Awais Mansoor

Deep Learning Guided Partitioned Shape Model for Anterior Visual Pathway Segmentation

Analysis of cranial nerve systems, such as the anterior visual pathway (AVP), from MRI sequences is challenging due to their thin long architecture, structural variations along the path, and low contrast with adjacent anatomic structures. Segmentation of a pathologic AVP (e.g., with low-grade gliomas) poses additional challenges. In this work, we propose a fully automated partitioned shape model segmentation mechanism for AVP steered by multiple MRI sequences and deep learning features. Employing deep learning feature representation, this framework presents a joint partitioned statistical shape model able to deal with healthy and pathological AVP. The deep learning assistance is particularly useful in the poor contrast regions, such as optic tracts and pathological areas. Our main contributions are: 1) a fast and robust shape localization method using conditional space deep learning, 2) a volumetric multiscale curvelet transform-based intensity normalization method for robust statistical model, and 3) optimally partitioned statistical shape and appearance models based on regional shape variations for greater local flexibility. Our method was evaluated on MRI sequences obtained from 165 pediatric subjects. A mean Dice similarity coefficient of 0.779 was obtained for the segmentation of the entire AVP (optic nerve only =0.791) using the leave-one-out validation. Results demonstrated that the proposed localized shape and sparse appearance-based learning approach significantly outperforms current state-of-the-art segmentation approaches and is as robust as the manual segmentation.

Quantitative MRI criteria for optic pathway enlargement in neurofibromatosis type 1

Objective: To determine quantitative size thresholds for enlargement of the optic nerve, chiasm, and tract in children with neurofibromatosis type 1 (NF1). Methods: Children 0.5–18.6 years of age who underwent high-resolution T1-weighted MRI were eligible for inclusion. This consisted of children with NF1 with or without optic pathway gliomas (OPGs) and a control group who did not have other acquired, systemic, or genetic conditions that could alter their anterior visual pathway (AVP). Maximum and average diameter and volume of AVP structures were calculated from reconstructed MRI images. Values above the 95th percentile from the controls were considered the threshold for defining an abnormally large AVP measure. Results: A total of 186 children (controls = 82; NF1noOPG = 54; NF1+OPG = 50) met inclusion criteria. NF1noOPG and NF1+OPG participants demonstrated greater maximum optic nerve diameter and volume, optic chiasm volume, and total brain volume compared to controls (p < 0.05, all comparisons). Total brain volume, rather than age, predicted optic nerve and chiasm volume in controls (p < 0.05). Applying the 95th percentile threshold to all NF1 participants, the maximum optic nerve diameter (3.9 mm) and AVP volumes resulted in few false-positive errors (specificity >80%, all comparisons). Conclusions: Quantitative reference values for AVP enlargement will enhance the development of objective diagnostic criteria for OPGs secondary to NF1.

A Statistical Modeling Approach to Computer-Aided Quantification of Dental Biofilm

Biofilm is a formation of microbial material on tooth substrata. Several methods to quantify dental biofilm coverage have recently been reported in the literature, but at best they provide a semiautomated approach to quantification with significant input from a human grader that comes with the graders bias of what is foreground, background, biofilm, and tooth. Additionally, human assessment indices limit the resolution of the quantification scale; most commercial scales use five levels of quantification for biofilm coverage (0%, 25%, 50%, 75%, and 100%). On the other hand, current state-of-the-art techniques in automatic plaque quantification fail to make their way into practical applications owing to their inability to incorporate human input to handle misclassifications. This paper proposes a new interactive method for biofilm quantification in Quantitative light-induced fluorescence (QLF) images of canine teeth that is independent of the perceptual bias of the grader. The method partitions a QLF image into segments of uniform texture and intensity called superpixels; every superpixel is statistically modeled as a realization of a single 2-D Gaussian Markov random field (GMRF) whose parameters are estimated; the superpixel is then assigned to one of three classes ( background, biofilm, tooth substratum) based on the training set of data. The quantification results show a high degree of consistency and precision. At the same time, the proposed method gives pathologists full control to postprocess the automatic quantification by flipping misclassified superpixels to a different state (background, tooth, biofilm) with a single click, providing greater usability than simply marking the boundaries of biofilm and tooth as done by current state-of-the-art methods.

Optic pathway glioma volume predicts retinal axon degeneration in neurofibromatosis type 1.

Objective: To determine whether tumor size is associated with retinal nerve fiber layer (RNFL) thickness, a measure of axonal degeneration and an established biomarker of visual impairment in children with optic pathway gliomas (OPGs) secondary to neurofibromatosis type 1 (NF1). Methods: Children with NF1-OPGs involving the optic nerve (extension into the chiasm and tracts permitted) who underwent both volumetric MRI analysis and optical coherence tomography (OCT) within 2 weeks of each other were included. Volumetric measurement of the entire anterior visual pathway (AVP; optic nerve, chiasm, and tract) was performed using high-resolution T1-weighted MRI. OCT measured the average RNFL thickness around the optic nerve. Linear regression models evaluated the relationship between RNFL thickness and AVP dimensions and volume. Results: Thirty-eight participants contributed 55 study eyes. The mean age was 5.78 years. Twenty-two participants (58%) were female. RNFL thickness had a significant negative relationship to total AVP volume and total brain volume (p < 0.05, all comparisons). For every 1 mL increase in AVP volume, RNFL thickness declined by approximately 5 microns. A greater AVP volume of OPGs involving the optic nerve and chiasm, but not the tracts, was independently associated with a lower RNFL thickness (p < 0.05). All participants with an optic chiasm volume >1.3 mL demonstrated axonal damage (i.e., RNFL thickness <80 microns). Conclusions: Greater OPG and AVP volume predicts axonal degeneration, a biomarker of vision loss, in children with NF1-OPGs. MRI volumetric measures may help stratify the risk of visual loss from NF1-OPGs.

Semi-automatic assessment of pediatric hydronephrosis severity in 3D ultrasound

Hydronephrosis is the most common abnormal finding in pediatric urology. Thanks to its non-ionizing nature, ultrasound (US) imaging is the preferred diagnostic modality for the evaluation of the kidney and the urinary track. However, due to the lack of correlation of US with renal function, further invasive and/or ionizing studies might be required (e.g., diuretic renograms). This paper presents a computer-aided diagnosis (CAD) tool for the accurate and objective assessment of pediatric hydronephrosis based on morphological analysis of kidney from 3DUS scans. The integration of specific segmentation tools in the system, allows to delineate the relevant renal structures from 3DUS scans of the patients with minimal user interaction, and the automatic computation of 90 anatomical features. Using the washout half time (T1/2) as indicative of renal obstruction, an optimal subset of predictive features is selected to differentiate, with maximum sensitivity, those severe cases where further attention is required (e.g., in the form of diuretic renograms), from the noncritical ones. The performance of this new 3DUS-based CAD system is studied for two clinically relevant T1/2 thresholds, 20 and 30 min. Using a dataset of 20 hydronephrotic cases, pilot experiments show how the system outperforms previous 2D implementations by successfully identifying all the critical cases (100% of sensitivity), and detecting up to 100% (T1/2 = 20 min) and 67% (T1/2 = 30 min) of non-critical ones for T1/2 thresholds of 20 and 30 min, respectively.

Severity quantification of pediatric viral respiratory illnesses in chest X-ray images.

Accurate assessment of severity of viral respiratory illnesses (VRIs) allows early interventions to prevent morbidity and mortality in young children. This paper proposes a novel imaging biomarker framework with chest X-ray image for assessing VRIs severity in infants, developed specifically to meet the distinct challenges for pediatric population. The proposed framework integrates three novel technical contributions: a) lung segmentation using weighted partitioned active shape model, b) obtrusive object removal using graph cut segmentation with asymmetry constraint, and c) severity quantification using information-theoretic heterogeneity measures. This paper presents our pilot experimental results with a dataset of 148 images and the ground-truth severity scores given by a board-certified pediatric pulmonologist, demonstrating the effectiveness and clinical relevance of the presented framework.

Generic method for intensity standardization of medical images using multiscale curvelet representation

Most computer-aided diagnosis (CAD) methods in medical imaging are finely tuned for the settings of training data, i.e., the acquisition protocol and machine settings. Therefore, they may fail to perform optimally on images acquired under a different protocol. Intensity standardization, or mapping the acquired data to a predefined intensity profile, can alleviate this challenge. In this work, we present a generic method for intensity standardization of 2D/3D medical images using localized subband energy scaling of the multi-scale curvelet transform. During the training phase, reference data are first decomposed into scale and orientation localized subbands using the multiscale curvelet transform, followed by calculating a reference energy value for each subband. During the testing stage, the localized energy of each subband is scaled to the reference localized energy value from the training stage through an iterative process. We validated our generic standardization method on 2D chest X-rays (CXR) and 3D T1-weighted MRI sequences acquired using different scanners on a group of both healthy and diseased subjects. A significant improvement (Dice coefficient of 0.91±0.05 versus 0.68 ± 0.13, p-value< 0.001) was obtained in the whole brain segmentation accuracy after standardization. Similarly, for air-trapping quantification, the standardization improved the correlation with the expert visual assessment of air-trapping from CXR from R = 0.32 to R = 0.93. The proposed intensity standardization technique could be adopted as a pre-processing step for 2D and 3D data to improve the accuracy of CAD on data obtained from variable sources.

Construction of a Spatiotemporal Statistical Shape Model of Pediatric Liver from Cross-Sectional Data

This paper proposes a spatiotemporal statistical shape model of a pediatric liver, which has potential applications in computer-aided diagnosis of the abdomen. Shapes are analyzed in the space of a level set function, which has computational advantages over the diffeomorphic framework commonly employed in conventional studies. We first calculate the time-varying average of the mean shape development using a kernel regression technique with adaptive bandwidth. Then, eigenshape modes for every timepoint are calculated using principal component analysis with an additional regularization term that ensures the smoothness of the temporal change of the eigenshape modes. To further improve the performance, we applied data augmentation using a level set-based nonlinear morphing technique. The proposed algorithm was evaluated in the context of a spatiotemporal statistical shape modeling of a liver using 42 manually segmented livers from children whose age ranged from approximately 2 weeks to 95 months. Our method achieved a higher generalization and specificity ability compared with conventional methods.

Automatic kidney segmentation in 3D pediatric ultrasound images using deep neural networks and weighted fuzzy active shape model

Automatic kidney segmentation in 3D ultrasound (3DUS) images is clinically important to provide a fast and reliable diagnosis of diseased kidneys. US imaging is a challenging modality for organ evaluation, especially for pediatric kidneys with different shape, size, and texture characteristics. The aim of this study is to present an automatic kidney segmentation method in pediatric 3DUS images using the combination of deep neural networks and weighted fuzzy active shape model. We used deep neural networks to localize the kidney bounding box. The box was then used to initialize the weighted fuzzy active shape model and complete the fully automatic segmentation of the kidney capsule in 3DUS. The performance of the method was evaluated using a dataset of 45 kidneys, showing an average Dice similarity score of 0.82 ± 0.06 and average symmetric surface distance of 1.94 ± 0.74 mm.

Automatic detection of kidney in 3D pediatric ultrasound images using deep neural networks.

Ultrasound (US) imaging is the routine and safe diagnostic modality for detecting pediatric urology problems, such as hydronephrosis in the kidney. Hydronephrosis is the swelling of one or both kidneys because of the build-up of urine. Early detection of hydronephrosis can lead to a substantial improvement in kidney health outcomes. Generally, US imaging is a challenging modality for the evaluation of pediatric kidneys with different shape, size, and texture characteristics. The aim of this study is to present an automatic detection method to help kidney analysis in pediatric 3DUS images. The method localizes the kidney based on its minimum volume oriented bounding box) using deep neural networks. Separate deep neural networks are trained to estimate the kidney position, orientation, and scale, making the method computationally efficient by avoiding full parameter training. The performance of the method was evaluated using a dataset of 45 kidneys (18 normal and 27 diseased kidneys diagnosed with hydronephrosis) through the leave-one-out cross validation method. Quantitative results show the proposed detection method could extract the kidney position, orientation, and scale ratio with root mean square values of 1.3 ± 0.9 mm, 6.34 ± 4.32 degrees, and 1.73 ± 0.04, respectively. This method could be helpful in automating kidney segmentation for routine clinical evaluation.