B. Karthikeyan

Bio: B. Karthikeyan is an academic researcher from VIT University. The author has contributed to research in topics: Steganography & Encryption. The author has an hindex of 10, co-authored 64 publications receiving 330 citations. Previous affiliations of B. Karthikeyan include Viswajyothi College of Engineering and Technology & Shanmugha Arts, Science, Technology & Research Academy.

NIR Based Non-Invasive Blood Glucose Measurement

Abstract: Objectives: This paper describes the method of measurement of glucose concentration in the human blood non-invasively using the near infrared optical technique. Methods/Analysis: In recent medical practice, the concentration of glucose in blood is measured using an invasive techniques which generally involves puncturing finger. In generic few ml of blood whereas in recent practice less than a drop of blood is taken out and passed through the standard chemical tests to measure glucose concentration. These methods are expensive as well as painful. The frequent finger puncturing causes calluses on the skin and also increases the risk of spreading infectious diseases. Findings: So, the development of a non-invasive blood glucose measurement system will be boon to the diabetic patients. This paper describes the method of blood sugar measurement in the human blood non-invasively using the painless near infrared based optical technique. The designed system consists of LED emitting signals of 940 nm wavelength. These optical signals are sent through the fingertip and reflected signals are detected by phototransistor placed beside the LED. The glucose concentration in the blood is determined by analyzing the variation in the intensity of received signal obtained after reflection. The results obtained from the designed system shows the feasibility of using NIR based non-invasive method for the measurement of blood glucose. Applications/Improvements: The described system is majorly useful for diabetic patients. The measurement accuracy of the proposed system can be improved by incorporating it with noise filtering techniques.

Breast Cancer Detection Using Machine Learning

Abstract: In the contemporary world Breast Cancer has become the second leading cause of death among women. Early detection, assessment and followed by appropriate treatment of the cancer can reduce the deadly risk. Technology such as data mining and machine learning can substantially improve the diagnosis accuracy and reduce errors that can be made by medical professionals. This paper emphasizes on the algorithm on which machine can learn to accurately detect breast cancer. Machine Learning algorithm that are composed in this paper are Support Vector Machine, Logistic Regression, K-Nearest Neighbour Classifier, Decision Tree and Random Forest. The Algorithm will be tested accordingly with the World Breast Cancer (WBC) datasets. Benchmarking the aforementioned algorithm and picking the most efficient amongst the algorithm is the aim of this paper.

A novel LSB based image steganography with multi-level encryption

Abstract: Steganography is an art of hiding the existence of secret information by embedding it in a cover and hence preventing the unauthorized access of confidential information. This paper proposes a novel approach of encrypting the plain text into cipher text and embedding it into a color image. Encryption is done in two stages, during first stage it is encrypted by Ceaser cipher technique and in the second stage it is encrypted based on chaos theory. The cipher text obtained after encryption is embedded using 3, 3, 2 LSB replacement algorithm.

A fuzzy based approach for privacy preserving clustering

Abstract: Extracting previously unknown patterns from huge volume of data is the primary objective of any data mining algorithm. In recent days there is a tremendous growth in data collection due to the advancement in the field of information technology. The patterns revealed by data mining algorithm can be used in various domains like Image Analysis, Marketing and weather forecasting. As a side effect of the mining algorithm some sensitive information is also revealed. There is a need to preserve the privacy of individuals which can be achieved by using privacy preserving data mining. In this paper we propose a new approach to preserve sensitive information using fuzzy logic. First we perform clustering on the original data set then we add noise to the numeric data using a fuzzy membership function that results in distorted data. Set of Clusters generated using the distorted data is also relative to the original cluster as well as privacy is also achieved. It is also proved that the number of iterations for performing the clustering process is less in our approach when compared with the traditional approach.

Meta-data based secret image sharing application for different sized biomedical images

Abstract: Usually in image sharing schemes, shares are generated first for a given secret image and then embedded into cover images to produce stego images. These two steps are done sequentially. There exist some relationship in the first step, the size of the secret image and size of the shares which are derived from them. In the proposed method, these two steps are done concurrently. A cover image is chosen and according to its embedding capacity, share is generated and subsequently embedded into chosen cover to produce the stego image. This process is repeated till all the image portions are embedded. While generating share, meta-data (i.e.) header is created for each shares and appended to shares before being embedded. At the destination end, shares are extracted from each stego images and are reassembled into a single original secret image according to the meta-data present in each share. Methods available in the literature embeds uniform sized secret image into cover images of uniform sizes. Using proposed method different sized secret images have been embedded into cover images of varying sizes.

Subset Selection in Regression

Abstract: 8. Subset Selection in Regression (Monographs on Statistics and Applied Probability, no. 40). By A. J. Miller. ISBN 0 412 35380 6. Chapman and Hall, London, 1990. 240 pp. £25.00.

Knowledge and Data Engineering for e-Learning Special Issue of IEEE Transactions on Knowledge and Data Engineering

Abstract: With the advent of the Internet, we are seeing more sophisticated techniques being developed to support e-Iearning. The rapid developme nt of Web-based learning and new concepts like virtual classrooms, virtual laboratories and virtual universities introduces many new issues to be addressed. On the technical side, we need to develop effective e-technologies for supporting distance education. On the learning and management side, we need to consider issues such as new style of learning and different system set-u p requirements. Finally, the issue of standardization of e-Iearning systems should also be considered. In this special issue, our focus will be on the technical side, although other issues related to knowledge and data engineering for e-Iearning may also be considered. Topics: In this special issue, we call for original papers describing novel knowledge and data engineering techniques that support e-Iearning. Preference will be given to papers that include an evaluation of users' experience in using the proposed methods. Areas of interests include, but are not limited to: • Semantic Web technology for e-Iearning • Data modeling (eg., XML) for efficient management of course materials • Searching and indexing techniques to suppo rt effective course notes retrieval • User-centric e-Iearning systems and user interaction management • Profiling techniques to support grading and learning recommendation • Data and knowledge base suppo rt for pervasive e-Iearning • Course material analysis and understanding • Automatic generation of questions and answers • Collaborative communities for e-Iearning

A Review of Wearable Sensor Systems for Monitoring Body Movements of Neonates

Abstract: Characteristics of physical movements are indicative of infants' neuro-motor development and brain dysfunction. For instance, infant seizure, a clinical signal of brain dysfunction, could be identified and predicted by monitoring its physical movements. With the advance of wearable sensor technology, including the miniaturization of sensors, and the increasing broad application of micro- and nanotechnology, and smart fabrics in wearable sensor systems, it is now possible to collect, store, and process multimodal signal data of infant movements in a more efficient, more comfortable, and non-intrusive way. This review aims to depict the state-of-the-art of wearable sensor systems for infant movement monitoring. We also discuss its clinical significance and the aspect of system design.