Abstract

Population screening for sight threatening diseases based on fundus imaging is in place or being considered worldwide. Most existing programs are focussed on a specific disease and are based on manual reading of images, though automated image analysis based solutions are being developed. Exudates and drusen are bright lesions which indicate very different diseases, but can appear to be similar. Discriminating between them is of interest to increase screening performance. In this paper, we present a Bag of Words approach which can be used to design a system that can play the dual role of content based retrieval (of images with exudates or drusen) system and a decision support system to address the problem of bright lesion discrimination. The approach consists of a novel partitioning of an image into patches from which color, texture, edge and granulometry based features are extracted to build a dictionary. A bag of Words approach is then employed to help retrieve images matching a query image as well as derive a decision on the type of bright lesion in the given (query) image. This approach has been implemented and tested on a combination of public and local dataset of 415 images. The area under the curve for image classification is 0.90 and retrieved precision is 0.76.

Abstract

Diffusion Weighted Magnetic Resonance Imaging (DWI) is routinely used for early detection of cerebral ischemic changes in acute stroke. Fast acquisition with a standard echoplanar imaging technique generally compromises the image signal-to-noise ratio and in-plane resolution resulting in a reduction of the conspicuity and definition of lesions in the acquired data when viewed on a standard 8-bit display. We present a novel method for automatically and adaptively determining the window settings that enhance the contrast of the image relative to the ischemic lesions. The method performs a coarse segmentation of the lesions followed by contrast-to-noise ratio based computation of the optimal window parameters. The proposed method was tested on 24 datasets acquired with different protocols. The contrast improvement of the lesions is validated through a mirror region of interest analysis and by using the contrast improvement ratio metric. The average obtained improvement in contrast ranges from 25% to 60%. Preliminary results of segmentation showed a good reduction in the false positives and improvement in the lesion boundaries. A perception study of the windowed results against 8 radiologists was conducted. Reduction of 14.17% in the mean response time of detection was observed. Statistical analysis performed using t-test validates the reduction in mean response time to be significant. Results presented in the study show promise in the method.

Abstract

This paper describes a novel tortuosity measure, based on the premise that tortuosity is a measure of deviation from an ideal non-tortuous vessel. Hence, we propose to model the overall shape of an ideal vessel as a quadratic polynomial at a larger scale while the deviations are modeled as quadratic polynomials at smaller scales. Thus, a given vessel center-line is decomposed as a sum of quadratic polynomials of decreasing scale. This Quadratic Polynomial Decomposition is used as a framework for defining a quantitative measure of tortuosity. As opposed to the existing proposed measures, our method can distinguish between the relative size, shapes and orientations of the vessel bends. The measure is position and scale invariant and satisfies two key desired properties: it varies directly with frequency of twists at fixed amplitude and it varies directly with amplitude of twists when their frequency is fixed. The proposed method has been tested on a standard data set containing 30 artery and 30 vein vessel segments, and shown to be among the best measures as compared to the results of existing methods.

Abstract

The blood perfusion measurement is done by injecting a bolus of contrast agent in the brain followed by imaging over a period of time (scan). This process can extend into minutes and hence any patient motion mid-scan results in corrupted data. This is often observed in dynamic magnetic resonance(MR) imaging for both susceptibility (DSC) and contrast enhanced (DCE) scans. Motion correction done after scanning is typically the most time-intensive step in the entire measurement process since it involves registering each volume in the time-series to a reference volume. We argue that detecting the presence of motion prior to correction can mitigate this problem by reducing the number of volumes to be corrected. The challenge in motion detection is that the injected contrast alters the signal intensity as a function of time leading to false alarms. We present a robust multi-stage method: subdivision of the time series data into bolus and non bolus phases; clustering-based identification of bolus-affected pixels followed by correction of their intensity using a Gamma variate function fitting-based method and a 2Dblock-wise phase correlation for detecting motion between adjacent volumes in DSC-MRI data. The proposed method was tested on a DSC MR sequence with simulated motion of varying degrees. The experimental results show that the entropy of the derived motion fields is a good metric for detecting and categorizing the motion. The proposed scheme when applied prior to correction can achieve on average a37% reduction in the time required for motion correction

Abstract

Glaucoma is an eye disorder that causes irreversible loss of vision and is prevalent in the aging population. Glaucoma is indicated both by structural changes and presence of atrophy in retina. In retinal images, these appear in the form of subtle variation of local intensities. These variations are typically described using local shape based statistics which are prone to error. We propose an automated, global feature based approach to detect glaucoma from images. An image representation is devised to accentuate subtle indicators of the disease such that global image features can discriminate between normal and glaucoma cases effectively.The proposed method is demonstrated on a large image dataset annotated by 3 medical experts. The results show the method to be effective in detecting subtle glaucoma indi-cators. The classification performance on a dataset of 1186color retinal images containing a mixture of normal, suspect and confirmed cases of glaucoma is 97 percent sensitivity at 87 percent specificity. This improves further when the suspect cases are removed from the abnormal cases. Thus,the proposed method offers a good solution for glaucoma screening from retinal images.

Abstract

Several computational visual saliency models have been pro-posed in the context of viewing natural scenes. We aim to investigate the relevance of computational saliency models in medical images in the context of abnormality detection.We report on two studies aimed at understanding the role of visual saliency in medical images. Diffuse lesions in ChestX-Ray images, which are characteristic of Pneumoconiosis and high contrast lesions such as ‘Hard Exudates’ in retinal images were chosen for the study. These approximately correspond to conjunctive and disjunctive targets in a visual search task. Saliency maps were computed using three popular models namely Itti-Koch [7], GBVS [3] and SR [4].The obtained maps were evaluated against gaze maps and ground truth from medical experts. Our results show that GBVS is seen to perform the best(Mdn. ROC area= 0.77) for chest X-Ray images while SRperforms the best (ROC area= 0.73) for retinal images,thus asserting that searching for conjunctive targets calls for a more local examination of an image while disjunctive targets call for a global examination. Based on the results of the above study, we propose extensions for the two best performing models. The first extension makes use of top down knowledge such as lung segmentation. This is shown to improve the performance of GBVS to some extent. Inthe second case the extension is by way of including multi-scale information. This is shown to significantly (by 28.76%)improve abnormality detection. The key insight from these studies is that bottom saliency continues to play a predominant role in examining medical images.

Abstract

This paper addresses the data corruption that occurs due to patient motion during a scan which is particularly a problem in perfusion weighted MRI due to long scan times. Motion correction is typically the rate-limiting step in processing as each volume has to be registered to a reference volume. This is compounded by the dynamically varying contrast in the volume series due to passage of an injected contrast agent. We propose an efficient two stage motion correction method, consisting of motion detection and a 2-pass registration method for aligning the motion-corrupted volumes. A 2D block-wise phase correlation in central slices is used for the first stage. Alignment employs a strategy which is sensitive to the status of the bolus in the volume and is based on gamma-variate function fitting for intensity correction to handle dynamic contrast in DSC-MRI. Evaluation of the approach shows that it is fast and accurate.

Abstract

Diffusion Weighted MR Imaging (DWI) is routinely used for early detection of cerebral ischemic stroke.DWI with higher b-values (b=2000) provide improved sensitivity, higher conspicuity and reduced artifacts and thus improve the detectability of smallest infarcts than conventional DWI (b=1000). We propose a novel framework for accurately detecting stroke regions by combining information from multiple sources:b2000,b1000 data and the apparent diffusion coefficient map.The detected lesions are finally segmented using an active contour approach. The proposed method was tested on 41 datasets acquired with different protocols. A comparison of our method with a leading method [3] validates the effectiveness of our approach. The mediandice coefficient, sensitivity and specificity for stroke segmentation were 0.84, 87.07% and 99.90% respectively.The strength of the proposed method is its ability to capture (and accurately segment) the small (and large) lesions in the data which are often missed by segmentation methods operating on a single b-value data.

Abstract

Detection of abnormalities from medical images is of key interest in developing computer-aided diagnostic tools. In this paper, we observe the key challenges for representation and feature extraction schemes to be met for detection of abnormalities by learning normal cases. We introduce an image representation, motivated by the effect of motion on perception of structures. This representation is based on a set of patterns called generalized moment patterns (GMP) generated via induced motion over regions of interest, for learning normal. The proposed GMP has been utilized to develop a scheme for addressing two well-known problems: lesion classification in mammograms and detection of macular edema in color fundus images. The strengths of this scheme are that it does not require anylesion-level segmentation and relies largely on normal images for training which is attractive for developing screening tools. The proposed scheme has been assessed on two public domain datasets,namely, MIAS and MESSIDOR. A comparison against the performance of state of the art method sindicates the proposed scheme to be superior

Abstract

Colorization is the process of adding colors to gray scale images. This is done to restore or enhance old films or photographs. Most of the techniques for the colorization of an image or video require manual designation of the locations to be colored and the colors themselves which is an expensive and time consuming process. In this paper, we propose a fast but effective fully automated technique for coloring the gray scale image sequences. We define a notion of a most informative frame which is to be coloured manually and exploit the motion field between frames for propagation of the colors to the remaining frames. The proposed technique attempts to provide a method to minimize the amount of labour required for colorization and to decrease the computational cost of this task. The most informative frame is one which has almost all the objects present in that scene. The motion field estimation is based on optical flow. A final refinement step uses similarity based colour filling. Extensive testing of the proposed technique on a large set of videos from movies and animation confirms that it is efficient and effective without any loss of quality.