Speckle reduction using an artificial neural network algorithm.

This paper presents an algorithm for reducing speckle noise from optical coherence tomography (OCT) images using an artificial neural network (ANN) algorithm. The noise is modeled using Rayleigh distribution with a noise parameter, sigma, estimated by the ANN. The input to the ANN is a set of intensity and wavelet features computed from the image to be processed, and the output is an estimated sigma value. This is then used along with a numerical method to solve the inverse Rayleigh function to reduce the noise in the image. The algorithm is tested successfully on OCT images of Drosophila larvae. It is demonstrated that the signal-to-noise ratio and the contrast-to-noise ratio of the processed images are increased by the application of the ANN algorithm in comparison with the respective values of the original images.

Investigation of basal cell carcinoma using dynamic focus optical coherence tomography.

Optical coherence tomography (OCT) is becoming a popular modality for skin tumor diagnosis and assessment of tumor size and margin status. We conducted a number of imaging experiments on periocular basal cell carcinoma (BCC) specimens using an OCT configuration. This configuration employs a dynamic focus (DF) procedure where the coherence gate moves synchronously with the peak of the confocal gate, which ensures better signal strength and preservation of transversal resolution from all depths. A DF-OCT configuration is used to illustrate morphological differences between the BCC and its surrounding healthy skin in OCT images. The OCT images are correlated with the corresponding histology images. To the best of our knowledge, this is the first paper to look at DF-OCT imaging in examining periocular BCC.

OCT skin image enhancement through attenuation compensation.

The enhancement of optical coherence tomography (OCT) skin images can help dermatologists investigate the morphologic information of the images more effectively. In this paper, we propose an enhancement algorithm with the stages that includes speckle reduction, skin layer detection, and attenuation compensation. A weighted median filter is designed to reduce the level of speckle while preserving the contrast. A novel skin layer detection technique is then applied to outline the main skin layers: stratum corneum, epidermis, and dermis. The skin layer detection algorithm does not make any assumption about the structure of the skin. A model of the light attenuation is then used to estimate the attenuation coefficient of the stratum corneum, epidermis, and dermis layers. The performance of the algorithm has been evaluated qualitatively based on visual evaluation and quantitatively using two no-reference quality metrics: signal-to-noise ratio and contrast-to-noise ratio. The enhancement algorithm is tested on 35 different skin OCT images, which show significant improvements in the quality of the images, especially in the structures at deeper levels.

Two-tensor model-based bootstrapping on classified tensor morphologies: estimation of uncertainty in fiber orientation and probabilistic tractography.

In this manuscript, fast and clinically feasible model-based bootstrapping algorithms using a geometrically constrained two-tensor diffusion model are employed for estimating uncertainty in fiber orientation. A Monte-Carlo-based tensor morphology voxel classification algorithm is initially applied using single-tensor bootstrap samples before the use of a two-tensor model-based bootstrapping algorithm. Classification of tensor morphologies allows the tensor morphology to be considered when selecting the most appropriate bootstrap procedure. A constrained two-tensor model approach can greatly reduce data acquisition and computational times for whole bootstrap data volume generation compared to other multifiber model techniques, facilitating widespread clinical use. For comparison, we propose a new repetition-bootstrap algorithm based on classified voxels and the constrained two-tensor model. Tractography with these bootstrapping algorithms is also developed to estimate the connection probabilities between brain regions, especially regions with complex fiber configurations. Experimental results on synthetic data, a hardware phantom and human brain data demonstrate the superior performance of our algorithms compared to conventional approaches.

A novel approach for improved tractography and quantitative analysis of probabilistic fibre tracking curves.

This paper presents a novel approach for improved diffusion tensor fibre tractography, aiming to tackle a number of the limitations of current fibre tracking algorithms, and describes a quantitative analysis tool for probabilistic tracking algorithms. We consider the sampled random paths generated by a probabilistic tractography algorithm from a seed point as a set of curves, and develop a statistical framework for analysing the curve-set geometrically that finds the average curve and dispersion measures of the curve-set statistically. This study is motivated firstly by the goal of developing a robust fibre tracking algorithm, combining the power of both deterministic and probabilistic tracking methods using average curves. These typical curves produce strong connections to every anatomically distinct fibre tract from a seed point and also convey important information about the underlying probability distribution. These single well-defined trajectories overcome a number of the limitations of deterministic and probabilistic approaches. A new clustering algorithm for branching curves is employed to separate fibres into branches before applying the averaging methods. Secondly, a quantitative analysis tool for probabilistic tracking methods is introduced using statistical measures of curve-sets. Results on phantom and in vivo data confirm the efficiency and effectiveness of the proposed approach for the tracking algorithm and the quantitative analysis of the probabilistic methods.

Probabilistic clustering and shape modelling of white matter fibre bundles using regression mixtures.

We present a novel approach for probabilistic clustering of white matter fibre pathways using curve-based regression mixture modelling techniques in 3D curve space. The clustering algorithm is based on a principled method for probabilistic modelling of a set of fibre trajectories as individual sequences of points generated from a finite mixture model consisting of multivariate polynomial regression model components. Unsupervised learning is carried out using maximum likelihood principles. Specifically, conditional mixture is used together with an EM algorithm to estimate cluster membership. The result of clustering is a probabilistic assignment of fibre trajectories to each cluster and an estimate of cluster parameters. A statistical shape model is calculated for each clustered fibre bundle using fitted parameters of the probabilistic clustering. We illustrate the potential of our clustering approach on synthetic and real data.

A novel white matter fibre tracking algorithm using probabilistic tractography and average curves.

This paper presents a novel white matter fibre tractography approach using average curves of probabilistic fibre tracking measures. We compute "representative" curves from the original probabilistic curve-set using two different averaging methods. These typical curves overcome a number of the limitations of deterministic and probabilistic approaches. They produce strong connections to every anatomically distinct fibre tract from a seed point and also convey important information about the underlying probability distribution. A new clustering algorithm is employed to separate fibres into branches before applying averaging methods. The performance of the technique is verified on a wide range of seed points using a phantom dataset and an in vivo dataset.