Carlos S. Mendoza

Extraction of Airways From CT (EXACT'09)

This paper describes a framework for establishing a reference airway tree segmentation, which was used to quantitatively evaluate fifteen different airway tree extraction algorithms in a standardized manner. Because of the sheer difficulty involved in manually constructing a complete reference standard from scratch, we propose to construct the reference using results from all algorithms that are to be evaluated. We start by subdividing each segmented airway tree into its individual branch segments. Each branch segment is then visually scored by trained observers to determine whether or not it is a correctly segmented part of the airway tree. Finally, the reference airway trees are constructed by taking the union of all correctly extracted branch segments. Fifteen airway tree extraction algorithms from different research groups are evaluated on a diverse set of twenty chest computed tomography (CT) scans of subjects ranging from healthy volunteers to patients with severe pathologies, scanned at different sites, with different CT scanner brands, models, and scanning protocols. Three performance measures covering different aspects of segmentation quality were computed for all participating algorithms. Results from the evaluation showed that no single algorithm could extract more than an average of 74% of the total length of all branches in the reference standard, indicating substantial differences between the algorithms. A fusion scheme that obtained superior results is presented, demonstrating that there is complementary information provided by the different algorithms and there is still room for further improvements in airway segmentation algorithms.

Fast parameter-free region growing segmentation with application to surgical planning

In this paper, we propose a self-assessed adaptive region growing segmentation algorithm. In the context of an experimental virtual-reality surgical planning software platform, our method successfully delineates main tissues relevant for reconstructive surgery, such as fat, muscle, and bone. We rely on a self-tuning approach to deal with a great variety of imaging conditions requiring limited user intervention (one seed). The detection of the optimal parameters is managed internally using a measure of the varying contrast of the growing region, and the stopping criterion is adapted to the noise level in the dataset thanks to the sampling strategy used for the assessment function. Sampling is referred to the statistics of a neighborhood around the seed(s), so that the sampling period becomes greater when images are noisier, resulting in the acquisition of a lower frequency version of the contrast function. Validation is provided for synthetic images, as well as real CT datasets. For the CT test images, validation is referred to manual delineations for 10 cases and to subjective assessment for another 35. High values of sensitivity and specificity, as well as Dice’s coefficient and Jaccard’s index on one hand, and satisfactory subjective evaluation on the other hand, prove the robustness of our contrast-based measure, even suggesting suitability for calibration of other region-based segmentation algorithms.

Automatic Analysis of Pediatric Renal Ultrasound Using Shape, Anatomical and Image Acquisition Priors

In this paper we present a segmentation method for ultrasound (US) images of the pediatric kidney, a difficult and barely studied problem. Our method segments the kidney on 2D sagittal US images and relies on minimal user intervention and a combination of improvements made to the Active Shape Model (ASM) framework. Our contributions include particle swarm initialization and profile training with rotation correction. We also introduce our methodology for segmentation of the kidneys collecting system (CS), based on graph-cuts (GC) with intensity and positional priors. Our intensity model corrects for intensity bias by comparison with other biased versions of the most similar kidneys in the training set. We prove significant improvements (p < 0.001) with respect to classic ASM and GC for kidney and CS segmentation, respectively. We use our semi-automatic method to compute the hydronephrosis index (HI) with an average error of 2.67 +/- 5.22 percentage points similar to the error of manual HI between different operators of 2.31 +/- 4.54 percentage points.

What's in a Name? Accurately Diagnosing Metopic Craniosynostosis Using a Computational Approach.

Background: The metopic suture is unlike other cranial sutures in that it normally closes in infancy. Consequently, the diagnosis of metopic synostosis depends primarily on a subjective assessment of cranial shape. The purpose of this study was to create a simple, reproducible radiographic method to quantify forehead shape and distinguish trigonocephaly from normal cranial shape variation. Methods: Computed tomography scans were acquired for 92 control patients (mean age, 4.2 ± 3.3 months) and 18 patients (mean age, 6.2 ± 3.3 months) with a diagnosis of metopic synostosis. A statistical model of the normal cranial shape was constructed, and deformation fields were calculated for patients with metopic synostosis. Optimal and divergence (simplified) interfrontal angles (IFA) were defined based on the three points of maximum average deformation on the frontal bones and metopic suture, respectively. Statistical analysis was performed to assess the accuracy and reliability of the diagnostic procedure. Results: The optimal interfrontal angle was found to be significantly different between the synostosis (116.5 ± 5.8 degrees; minimum, 106.8 degrees; maximum, 126.6 degrees) and control (136.7 ± 6.2 degrees; minimum, 123.8 degrees; maximum, 169.3 degrees) groups (p < 0.001). Divergence interfrontal angles were also significantly different between groups. Accuracy, in terms of available clinical diagnosis, for the optimal and divergent angles, was 0.981 and 0.954, respectively. Conclusions: Cranial shape analysis provides an objective and extremely accurate measure by which to diagnose abnormal interfrontal narrowing, the hallmark of metopic synostosis. The simple planar angle measurement proposed is reproducible and accurate, and can eliminate diagnostic subjectivity in this disorder. CLINICAL QUESTION/LEVEL OF EVIDENCE: Diagnostic, IV.

Kidney segmentation in ultrasound via genetic initialization and Active Shape Models with rotation correction

In this paper we present a segmentation method for 2D ultrasound images of the pediatric kidney. Our method relies on minimal user intervention and produces accurate segmentations thanks to a combination of improvements made to the Active Shape Model (ASM) framework. The initialization of the ASM module is based on a Covariance Matrix Adaptation Evolution Strategy (CMA-ES) genetic algorithm that optimizes the pose and the main shape variation modes of the kidney shape model. In order to account for the image formation process in ultrasound, the appearance model is obtained not according to the anatomically corresponding contour landmarks, but to those that exhibit a similar angle of incidence with respect to the wavefront traveling from the probe. The results indicate a median Dices coefficient of 90.2% and a relative area difference of 10.8% for segmentation of a set of 80 kidney images.

Scale invariant descriptors in pattern analysis of melanocytic lesions

In this paper we introduce the importance of scale invariance in properly discriminating some of the typical patterns found in melanocytic lesions, by dermatoscopic image analysis. Pattern discrimination is a necessary step before pattern irregularity (an indicator of malignancy) can be quantified. We propose a set of features that allows for the discrimination of such patterns even when they appear in different degrees of magnification. We show how an automated feature selection stage produces a preferred scale invariant set of features among non-invariant features, yielding the best classification rate for those features. The average correct classification rate for the five kinds of classified patterns rises up to 94%.

Pattern Analysis of Dermoscopic Images Based on FSCM Color Markov Random Fields

In this paper a method for pattern analysis in dermoscopic images of abnormally pigmented skin (melanocytic lesions) is presented. In order to diagnose a possible skin cancer, physicians assess the lesion according to different rules. The new trend in Dermatology is to classify the lesion by means of pattern irregularity. In order to analyze the pattern turbulence, lesions ought to be segmented into single pattern regions. Our classification method, when applied on overlapping lesion patches, provides a pattern chart that could ultimately allow for in-region single-texture turbulence analysis. Due to the color-textured appearance of these patterns, we present a novel method based on a Finite Symmetric Conditional Model (FSCM) Markov Random Field (MRF) color extension for the characterization and discrimination of pattern samples. Our classification success rate rises to 86%.

Self-assessed Contrast-Maximizing Adaptive Region Growing

In the context of an experimental virtual-reality surgical planning software platform, we propose a fully self-assessed adaptive region growing segmentation algorithm. Our method successfully delineates main tissues relevant to head and neck reconstructive surgery, such as skin, fat, muscle/organs, and bone. We rely on a standardized and self-assessed region-based approach to deal with a great variety of imaging conditions with minimal user intervention, as only a single-seed selection stage is required. The detection of the optimal parameters is managed internally using a measure of the varying contrast of the growing regions. Validation based on synthetic images, as well as truly-delineated real CT volumes, is provided for the reader’s evaluation.

Linearized Multidimensional Earth-Mover's-Distance Gradient Flows

This paper presents the first framework capable of performing active contour segmentation using Earth Movers Distance (EMD) to measure dissimilarity between multidimensional feature distributions. EMD is the best known and understood cross-bin histogram distance measure, and as such it allows for meaningful comparisons between distributions, unlike bin-to-bin measures that only account for discrepancies on a bin-to-bin basis. Because EMD is obtained with linear programming techniques, its differential structure with respect to variations in bin weights as the active contour evolves is expressed through sensitivity analysis. Euler-Lagrange equations are then derived from the computed sensitivity at every iteration to produce gradient descent flows. We validate our approach with color image segmentation, in comparison with state-of-the-art Bhattacharyya (bin-to-bin) and 1D EMD (cross-bin) active contours. Some unique advantages of cross-bin comparison are highlighted in our segmentation results: better perceptual value and increased robustness with respect to the initialization.

An optimal set of landmarks for metopic craniosynostosis diagnosis from shape analysis of pediatric CT scans of the head

Craniosynostosis (premature fusion of skull sutures) is a severe condition present in one of every 2000 newborns. Metopic craniosynostosis, accounting for 20-27% of cases, is diagnosed qualitatively in terms of skull shape abnormality, a subjective call of the surgeon. In this paper we introduce a new quantitative diagnostic feature for metopic craniosynostosis derived optimally from shape analysis of CT scans of the skull. We built a robust shape analysis pipeline that is capable of obtaining local shape differences in comparison to normal anatomy. Spatial normalization using 7-degree-of-freedom registration of the base of the skull is followed by a novel bone labeling strategy based on graph-cuts according to labeling priors. The statistical shape model built from 94 normal subjects allows matching a patients anatomy to its most similar normal subject. Subsequently, the computation of local malformations from a normal subject allows characterization of the points of maximum malformation on each of the frontal bones adjacent to the metopic suture, and on the suture itself. Our results show that the malformations at these locations vary significantly (p<0.001) between abnormal/normal subjects and that an accurate diagnosis can be achieved using linear regression from these automatic measurements with an area under the curve for the receiver operating characteristic of 0.97.