Kalavathi Palanisamy | GANDHIGRAM RURAL INSTITUTE (original) (raw)

Papers by Kalavathi Palanisamy

International Journal of Imaging Systems and Technology, 2016

A study that was published by pubmed in September 2010 concluded that Obesity is more common in c... more A study that was published by pubmed in September 2010 concluded that Obesity is more common in children with CD than previously recognized. In the appropriate clinical setting, CD must be considered even in obese children. Untreated celiac disease symptoms can vary greatly from person to person, it can cause many problems to the patient, migraine, abortion, unexplained anemia, brain and nervous disorders. Referring also to a study published by Ncpi-Nutrients 2014, 3 out of 17 patients were obese and presented with BMI≥30 (Figure 2). Gluten Enteropathy is a common cause of weight issues in populations that consume grains as a diet staple. As a conclusion, consider celiac (coeliac) disease when you study obesity.

Proceedings of the Eighth Indian Conference on Computer Vision, Graphics and Image Processing - ICVGIP '12, 2012

Automatic segmentation of brain tissues from magnetic resonance images still remains as a challen... more Automatic segmentation of brain tissues from magnetic resonance images still remains as a challenge due to variations in shape and size, use of different pulse sequences, overlapping signal intensities and imaging artifacts. In this paper, an automatic method to segment the brain regions from T1weighted MRI human head scans is proposed. This method consists of two stages. In Stage-1, the brain region in the middle slice of the volume is extracted. The brain regions in the remaining slices are extracted in Stage-2. In each stage the binary form of the brain image is processed to find the rough brain mask. The boundary of the fine brain region in the rough brain is detected using the contour lines. The proposed method is robust to the variability of brain anatomy and image orientation. The experimental results using 60 volumes of T1-weighted brain images show that the proposed method extracts the brain regions more accurately than by the popular methods BET, BSE, WAT, HWA and GCUT.

Communications in Computer and Information Science, 2011

This paper describes a new approach for medical image binarization based on square wave represent... more This paper describes a new approach for medical image binarization based on square wave representation. A square wave is a type of wave form, where the input signal has two levels +1 (foreground) and-1 (background). The signal switches between these levels based on the threshold value computed at that level with the specified time interval. In this method, a local threshold value is calculated at every interval using the current intensity value. Then, the image pixel is assigned with a value +1 or-1 using this local threshold value. The experimental results show that the proposed method reduces the complexity and increases the seperability factor in medical image segmentation. The result obtained by our method is comparable to or better than Otsu's thresholding method.

2013 8th International Conference on Computer Science & Education, 2013

In this paper a new method is devised to segment the brain tissue from Tl-weighted MR brain image... more In this paper a new method is devised to segment the brain tissue from Tl-weighted MR brain images. The proposed method selects optimal threshold values based on Otsu's multiple thresholding technique to segment WM, GM and CSF from MR brain images. The segmentation results obtained by the proposed method is compared with the manually segmented images and have produced best results in terms of overlapping measure. The experimental results using 20 volumes of brain images show that the proposed method accurately segmented the brain tissues than the existing methods AMAP, BMAP, FCM, MAP, ML and TKmean.

All medical image processing techniques need to extract meaningful information from medical image... more All medical image processing techniques need to extract meaningful information from medical images. However, the noise generated during image acquisition and transmission degrades the human interpretation, or computer-aided analysis of these images. Therefore, denoising should be performed to improve the image quality for more accurate analysis and diagnosis. In this paper we propose a medical image denoising technique using three spatial mean filters and the performance of these filters are evaluated using the Canny edge detector by computing the edge image difference between the original and the denoised image.

Journal of Computer Assisted Tomography, 2013

The high-resolution magnetic resonance brain images often contain some nonbrain tissues (ie, skin... more The high-resolution magnetic resonance brain images often contain some nonbrain tissues (ie, skin, fat, muscle, neck, eye balls, etc) compared with the functional images such as positron emission tomography, single-photon emission computed tomography, and functional magnetic resonance imaging (MRI) scans, which usually contain few nonbrain tissues. Automatic segmentation of brain tissues from MRI scans remains a challenging task due to the variation in shape and size, use of different pulse sequences, overlapping signal intensities and imaging artifacts. This article presents a contour-based automatic brain segmentation method to segment the brain regions from T1-, T2-, and proton densityYweighted MRI of human head scans. The proposed method consists of 2 stages. In stage 1, the brain regions in the middle slice is extracted. Many of the existing methods failed to extract brain regions in the lower and upper slices of the brain volume, where the brain appears in more than 1 connected region. To overcome this problem, in the proposed method, a landmark circle is drawn at the center of the extracted brain region of a middle slice and is likely to pass through all the brain regions in the remaining lower and upper slices irrespective of whether the brain is composed of 1 or more connected components. In stage 2, the brain regions in the remaining slices are extracted with reference to the landmark circle obtained in stage 1. The proposed method is robust to the variability of brain anatomy, image orientation, and image type, and it extracts the brain regions accurately in T1-, T2-, and proton densityYweighted normal and abnormal brain images. Experimental results by applying the proposed method on 100 volumes of brain images show that the proposed method exhibits best and consistent performance than by the popular existing methods brain extraction tool, brain surface extraction, watershed algorithm, hybrid watershed algorithm, and skull stripping using graph cuts.

The Imaging Science Journal, 2013

This paper presents a skull stripping method to segment the brain from MRI human head scans using... more This paper presents a skull stripping method to segment the brain from MRI human head scans using multi-seeded region growing technique. The proposed method has two stages. In Stage-1, the brain in the middle slice is segmented, the brains in the remaining slices are segmented in Stage-2. In each stage, the proposed method is required to identify the rough brain mask. The fine brain region in the rough brain mask is segmented using multi-seeded region growing approach. The proposed method uses multiple seed points which are selected automatically based on the intensity profile of grey matter (GM), white matter (WM) and cerebrospinal fluid (CSF) of the brain image. The proposed brain segmentation method using multi-seeded region growing (BSMRG) was validated using 100 volumes of T1, T2 and PD-weighted MR brain images obtained from Internet Brain Segmentation Repository (IBSR), LONI and Whole Brain Atlas (WBA). The best Dice (D) value of 0?971 and Jaccard (J) value of 0?944 were recorded by the proposed BSMRG method on IBSR dataset. For LONI dataset, the best values of D50?979 and J50?960 were obtained for the sagittal oriented images by the proposed method. The performance consistency of the proposed method was tested on the brain images of all types and orientation and have and produced better and stable results than the existing methods Brain Extraction Tool (BET), Brain Surface Extraction (BSE), Watershed Algorithm (WAT), Hybrid Watershed Algorithm (HWA) and Skull Stripping using Graph Cuts (GCUT).

MRI brain images are widely used in medical applications for research, diagnosis, treatment, surg... more MRI brain images are widely used in medical applications for research, diagnosis, treatment, surgical planning and image guided surgeries. These MR brain images are often corrupted with various imaging artifacts and may affect the performance of brain image processing techniques. In this paper, we listed and identified the causes of the common imaging artifacts in MR brain images.

In this study an attempt was made to characterize the lignolytic fungi against various growth con... more In this study an attempt was made to characterize the lignolytic fungi against various growth conditions. The decayed sawdust sample was collected and screened for lignolytic fungi using Malt extract medium containing of tannic acid as lignin equivalent. Seven fungal isolates such as and Trichoderma spp, Penicillium spp, Aspergillus niger, Botrytis spp, Rhizopus spp, Aspergillus fumigatus and Cladosporium spp2 were selected based on their solubilization index against tannic acid. All these isolates were further characterized for their activity to degrade lignin analogue i.e., tannic acid with various growth conditions such as concentration of tannic acid (0.25, 0.50, 0.75, 1.0 and 1.25%) different temperature (28°C, 37°C and 45°C) and various pH ( acidic, neutral and alkaline) conditions. The bacterial isolates such as Penicillium spp, Aspergillus niger, Aspergillus fumigatus and Trichoderma spp were showed the better lignolytic activity at a concentration of 1.0 percent tannic acid...

Int. Arab J. Inf. Technol., 2016

In this article, a new segmentation method to extract the brain from T1, T2 and PD-weighted Magne... more In this article, a new segmentation method to extract the brain from T1, T2 and PD-weighted Magnetic Resonance Image (MRI) of human head images based on Modified Chan-Vese (MCV) active contour model is proposed. This method first segment the brain in the middle slice of the brain volume. Then, the brain regions of the remaining slices are segmented using the extracted middle brain as a reference. The input brain image is pre-processed to find the rough brain. The initial contour for the MCV method is drawn at the center of the segmented rough brain image and is then propagated to reach the brain boundary. The result of this proposed method is compared with the hand stripped images and found to produce significant results. The proposed method was tested with 100 volumes of brain images and had accurately segmented the brain regions which are better than the existing methods such as Brain Extraction Tool (BET), Brain Surface Extraction (BSE), Watershed Algorithm (WAT), Hybrid Watershe...

Current medical imaging, 2020

BACKGROUND AND OBJECTIVE In order to reduce time complexity and to improve the computational effi... more BACKGROUND AND OBJECTIVE In order to reduce time complexity and to improve the computational efficiency in diagnosing process, automated brain tissue segmentation for magnetic resonance brain images is proposed in this paper. METHODS This method incorporates two processes, the first one is preprocessing and the second one is segmentation of brain tissue using Histogram based Swarm Optimization techniques. The proposed method was investigated with images obtained from twenty volumes and eighteen volumes of T1-Weighted images obtained from Internet Brain Segmentation Repository (IBSR), Alzheimer disease images from Minimum Interval Resonance Imaging in Alzheimer's Disease (MIRIAD) and T2-Weighted real-time images collected from SBC Scan Center Dindigul. RESULTS The proposed technique was tested with three brain image datasets. Quantitative evaluation was done with Jaccard (JC) and Dice (DC) and also it was compared with existing swarm optimization techniques and other methods like...

Communications in Computer and Information Science

Low power and lossy networks (LLNs) which are interconnected with internet to collect data throug... more Low power and lossy networks (LLNs) which are interconnected with internet to collect data through sensors and store them over the cloud make the Internet of Things (IoT). The routing protocols in LLNs play the essential role of forwarding and routing the packets. IPv6 routing protocol for Low power and lossy networks (RPL), used in LLNs has the key features of topology formation, control messages, objective function and Trickle algorithm. The trickle algorithm is a dynamic algorithm controlling the timer in RPL. There are some key parameters in the trickle algorithm that affect the functioning of the trickle timer and consequently the RPL itself. The efficiency, robustness and improvement of RPL depends to a great extent on the fine tuning of the trickle algorithm and there are no specific standard values provided for the attunement. This paper aims at creating a suitable simulation environment in Cooja Simulator over the Contiki operating system and attuning the key parameters of trickle algorithm, namely minimum interval (Imin), maximum interval (Imax) and redundancy value (k) to find out the optimum reliability of RPL.

Journal of Applied Clinical Medical Physics

Abstract In the last two decades, incredible progress in various medical imaging modalities and s... more Abstract In the last two decades, incredible progress in various medical imaging modalities and sensing techniques have been made, leading to the proliferation of three‐dimensional (3D) imagery. Byproduct of such great progress is the production of huge volume of medical images and this big data place a burden on automatic image processing methods for diagnostic assistance processes. Moreover, large amount of medical imaging data needs to be transmitted with no loss of information for the purpose of telemedicine, remote diagnosis etc. In this work, we consider a hybrid lossless compression technique with object‐based features for three‐dimensional (3D) medical images. Our approach utilizes two phases as follows: first we determine the volume of interest (VOI) for a given 3D medical imagery using selective bounding volume (SBV) method, and second the obtained VOI is encoded using a hybrid lossless algorithm using Lembel‐Ziv‐Welch Coding (LZW) followed by arithmetic coding (L to A). Experimental results show that our proposed 3D medical image compression method is comparable with other existing standard lossless encoding methods such as Huffman Coding, Run Length Coding, LZW, and Arithmetic Coding and obtains superior results overall.

Journal of Digital Imaging, 2015

The high resolution magnetic resonance (MR) brain images contain some non-brain tissues such as s... more The high resolution magnetic resonance (MR) brain images contain some non-brain tissues such as skin, fat, muscle, neck, and eye balls compared to the functional images namely positron emission tomography (PET), single photon emission computed tomography (SPECT), and functional magnetic resonance imaging (fMRI) which usually contain relatively less non-brain tissues. The presence of these non-brain tissues is considered as a major obstacle for automatic brain image segmentation and analysis techniques. Therefore, quantitative morphometric studies of MR brain images often require a preliminary processing to isolate the brain from extra-cranial or non-brain tissues, commonly referred to as skull stripping. This paper describes the available methods on skull stripping and an exploratory review of recent literature on the existing skull stripping methods.

Automatic segmentation of brain tissue on magnetic resonance images is a challenging process due ... more Automatic segmentation of brain tissue on magnetic resonance images is a challenging process due to the variation in brain shapes and similarity of intensity values in the brain and non-brain tissues. Skull stripping is a process of segmenting brain and non-brain tissues in MR brain images. It is an important image processing step in many neuroimage studies. In this paper, we propose a new skull stripping method for magnetic resonance image (MRI) of human head scans based on image contour. We used hybrid method, which combines two or more methods to produce better result. This algorithm first preprocesses the image by denoising using mean filter. The denoised image is blurred to obtain a rough brain mask using image contour. The rough mask is further processed to produce final brain mask. The proposed algorithm is compared with standard manual stripping (Gold standard) images and produced significant result. The experimental results show that the proposed method extracted the brain accurately which are comparable to that of BSE, BET, WAT and HWA using IBSR data set.

Procedia Computer Science

International Journal of Computer Sciences and Engineering, 2017

Most of the remote sensing images are multispectral image where these images are acquired in the ... more Most of the remote sensing images are multispectral image where these images are acquired in the form of several bands that constitute a spectral direction. As large amount of data is represented by multispectral image, a lot of memory space is needed for storage and transmission. Hence, there is big need for compression methods for multispectral images. The prime factor of any image compression method is the redundancy as well as correlation on an image. In this way, the multispectral images having high degree of correlation on spatial domain and redundancy on spectral domain. This leads to conception of several compression methods for these multispectral images. Moreover, every tiny information from multispectral image is very important for efficient processing and so the lossless encoding is always preferable. In this paper, we proposed a hybrid lossless method using Lempel-Ziv-Welch (LZW) and Arithmetic Coding for compressing the multispectral Images. The performance of our method is compared with existing lossless compression methods such as Huffman Coding, Run Length Coding (RLE), LZW and Arithmetic Coding.

Segmentation of bacterial image is an important process in many bioimage applications. Automatic ... more Segmentation of bacterial image is an important process in many bioimage applications. Automatic counting of bacteria is also an essential step in microbiology to control environmental condition and enhance growth to obtain desired results. The manual counting is difficult and error prone. Therefore, it is necessary to develop a computational algorithm to count the bacteria present in the bacterial images. In this paper, we proposed an approach for automatic segmentation and counting of bacteria in Light Microscopy (LM) and Scanning Electron Microscopy (SEM) images. We used CV active contour model and labeling algorithm for segmentation and counting of bacteria. The count value obtained by the proposed method is compared with the manual count value.

International Journal of Imaging Systems and Technology, 2016

Proceedings of the Eighth Indian Conference on Computer Vision, Graphics and Image Processing - ICVGIP '12, 2012

Communications in Computer and Information Science, 2011

2013 8th International Conference on Computer Science & Education, 2013

Journal of Computer Assisted Tomography, 2013

The Imaging Science Journal, 2013

Int. Arab J. Inf. Technol., 2016

Current medical imaging, 2020

Communications in Computer and Information Science

Journal of Applied Clinical Medical Physics

Journal of Digital Imaging, 2015

Procedia Computer Science

International Journal of Computer Sciences and Engineering, 2017

Alzheimer Disease (AD) is one of the neurodegenerative disease occur due to death of the brain ce... more Alzheimer Disease (AD) is one of the neurodegenerative disease occur due to death of the brain cells, and leads to shrinkage in the brain tissues. MRI imaging technique is the tool used to diagnosis and detect the Alzheimer disease. In this paper, we proposed a method consists of two process, in the first process, the skull is removed from the brain image using Contour based brain segmentation method (CBSM), and then we applied clustering technique namely Fast Fuzzy C Means (FFCM) to segment the brain tissue such as White Matter (WM) and Gray Matter (GM). In the second step, the segmented WM and GM are analyzed to detect Alzheimer's Disease in MR brain images by computing the similarity measures such as Jaccard and Dice against the normal brain.

Biometric system provides automatic identification of an individual based on a unique features or... more Biometric system provides automatic identification of an individual based on a unique features or characteristic possessed by the individual. Iris recognition is one of the important biometric recognition systems that identify people based on their eyes and iris. In this paper, we proposed a method to segment the iris in the eye image using active contour method. We evaluated our proposed method using the image obtained from the internet.

Brain tissue segmentation plays a major role in medical image analysis for identifying various br... more Brain tissue segmentation plays a major role in medical image analysis for identifying various brain disorders. In this paper, we presented a method Wavelet Based Image Fusion with Clustering Technique (WBIFCT) for segmenting of brain tissues such as White Matter (WM) and Gray Matter (GM) used to identify the brain related diseases. The proposed method consists of two process, initially, the image is decomposed to give approximation and detail coefficients. From these coefficients, features vectors are extracted using Principal Component Analysis (PCA) and these features are combined with image fusion techniques. Finally, K-Means clustering technique is applied to the fused image to segment the brain tissues. The similarity measures such as Jaccard (J) and Dice (D) are computed for comparing the performance of manually segmented images. The results of proposed method gives better tissue classification on selected twenty volumes of T1 weighted MR brain images.

Automatic segmentation and detection of virus of different shapes in Electron microscopic images ... more Automatic segmentation and detection of virus of different shapes in Electron microscopic images are necessary for the microbiologist for its analysis and to control the environmental condition. Enveloped virus is one such type of viruses which are appeared as a circular shape in electron microscopic images. In this paper, we present a simple method to segment and detect enveloped virus using its morphological features such as shape and size. We used images obtained from Transmission Electron Microscopy (TEM) and Scanning Electron Microscopy (SEM) for automatic segmentation and detection. Our experimental results are better compared with manual detection.

Noise is one of the important factor which affects medical images during image acquisition and tr... more Noise is one of the important factor which affects medical images during image acquisition and transmission process. There are different types of techniques are used for producing these medical images such as X-ray, Ultrasound, CT and MRI. During these processes noise occurs due to some radiation problem. In this paper, we have used some selected spatial filtering techniques such as Kuwahara filter, Lee filter, Average filter, Median filter, Entropy filter, Order filter, Max filter and Min filter to denoise the ultrasound image. For comparative analysis, the performance of these filtering techniques was quantitatively evaluated using Peak Signal to Noise Ratio (PSNR). The observed results shown that Max filter gives high quality image compared to other filtering techniques.

The Image compression becomes an integral part of digital imaging field because of the tremendous... more The Image compression becomes an integral part of digital imaging field because of the tremendous advancements in digital image acquiring techniques. As the quality of the images is high, the complexity for storing and transmitting those images also increased. Even the image compression methodologies are also having an optimize growth such as transformation based compression methods which includes Fast Fourier transform (FFT), Discrete cosine transform (DCT), Discrete wavelet transform (DWT). In this paper, we proposed a wavelet based image compression technique on which the two-dimensional discrete wavelet transform is used to decompose the image and the wavelet coefficients are transmitted by an entropy encoding method after thresholding. The results of our proposed method are assessed in terms of compression ratio (CR), bits per pixel (BPP), peak signal to noise ratio (PSNR) and the Structural similarity index (SSIM).