markov random field model. The continuous updating of all parameters governing the diffusion process enables automatic denoising, and the partial volume effect is also addressed during the labeling step. To assess the relevance, the Jaccard similarity coefficient was computed. Experiments were conducted on synthetic data and breast magnetic resonance images extracted from a high-risk population. The relevance of the approach for the dataset is highlighted, and we demonstrate accuracy superior to that of traditional clustering algorithms. The results emphasize the benefits of both denoising guided by input data and the inclusion of spatial dependency through a Markov random field. For example, the Jaccard coefficient for the clinical data was increased by 114%, 109%, and 140% with respect to a K-means algorithm and, respectively, for the adipose, glandular and muscle and skin components. Moreover, the agreement between the manual segmentations provided by an experienced radiologist and the automatic segmentations performed with this algorithm was good, with Jaccard coefficients equal to 0.769, 0.756, and 0.694 for the above-mentioned classes.

Comparison of an alternative and existing binning methods to reduce the acquisition duration of 4D PET/CT.

PURPOSEnRespiratory motion is a source of artifacts that reduce image quality in PET. Four dimensional (4D) PET/CT is one approach to overcome this problem. Existing techniques to limiting the effects of respiratory motions are based on prospective phase binning which requires a long acquisition duration (15-25 min). This time is uncomfortable for the patients and limits the clinical exploitation of 4D PET/CT. In this work, the authors evaluated an existing method and an alternative retrospective binning method to reduce the acquisition duration of 4D PET/CT.nnnMETHODSnThe authors studied an existing mixed-amplitude binning (MAB) method and an alternative binning method by mixed-phases (MPhB). Before implementing MPhB, they analyzed the regularity of the breathing patterns in patients. They studied the breathing signal drift and missing CT slices that could be challenging for implementing MAB. They compared the performance of MAB and MPhB with current binning methods to measure the maximum uptake, internal volume, and maximal range of tumor motion.nnnRESULTSnMPhB can be implemented depending on an optimal phase (in average, the exhalation peak phase -4.1% of the entire breathing cycle duration). Signal drift of patients was in average 35% relative to the breathing amplitude. Even after correcting this drift, MAB was feasible in 4D CT for only 64% of patients. No significant differences appeared between the different binning methods to measure the maximum uptake, internal volume, and maximal range of tumor motion. The authors also determined the inaccuracies of MAB and MPhB to measure the maximum amplitude of tumor motion with three bins (less than 3 mm for movement inferior to 12 mm, up to 6.4 mm for a 21 mm movement).nnnCONCLUSIONSnThe authors proposed an alternative binning method by mixed-phase binning that halves the acquisition duration of 4D PET/CT. Mixed-amplitude binning was challenging because of signal drift and missing CT slices. They showed that more than three bins were necessary for a more accurate measurement of the maximum amplitude of the tumor motion. However, the current 4D-CT technology limits the increase of the number of bins in 4D PET/CT because of missing CT slices. One can reconstruct 4D PET images with more bins but without attenuation/scatter correction.

Support-vector-machines-based multidimensional signal classification for fetal activity characterization

Electronic fetal monitoring may be required during the whole pregnancy to closely monitor specific fetal and maternal disorders. Currently used methods suffer from many limitations and are not sufficient to evaluate fetal asphyxia. Fetal activity parameters such as movements, heart rate and associated parameters are essential indicators of the fetus well being, and no current device gives a simultaneous and sufficient estimation of all these parameters to evaluate the fetus well-being. We built for this purpose, a multi-transducer-multi-gate Doppler system and developed dedicated signal processing techniques for fetal activity parameter extraction in order to investigate fetuss asphyxia or well-being through fetal activity parameters. To reach this goal, this paper shows preliminary feasibility of separating normal and compromised fetuses using our system. To do so, data set consisting of two groups of fetal signals (normal and compromised) has been established and provided by physicians. From estimated parameters an instantaneous Manning-like score, referred to as ultrasonic score was introduced and was used together with movements, heart rate and associated parameters in a classification process using Support Vector Machines (SVM) method. The influence of the fetal activity parameters and the performance of the SVM were evaluated using the computation of sensibility, specificity, percentage of support vectors and total classification accuracy. We showed our ability to separate the data into two sets : normal fetuses and compromised fetuses and obtained an excellent matching with the clinical classification performed by physician.

Multidimensional Ultrasound Doppler Signal Analysis for Fetal Activity Monitoring.

Fetal activity parameters such as movements, heart rate and the related parameters are essential indicators of fetal wellbeing, and no device provides simultaneous access to and sufficient estimation of all of these parameters to evaluate fetal health. This work was aimed at collecting these parameters to automatically separate healthy from compromised fetuses. To achieve this goal, we first developed a multi-sensor-multi-gate Doppler system. Then we recorded multidimensional Doppler signals and estimated the fetal activity parameters via dedicated signal processing techniques. Finally, we combined these parameters into four sets of parameters (or four hyper-parameters) to determine the set of parameters that is able to separate healthy from other fetuses. To validate our system, a data set consisting of two groups of fetal signals (normal and compromised) was established and provided by physicians. From the estimated parameters, an instantaneous Manning-like score, referred to as the ultrasonic score, was calculated and was used together with movements, heart rate and the associated parameters in a classification process employing the support vector machine method. We investigated the influence of the sets of parameters and evaluated the performance of the support vector machine using the computation of sensibility, specificity, percentage of support vectors and total classification error. The sensitivity of the four sets ranged from 79% to 100%. Specificity was 100% for all sets. The total classification error ranged from 0% to 20%. The percentage of support vectors ranged from 33% to 49%. Overall, the best results were obtained with the set of parameters consisting of fetal movement, short-term variability, long-term variability, deceleration and ultrasound score. The sensitivity, specificity, percentage of support vectors and total classification error of this set were respectively 100%, 100%, 35% and 0%. This indicated our ability to separate the data into two sets (normal fetuses and pathologic fetuses), and the results highlight the excellent match with the clinical classification performed by the physicians. This work indicates the feasibility of detecting compromised fetuses and also represents an interesting method of close fetal monitoring during the entire pregnancy.

MO‐D‐141‐11: Registration of Breast Magnetic Resonance Imaging and X‐Ray Mammography Through a Biomechanical Model Based On Clinical Data and a Finite Element Method

PURPOSEnOvercoming the drawbacks of x-ray mammography and magnetic resonance imaging (MRI) by fusing the information in order to assist clinicians in the task of early detection of breast cancer.nnnMETHODSnA detailed 3D computer-generated breast phantom based on empirical data extracted from breast MRI was constructed for each patient. To achieve this goal, MRI data were classified into the different components of breast tissues (glandular, adipose, skin and eventually tumor) using a semi-automated segmentation algorithm based on voxel intensity. Then, a geometrical model of the breast was constructed through the isosurfaces of this segmented volume. In order to perform a study on breast deformation using the finite element method, the geometrical model was automatically meshed into tetrahedral elements and material properties were assigned to the different kinds of breast tissues. To represent the large deformation of breast during a mammography exam, a neo-Hookean hyperelastic model was chosen to describe the constitutive relations of breast tissues, and the compression was simulated using a stiff plate model. After compressing the phantom, mammograms were simulated based on the deformed configuration. During this step, a parametric optimization of the model was conducted (mesh refinement, mechanical properties and friction coefficient).nnnRESULTSnSmall variations of the model parameters strongly influence the deformation and modify significantly the resultant simulated images. During the optimization process, both a better conservation of details and a convergence toward a distribution of components were observed for finer meshes, whereas the friction coefficient affects mostly the skin deformation.nnnCONCLUSIONnThe phantom developed in this study allows the modeling of large deformations through the use of the finite element method, and also allows the simulation of mammographic images containing high-resolution details. Moreover, this phantom combines flexibility and realism, and can be used for multimodality imaging research but also for clinical performance optimization.