Syed Muhammad Anwar

Medical image retrieval using deep convolutional neural network

Abstract With a widespread use of digital imaging data in hospitals, the size of medical image repositories is increasing rapidly. This causes difficulty in managing and querying these large databases leading to the need of content based medical image retrieval (CBMIR) systems. A major challenge in CBMIR systems is the semantic gap that exists between the low level visual information captured by imaging devices and high level semantic information perceived by human. The efficacy of such systems is more crucial in terms of feature representations that can characterize the high-level information completely. In this paper, we propose a framework of deep learning for CBMIR system by using deep convolutional neural network (CNN) that is trained for classification of medical images. An intermodal dataset that contains twenty-four classes and five modalities is used to train the network. The learned features and the classification results are used to retrieve medical images. For retrieval, best results are achieved when class based predictions are used. An average classification accuracy of 99.77% and a mean average precision of 0.69 is achieved for retrieval task. The proposed method is best suited to retrieve multimodal medical images for different body organs.

A Novel Image Retrieval Based on a Combination of Local and Global Histograms of Visual Words

Content-based image retrieval (CBIR) provides a sustainable solution to retrieve similar images from an image archive. In the last few years, the Bag-of-Visual-Words (BoVW) model gained attention and significantly improved the performance of image retrieval. In the standard BoVW model, an image is represented as an orderless global histogram of visual words by ignoring the spatial layout. The spatial layout of an image carries significant information that can enhance the performance of CBIR. In this paper, we are presenting a novel image representation that is based on a combination of local and global histograms of visual words. The global histogram of visual words is constructed over the whole image, while the local histogram of visual words is constructed over the local rectangular region of the image. The local histogram contains the spatial information about the salient objects. Extensive experiments and comparisons conducted on Corel-A, Caltech-256, and Ground Truth image datasets demonstrate that the proposed image representation increases the performance of image retrieval.

Human emotion recognition and analysis in response to audio music using brain signals

Human emotion recognition using brain signals is an active research topic in the field of affective computing. Music is considered as a powerful tool for arousing emotions in human beings. This study recognized happy, sad, love and anger emotions in response to audio music tracks from electronic, rap, metal, rock and hiphop genres. Participants were asked to listen to audio music tracks of 1min for each genre in a noise free environment. The main objectives of this study were to determine the effect of different genres of music on human emotions and indicating age group that is more responsive to music. Thirty men and women of three different age groups (1525 years, 2635 years and 3650 years) underwent through the experiment that also included self reported emotional state after listening to each type of music. Features from three different domains i.e., time, frequency and wavelet were extracted from recorded EEG signals, which were further used by the classifier to recognize human emotions. It has been evident from results that MLP gives best accuracy to recognize human emotion in response to audio music tracks using hybrid features of brain signals. It is also observed that rock and rap genres generated happy and sad emotions respectively in subjects under study. The brain signals of age group (2635 years) gave best emotion recognition accuracy in accordance to the self reported emotions. Display Omitted Brain signals are recorded using EEG headset while listening to audio music.Four different emotions are recognized using hybrid features of EEG signals.Emotion recognition accuracy is analyzed for three different age groups.The relationship between music genres and human emotions is examined.

Facial Expression Recognition Using Stationary Wavelet Transform Features

Humans use facial expressions to convey personal feelings. Facial expressions need to be automatically recognized to design control and interactive applications. Feature extraction in an accurate manner is one of the key steps in automatic facial expression recognition system. Current frequency domain facial expression recognition systems have not fully utilized the facial elements and muscle movements for recognition. In this paper, stationary wavelet transform is used to extract features for facial expression recognition due to its good localization characteristics, in both spectral and spatial domains. More specifically a combination of horizontal and vertical subbands of stationary wavelet transform is used as these subbands contain muscle movement information for majority of the facial expressions. Feature dimensionality is further reduced by applying discrete cosine transform on these subbands. The selected features are then passed into feed forward neural network that is trained through back propagation algorithm. An average recognition rate of 98.83% and 96.61% is achieved for JAFFE and CK

Segmentation of glioma tumors in brain using deep convolutional neural network

Abstract Detection of brain tumor using a segmentation based approach is critical in cases, where survival of a subject depends on an accurate and timely clinical diagnosis. Gliomas are the most commonly found tumors, which have irregular shape and ambiguous boundaries, making them one of the hardest tumors to detect. The automation of brain tumor segmentation remains a challenging problem mainly due to significant variations in its structure. An automated brain tumor segmentation algorithm using deep convolutional neural network (DCNN) is presented in this paper. A patch based approach along with an inception module is used for training the deep network by extracting two co-centric patches of different sizes from the input images. Recent developments in deep neural networks such as dropout, batch normalization, non-linear activation and inception module are used to build a new ILinear nexus architecture. The module overcomes the over-fitting problem arising due to scarcity of data using dropout regularizer. Images are normalized and bias field corrected in the pre-processing step and then extracted patches are passed through a DCNN, which assigns an output label to the central pixel of each patch. Morphological operators are used for post-processing to remove small false positives around the edges. A two-phase weighted training method is introduced and evaluated using BRATS 2013 and BRATS 2015 datasets, where it improves the performance parameters of state-of-the-art techniques under similar settings.

Psychological stress measurement using low cost single channel EEG headset

This paper present, results of the study on noninvasive stress measurement using EEG signals recorded with a single electrode device. The process involves EEG data acquisition, feature extraction, and stress level classification. Psychologists have developed over a period of time, questionnaires that cover a wide range of symptoms associated with stress. In the first step, stress level of each participant was assessed using the Perceived Stress Scale (PSS) questionnaire. EEG signals of twenty eight participants were recorded using a single channel EEG headset for duration of three minutes. Feature vector based on frequency sub bands is used to train three different machine learning algorithms, to classify the stress level of participants. It is evident from results that psychological stress level can be measured by single channel EEG headset using machine learning algorithms with considerable accuracy. Moreover, increased Beta activity of subjects with high stress has been observed as compared to the subjects with no stress. This fact can be used as a key factor in classifying psychological stress with single channel EEG headset.

Autonomous Glaucoma detection from fundus image using cup to disc ratio and hybrid features

Glaucoma is a non-curable optic disease which can cause irreversible blindness if not detected at early stage. Progression of glaucoma occurs due to an increase in intraocular pressure and results in the damage of optic nerve. Progression of glaucoma can be stopped if detected at an early stage. There are no early symptoms of glaucoma and the only source to detect glaucoma at an early stage is the structural change that arises in the internal eye. Fundoscopy is one of the modern medical imaging techniques that enable Ophthalmologists to observe structural changes in the Optic Disc to detect glaucoma. Many autonomous glaucoma detection systems analyze fundus image by calculating Cup to Disc Ratio (CDR) and categorize the image as glaucoma or healthy. Glaucoma detection using machine learning is also being used widely to aid ophthalmologists. The proposed methodology provides a novel algorithm to detect glaucoma using a fusion of CDR and hybrid textural and intensity features. Image categorization (glaucoma, non-glaucoma, suspect) is done based on the results from both CDR and classifier. This fusion of CDR with hybrid features has improved the sensitivity of system to 1, specificity 0.88 and accuracy 92%.

Real time text speller based on eye movement classification using wearable EEG sensors

Making the machine more empathie to the user is one of the aspects of affective computing. Signal procured by Brain Computer Interface frameworks are utilized to process brain signals for controlling different gadgets for human ease. P300 based content speller is one of the device utilized for eye-writing, however eye movement based content speller can create fine and exact comes about as contrasted with P300 based content speller. This paper presents a robust solution for utilizing eye muscular-movement for typing text content. Eye movement is a brawny and husky movement, yet by utilizing EEG signals for eye movements, we have obtained EEG signal for diverse users and after that utilizing those signals we have finally created a framework for content spelling as a cursor movement over the letter sets. All analysis were executed on runtime procured datasets of numerous users to arrange the indicator in three classes lastly utilizing them as an aide sign to spell the content. The accuracy of the proposed system for right movement, left movement and blink is 70%, 80% and 79% respectively.

A review analysis on early glaucoma detection using structural features

Glaucoma is an eye disease that might cause severe destruction and permanent blindness if not detected at an early stage. Glaucoma is also called silent thief of sight. Glaucoma can be detected using structural and functional features. Functional features are observed by visual field testing and to observe structural features Optical Coherence Tomography (OCT) and Fundus images are the most widely used medical imaging techniques. Optic nerve head (ONH), Retinal layers are the key source and most repeatable structural features to detect structural changes in the retina of glaucomatous eyes. This paper presents a review on different glaucoma detection techniques from clinical and machine learning perspectives. The paper also highlights the functional and structural features and their significance with respect to digital fundus and OCT images for glaucoma detection. It concludes that structural features are more precise for early glaucoma detection as compared to functional features. Moreover, using hybrid features in training classifiers and correlating results of both fundus and OCT images can yield more accurate results.

Optic disc localization using local vessel based features and support vector machine

Optic disc is one of the fundamental regions located in the internal retina that helps ophthalmologists in analysis and early diagnosis of many retinal diseases such as optic atrophy, optic neuritis, papilledema, ischemic optic neuropathy, glaucoma and diabetic retinopathy. An accurate and early diagnosis requires an accurate optic disc examination. Presence of different retinal abnormalities and non-uniform illumination make optic disc localization a challenging task. There is a need to detect and localize optic disc from fundus images with high accuracy to make the diagnosis using Computer Aided Systems developed for ophthalmic disease diagnosis more reliable. Proposed algorithm provides a novel optic disc localization and segmentation technique that detects multiple candidate optic disc regions from fundus image using enhancement and segmentation. The proposed system then extracts a hybrid feature set for each candidate region consisting of vessel based and intensity based features which are finally fed to SVM classifier. Final decision of Optic disc region is done after computing Manhattan distance from the mean of training data feature matrix. The evaluation of proposed system has been done on publicly available datasets and one local dataset and results shows the validity of proposed system.