K Manivannan
@jainuniversity.ac.in
Professor, Department of CSE, School of CSE, JAIN Deemed to be University, JAIN Global Campus,
JAIN Deemed to be University
K Manivannan is an Professor of Computer Science and Engineering in
JAIN Deemed to be University,Bangalore, India. He received
B.E degree in Computer Science and Engineering from Anna University, Chennai and
M.E degree in Computer Science and Engineering from the same University. He has
A Ph.D in Computer Science and Engineering from Anna University, Chennai,
Tamilnadu, India. He has Published 37 National and International Journals, 10
Conferences and 4 books. He has successfully guided 5 Ph.D students in Anna
University, Chennai. His area of interests includes Medical Image Processing, High Performance
Computing, Distributed and Network Architecture.
EDUCATION
B.E CSE
M.E CSE
RESEARCH INTERESTS
Machine Learning, Medical Image Processing
FUTURE PROJECTS
Automatic and Reliable Plant Virus detection Using Deep Learning
Specifically we aim to 1.Create a computer-aided tool to automatically detect and classify the various levels of plant pathogens especially in virus 2.Train AI to identify additional features associated 3.Now a days various biosensors are available for the detection of plant pathogens in- suit analysis. 4.If plant pathogen detection can be done using deep learning techniques by analyzing the images of plant leaves, it will be beneficial for taking prevention methods in an early stage. Real time images of crops from large fields can be taken occasionally and can be analyzed for infection using the images taken will be really beneficial. 5.Thousands of images of plant leaves are kept for the analysis and the biosensor result can also be incorporated for fine tuning of the results. 6.The viruses that caused the infection can be identified and the disease can be detected.Early stage detection will prevent the spread of infection over a large field. . KeywordsPlant virus pathogens, Deep lea
Applications Invited
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
R. Rajkumar, D. Shanthi, and K. Manivannan
Kaunas University of Technology (KTU)
Deep learning-based anomaly detection in images has recently gained popularity as an investigative field with many global submissions. To simplify complex data analysis, researchers in the deep learning subfield of machine learning employ Artificial Neural Networks (ANNs) with many hidden layers. Finding data occurrences that significantly differ from generalizable to most data sets is the primary goal of anomaly detection. Many medical imaging applications use convolutional neural networks (CNNs) to examine anomalies automatically. While CNN structures are reliable feature extractors, they encounter challenges when simultaneously classifying and segmenting spots that need removal from scans. We suggest a separate and integration system to solve these issues, separated into two distinct departments: classification and segmentation. Initially, many network architecturesare taught independently for each abnormality, and these networks’ main components are combined. A sharedcomponent of the branched structure functions for all abnormalities. The final structure has two branches: onehas distinct sub-networks, each intended to classify a particular abnormality, and the other for segmenting various abnormalities. Deep CNNs training directly on high-resolution images necessitate input layer image compression, which results in the loss of information necessary for detecting medical abnormalities. A guided GradCAM (GCAM) tuned patch neural network is applied to full-size images for anomaly localization. Therefore, the suggested approach merges the pre-trained deep CNNs with class activation mappings and area suggestion systems to construct abnormality sensors and then fine-tunes the CNNs on picture patches, focusing on medical abnormalities instead of training on whole images. A mammogram data set was used to test the deep patch classifier, which had a 99% overall classification accuracy. A Brain tumor image data set was used to test the integrateddetector’s ability to detect abnormalities, and it did so with an average precision of 0.99.
J Kavitha, Shabnam Kumari, K. Manivannan, and Amit Kumar Tyagi
IGI Global
The role of emerging technologies has gained huge attention in recent years. With the advancements in sensor technologies, internet of things (IoT), and remote sensing, large amounts of data are being generated in agriculture. However, making sense of this data and extracting valuable information offers major/ important challenges. Data visualization techniques offer a solution by providing intuitive and interactive ways to represent complex agricultural data. Furthermore, big data analytics techniques enable the extraction of valuable information and patterns from large datasets, contributing to informed decision-making in smart agriculture. This chapter discusses the importance/benefits of data visualization and big data analytics in smart agriculture. It discusses various data sources in agriculture, including sensor technologies, IoT, and remote sensing. This work also provides an overview of the tools and technologies available for data visualization in smart agriculture, including data visualization software, etc., in detail.
K. Manivannan and S. Sathiamoorthy
IOS Press
In the last decades, Tuberculosis (TB) can be considered a serious illness affecting people over the globe and it leads to mortality when left untreated. Chest X-Ray (CXR) is the topmost selection for the recognition of pulmonary diseases in hospitals since it can be cost-efficient and easily available in many nations. But, manual CXR image screening is a huge load for radiologists, which results in a maximum inter-observer discrepancy rate. At present, Computer-Aided Detection (CAD) is a powerful imaging equipment for detecting and screening dangerous ailments. In recent times, Deep Learning (DL) based CAD schemes have demonstrated positive outcomes in the recognition of TB diseases. This study introduces an Egret Swarm Optimization Algorithm with Deep Feature Fusion based Tuberculosis Classification (ESOA-DFFTC) technique on CXR Images. The presented ESOA-DFFTC technique utilizes feature fusion and tuning processes for the classification of TB. To accomplish this, the ESOA-DFFTC model first exploits the Gaussian Filtering (GF) approach for image denoising purposes. Next, the ESOA-DFFTC model performs a feature fusion process using three DL models namely ResNeXt-50, MobileNetv2, and Xception. To enhance the achievement of the DL models, the ESOA-based hyperparameter optimizer is implemented in the study. For TB classification, the ESOA-DFFTC methodology uses an Arithmetic Optimization Algorithm (AOA) with Weight-Dropped Long Short-Term Memory (WDLSTM) methodology. The investigational output of the ESOA-DFFTC system was examined on a benchmark medical imaging dataset. A wide comparative investigation stated the greater achievement of the ESOA-DFFTC system over other current algorithms.
Manivannan K, Suresh T, and Parthiban M
Seventh Sense Research Group Journals
- Sentiment Analysis (SA) may extract data from various text sources like blogs, reviews, and news; later, it categorizes them based on the polarity. Furthermore, big data is generated via social media and mobile networks. The implementation of SA on big data was found to be valuable for the business to take helpful commercial insights from textual-related content. Implementing SA on big data is utilized as a method to classify opinions into different sentiments. This article introduces a new Big Data Analytics Assisted Arithmetic Optimization with Deep Learning Model for Sentiment Classification (BDA-AODLSC) approach. The presented BDA-AODLSC approach exploits BDA tools for sentiment classification. Initially, the BDA-AODLSC approach performs data preprocessing to transform it into a compatible format, and the TF-IDF method is utilized for the word embedding process. An Attention-based Long Short-Term Memory (ALSTM) method is utilized for classifying sentiments, and its hyperparameters can be selected by an Arithmetic Optimization Algorithm (AOA). For managing big data, the Hadoop MapReduce tool is employed. A far-reaching analysis is accomplished to demonstrate the superior accomplishment of the BDA-AODLSC method. The extensive output demonstrates the significant accomplishment of the BDA-AODLSC method over other existing techniques.
Chandramohan Kanmani Pappa, Dasthegir Nasreen Banu, Kumar Vaishnavi, Susila Nagarajan, Manivannan Karunakaran, and Perisetla Kandaswamy Hemalatha
Akademia Baru Publishing
Block chain is generic name to describe the technology used by Bitcoin and other digital currency to record and secure transaction. This technology enables a highly accessible ledger with greatly reduced risk for tampering. The dynamic immutable, data ledger makes ideal for real time monitoring of the shipment of goods. Cloud is an important of distributed storage system of networking. Cloud system need for security, storage management, minimize the cloud cost and fast storage could be improved. The security using new most development security system of block chain technology is used to improve the cloud security. The data owner to be uploads the data on web page and access the folder. The user has been accessing the data on cloud storage using encryption and decryption using block chain based cyber security system. the problem for security in cloud storage because data transmission and data sharing, the alternate for security solution using Post Quantum-proof cryptography algorithm is used to improve the encryption and decryption process and more tight security for block chain technology for the cloud system. The SHA-3 512 Hash function algorithm automatically generate the key for data security enhancements of cloud networks. Post Quantum-proof cryptography algorithm has been improved the encryption Performance and reduced the power consumption, and increase the Latency performance, and boost up the security performances. Finally Post Quantum-proof cryptography algorithm for well support for security system of the cloud networks. Even some passwords, which are often cited as the weakest link in cyber security, may not be necessary.
S. Krishnakumar and K. Manivannan
Springer Science and Business Media LLC
M. Sahaya Sheela, M. Balasubramani, J. J. Jayakanth, R. Rajalakshmi, K. Manivannan, and D. Suresh
Auricle Technologies, Pvt., Ltd.
The wireless sensor network is the most significant largest communication device. WSN has been interfacing with various wireless applications. Because the wireless application needs faster communication and less interruption, the main problem of jamming attacks on wireless networks is that jamming attack detection using various machine learning methods has been used. The reasons for jamming detection may be user behaviour-based and network traffic and energy consumption. The previous machine learning system could not present the jamming attack detection accuracy because the feature selection model of Chi-Squared didn’t perform well for jamming attack detections which determined takes a large dataset to be classified to find the high accuracy for jamming attack detection. To resolve this problem, propose a CNN-based quantum leap method that detects high accuracy for jamming attack detections the WSN-DS dataset collected by the Kaggle repository. Pre-processing using the Z-score Normalization technique will be applied, performing data deviations and assessments from the dataset, and collecting data and checking or evaluating data. Fisher’s Score is used to select the optimal feature of a jamming attack. Finally, the proposed CNN-based quantum leap is used to classify the jamming attacks. The CNN-based quantum leap simulation shows the output for jamming attacks with high precision, high detection, and low false alarm detection.
Manivannan K and Sathiamoorthy S
Seventh Sense Research Group Journals
K Manivannan, Chandrasekar Venkatachalam, S Nagaraj, and S Jeeva
IEEE
Chandrasekar Venkatachalam, K. Manivannan, and Shanmugavalli Venkatachalam
Springer Nature Singapore
K. Manivannan and S. Sathiamoorthy
The Intelligent Networks and Systems Society
: Tuberculosis (TB) detection and classification on chest X-ray (CXR) images remains the most significant task in medical diagnosis. TB is a contagious disorder that affects the pulmonary region, and its diagnosing process often depends on CXR. CXR images are utilized for classifying and detecting TB lesions, including infiltrates, cavities, pleural effusions, and nodules. Manual analysis by radiologists includes a visual assessment of the X-ray images by a skilled physician or radiologist. There were many techniques to automatically classify and detect TB on CXR, including deep learning-based approaches, manual interpretation by radiologists, and computer-aided diagnosis (CAD) systems. This manuscript offers the design of pelican optimization with majority voting ensemble model for tuberculosis detection and classification (POMVE-TDC) technique on the CXR images. The core objective of the POMVE-TDC approach is to classify the incidence of TB on the CXR images. At the primary stage, the POMVE-TDC technique undergoes a contrast enhancement process. Besides, the densely connected network (DenseNet-161) model is applied for the extraction of feature vectors. Meanwhile, a pelican optimization algorithm (POA) based hyperparameter optimizer is designed for the DenseNet-161 model. Finally, a majority voting ensemble classifier comprising graph convolution network (GCN), autoencoder (AE), and extreme learning machine (ELM) models are used. The performance evaluation of the POMVE-TDC technique on the medical dataset highlights the significant outcomes with maximum accuracy of 98.83%
K. Manivannan and S. Sathiamoorthy
IEEE
Accurate Tuberculosis (TB) screening using chest X-rays and artificial intelligence (AI) has the potential in increasing the quality of the healthcare services. Early detection of TB using automated tools find beneficial to decrease the severity level of the diseases. Therefore, the recent developments of the deep learning (DL) models are used in the design of automated TB detection tools. With this motivation, this article focuses on the design of new Harris Hawks optimization with Deep Learning Enabled Tuberculosis Classification (HHODL-TBC) model on chest X-rays. The proposed HHODL-TBC model focuses on the recognition and classification of TB effectually. It follows a three stage process: median filtering based noise removal, U-Net segmentation, MobileNetv2 feature extraction, HHO based hyperparameter tuning, and gated recurrent unit (GRU) classifier. The design of the HHO algorithm assist in the optimal hyperparameter selection of the GRU model. A comprehensive set of simulations were performed for illustrating the improvised results of the HHODL-TBC model, and the results demonstrate the improved outcomes of the HHODL-TBC model with higher accuracy of 99.33%.
S. Krishnakumar and K. Manivannan
IOS Press
The meningioma brain tumor detection is more important than the other tumor detection such as Glioma and Glioblastoma, due to its high severity level. The tumor pixel density of meningioma tumor is high and it leads to sudden death if it is not detected timely. The meningioma images are detected using Modified Empirical Mode Decomposition- Convolutional Neural Networks (MEMD-CNN) classification approach. This method has the following stages data augmentation, spatial-frequency transformation, feature computations, classifications and segmentation. The brain image samples are increased using data augmentation process for improving the meningioma detection rate. The data augmented images are spatially transformed into frequency format using MEMD transformation method. Then, the external empirical mode features are computed from this transformed image and they are fed into CNN architecture to classify the source brain image into either meningioma or non-meningioma. The pixels belonging tumor category are segmented using morphological opening-closing functions. The meningioma detection system obtains 99.4% of Meningioma Classification Rate (MCR) and 99.3% of Non-Meningioma Classification Rate (NMCR) on the meningioma and non-meningioma images. This MEMD-CNN technique for meningioma identification attains 98.93% of SET, 99.13% of SPT, 99.18% of MSA, 99.14% of PR and 99.13% of FS. From the statistical comparative analysis of the proposed MEMD-CNN system with other conventional detection systems, the proposed method provides optimum tumor segmentation results.
K. Manivannan and S. Sathiamoorthy
IEEE
Tuberculosis (TB) is the fifth leading cause of mortality rates across the world, adding nearly 10 million new cases and 1.5 million deaths annually. As TB caused by the bacteria that mostly affect the lungs is prevented and cured, the World Health Organization (WHO) reported a systematic and broad screening for eradicating the disease. Despite its interpretational difficulty and low specificity, poster anterior (PA) chest radiography becomes one of the preferred TB screening techniques. TB is majorly a disease in poor nations; thus, medical practitioners trained to interpret such CXRs were rare. Numerous computer-aided diagnosis (CAD) researches which deal with CXR abnormalities do not give more attention to other diseases (i.e., non-TB). This article devises an Optimal Deep Transfer Learning Model for Automated Tuberculosis Classification (ODTLATC) model. The presented ODTLATC model majorly concentrates on the identification of TB on chest radiographs. To attain this, the ODTLATC model follows a three-stage process such as pre-processing, feature extraction, and classification. At the initial stage, the ODTLATC model employs Weiner filtering (WF) approach for image denoising process. For feature extraction, deep convolutional neural network based residual network (ResNet50) model is utilized. At last, whale optimization algorithm (WOA) with bidirectional recurrent neural network (BiRNN) model is exploited for TB classification purposes. To demonstrate the better performance of the ODTLATC model, a extensive variety of simulations are conducted and the outcomes were inspected on chest radiographs. The comparative study reported the improved performance of the ODTLATC model over other DL models.
S. Krishnakumar and K. Manivannan
Springer Science and Business Media LLC
Balakumaresan Ragupathy and Manivannan Karunakaran
Wiley
In medical imaging, segmenting brain tumor becomes a vital task, and it provides a way for early diagnosis and treatment. Manual segmentation of brain tumor in magnetic resonance (MR) images is a time‐consuming and challenging task. Hence, there is a need for a computer‐aided brain tumor segmentation approach. Using deep learning algorithms, a robust brain tumor segmentation approach is implemented by integrating convolution neural network (CNN) and multiple kernel K means clustering (MKKMC). In this proposed CNN‐MKKMC approach, classification of MR images into normal and abnormal is performed by CNN algorithm. At next, MKKMC algorithm is employed to segment the brain tumor from the abnormal brain image. The proposed CNN‐MKKMC algorithm is evaluated both visually and objectively in terms of accuracy, sensitivity, and specificity with the existing segmentation methods. The experimental results demonstrate that the proposed CNN‐MKKMC approach yields better accuracy in segmenting brain tumor with less time cost.
Balakumaresan Ragupathy and Manivannan Karunakaran
Wiley
This article develops a methodology for meningioma brain tumor detection process using fuzzy logic based enhancement and co‐active adaptive neuro fuzzy inference system and U‐Net convolutional neural network classification methods. The proposed meningioma tumor detection process consists of the following stages as, enhancement, feature extraction, and classifications. The enhancement of the source brain image is done using fuzzy logic and then dual tree‐complex wavelet transform is applied to this enhanced image at different levels of scale. The features are computed from the decomposed sub band images and these features are further classified using CANFIS classification method which identifies the meningioma brain image from nonmeningioma brain image. The performance of the proposed meningioma brain tumor detection and segmentation system is analyzed in terms of sensitivity, specificity, segmentation accuracy, and Dice coefficient index with detection rate.
, K. Manivannan, L. Lakshminarasimman, , M.Janaki Rani, and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
In this research, a highly efficient desensitized FIR filter is designed to enhance the performance of digital filtering operation. With regard to FIR filter design, Multiplication and Accumulation component (MAC) forms the core processing entity. Half-band filters employing Ripple Carry Adder (RCA) based MAC structures have a sizeable number of logical elements, leading to high delay and high power consumption. To minimize these issues, a modified Booth multiplier encompassing SQRT Carry Select Adder (CSLA) based MAC component is proposed for the desensitized filter with reduced coefficients and employing lesser number of logical elements forgiving optimum performance with respect to delay and power consumption. The suggested FIR filter is simulated and assessed using EDA simulation tools from Modelsim 6.3c and Xilinx ISE. The results obtained from the proposed Desensitized FIR filter employing the modified booth multiplier with reduced complexity based SQRT CSLA show encouraging signs with respect to 12.08% reduction in delay and 2.2% reduction in power consumption when compared with traditional RCA based digital FIR filter.
Big data in mortality prediction is rationed with enormous amount of dataset of patients admitted in ICU for the healthcare providers to clarify and interpret about the status of the patients. However, it is difficult to process these large datasets for which big data is used. Mortality prediction of patients admitted in ICU faces many challenges such as imbalance distribution, high dimensionality etc. This paper focuses on overcoming the challenges that arise during the prediction of mortality of ICU patients through pre-processing, feature selection, feature extraction, and classification have been developed. The performance of classifiers has been affected by the high dimensional and unbalanced data of patients. Therefore, a classifier called Extreme Learning Machine has been used for a generalized performance of the classification. In order to predict the rate of mortality for the patients admitted in the ICU by solving the challenges using various methods and tools. For this work, the dataset is collected from a rural hospital that provides medical services in the particular locality. To evaluate the performance of the proposed model, various algorithms have been used and the obtained results are compared. The proposed approach is implemented and experimented in MATLAB software. Various statistical reports are obtained as results and verified. From the results and comparison, it is noticed that the proposed method outperforms than other approaches.
K. Venkatesan, K. Manivannan, S. Devendiran, Arun Tom Mathew, Nouby M Ghazaly, Aadhavan, and S.M. Neha Benny
Elsevier BV
Asif Qureshi, K. Manivannan, Vivek Khanzode, and Sourabh Kulkarni
Inderscience Publishers
A. Selvapandian and K. Manivannan
Wiley
Detection of abnormal regions in brain image is complex process due to its similarity between normal and abnormal regions. This article proposes an automated technique for the detection of meningioma tumor using Gradient Boosting Machine Learning (GBML) classification method. This proposed system consists of preprocessing, feature extraction and classification stages. In this article, Grey Level Co occurrence Matrix (GLCM) features, intensity features, and Gray Level Run Length Matrix features are derived from the test brain MRI image. These derived feature set are classified using GBML classification approach. Morphological functions are used to segment the tumor region in classified abnormal brain image. The performance of the proposed system is evaluated on brain MRI images which are obtained from open access data set. The proposed methodology stated in this article achieves 93.46% of sensitivity, 96.54% of specificity, and 97.75% of accuracy with respect to ground truth images.
A Selvapandian and K Manivannan
Elsevier BV