Qualification: 
Ph.D, M.E
r_lavanya@cb.amrita.edu
Phone: 
+91 9865816977, +91 422 2685000 Ext. 5727

Dr. R. Lavanya joined Amrita Vishwa Vidyapeetham in the year 2006, following her experience as a lecturer in Coimbatore Institute of Technology during the period 2001-2004. She is currently serving as Assistant Professor (Selection Grade) in the department of Electronics and Communication Engineering, Amrita School of Engineering, Coimbatore campus. She completed her B. E. in Electronics and Communication Engineering from Kumaraguru College of Technology in 2001 and M. E. in Applied Electronics from Coimbatore Institute of Technology in 2005. She earned her Ph. D. degree from Anna University, Chennai, in April 2015. Her Ph.D. thesis was on "Computer Aided Diagnosis of Breast Cancer". She is the coordinator of M.Tech Biomedical Engineering program at Amrita. She is actively involved in research and guidance, and supervising PhD scholars as well as PG and UG students. She has published her research work in journals and conferences. She is a member of the Institution of Electronics and Telecommunication Engineers.

Education

  • 2015: Ph. D. in Biomedical Image Processing
    Anna University
  • 2006: M. Tech. in Applied Electronics
    Coimbatore Institute of Technology
  • 2001: B. Tech. in Electronics and Communication Engineering
    Kumaraguru College of Technology

Professional Experience

Year Affiliation
July 2015 till date Assistant Professor (Selection Grade), Amrita Vishwa Vidyapeetham
Domain : Signal processing, Biomedical Image Processing, Computational Intelligence
Role : Teaching, Research and Guiding Projects
July 2008 to June 2015 Assistant Professor (Senior Grade), Amrita Vishwa Vidyapeetham
Domain : Signal processing, Biomedical Image Processing, Machine Learning, Soft computing
Role : Teaching, Research and Guiding Projects
July 2007 to June 2008 Senior Lecturer, Amrita Vishwa Vidyapeetham
Domain : Signal Processing, Machine Learning
Role : Teaching, Research and Guiding Projects
July 2006 to June 2007 Lecturer, Amrita Vishwa Vidyapeetham
Domain : Signal Processing
Role : Teaching, Research and Guiding Projects
June 6 to October 2004 Lecturer, Coimbatore Institute of Technology
Domain : Electronic circuits and systems
Role : Teaching and Research

Academic Responsibilities

SI.No. Position Class / Batch Responsibility
1 In-charge, Intelligent Electronics Systems Lab/Maker’s space Lab Since October 2020 In-charge and sharing facilities of the lab to students and research scholars for project / research work 
2 TAG Lead - Computational Intelligence for Electronics Systems Since Jan 2019 Guide and encourage TAG members towards active research, publications and funded project
2 PG Coordinator M.Tech BME (2016-18, 2017-19, 2018-2020, 2019-21 batches)
  • Curriculum design
  • Admission process
  • Student induction
  • Counseling
  • Academic Coordination
  • Initiating industry collaboration
  • Periodic review of Projects/Internships
  • Guidance for placement and exchange programs
  • Organizing conferences and talks
3 PG Co-coordinator M.Tech BME (2015-17 batch)
  • Curriculum design
  • Admission process
  • Student induction
  • Counseling
  • Periodic review of Projects
4 Program Planning and Coordination Committee member 2019-2020

The committee is responsible for preparing detailed proposal on new programmes and revamping of existing programmes to support the students for adapting to new technological trends and career path

5 Department Publication coordinator (2014-15) All ECE related programs Ensuring Quality of publication
6 Department Academic Coordinator (2010-2012) All ECE related programs Coordinating regular courses, Runtime redo, Vacation and re-registration courses
7 Class Counsellor 2013 B.Tech EIE
  • Counseling
  • Supporting academic coordination
  • Project review
8 Class Advisor 2010-2014 B.Tech ECE ‘B’
  • Counseling
  • Supporting academic coordination
  • Project review
9 Class Advisor 2007-2010 B.Tech ECE ‘C’
  • Counseling
  • Supporting academic coordination
  • Project review

Courses Handled

  • Signals and Systems
  • Digital Signal Processing
  • Introduction to Digital Signal Processing
  • Introduction to Soft Computing
  • Analog Communication
  • Digital Communication
  • VLSI Circuit Design
  • Signals and Systems Lab
  • Digital Signal Processing Lab
  • Hardware Design Lab
  • Computer Communication and Networking Lab
  • Analog Communication Lab
  • Digital Communication Lab
  • Computer Programming
  • Solid State Devices and Circuits Lab
  • Digital Systems Lab
  • Electrical Engineering and Electronics Lab
  • Project
  • Signal Processing
  • Digital Signal Processing
  • Soft Computing
  • Biomedical Signal Analysis
  • Biomedical Image Processing
  • Computational Medical Diagnostics
  • Multidimensional Signal Processing
  • Research Methodology and Medical Ethics
  • Advanced Signal Processing
  • Deep Learning Techniques
  • Biomedical Engineering Lab
  • Open/Live-in Labs
  • Internship
  • Dissertation

Specialized Courses Developed

Course Name Specialization Programme Outcome
16BM763 Mammogram Image Analysis Knowledge in applying appropriate image processing for mammograms with the ability to develop automated breast cancer diagnostic systems aiming superior performance
18BM702 Special Topics in Biomedical Image Processing Knowledge in advanced algorithms for image segmentation, representation, reconstruction, registration and restoration
18BM741 Essentials of Telemedicine Knowledge in telemedical technology , telemedicine standards, m-health, e-health and its applications

Innovations in Teaching - Learning

SNo Innovation Method Description with Tools used
1. Online course design based on NPTEL Introduced online course delivery for M.Tech BME (2018 revision) in18BM611 Anatomy and Physiology based on video lectures from reputed institutions, by adopting the syllabus accordingly
2. Fractal courses Introduced fractal courses for M.Tech BME (2018 revision) to enable exposure to diverse and specialized areas

Participation in Faculty Development / STTP / Workshops /Conferences

SNo Title Organization Period Outcome
1. Two-day workshop on “Recent Trends in image Analysis” Amrita School of Engineering, Coimbatore. June 2009 Insight on basics of Image Processing
2. Two-day workshop on Research Issues in Digital Image processing Dr.Mahalingam College of Engineering and Technology, Coimbatore. Nov 2009 Insight on basics of Image Processing
3. One-day workshop on MATLAB and Simulink for Engineering Applications Amrita School of Engineering, Coimbatore Apr 2014 Insight on basics of MATLAB and Simulink
4. Two-day Workshop on Signal and Image Processing Applications Using Xilinx System Generator Amrita School of Engineering, Coimbatore. Mar 2013 Insight on hardware implementation of signal processing algorithms
5. International Conference on Artificial Intelligence and Evolutionary Algorithms in Engineering Systems Noorul Islam College of Engineering, Kanyakumari. November 2014 Paper presentation and exchange of ideas on Evolutionary algorithms

Organizing Faculty Development / STTP / Workshops /Conferences

SNo Title Organization Period Outcome
1. Ten-day UGC sponsored Refresher course on VLSI Design (Coordinator) Amrita School of Engineering, Coimbatore Dec 2007 Insight on basics of VLSI Design
2. Four-day National Workshop on Signal Processing for Biomedical Applications (Coordinator) Amrita School of Engineering, Coimbatore June 2009 Insight on Pre-processing and Segmentation of biomedical images, feature extraction, Machine learning Techniques, Tools for Processing and Analysis of Biomedical signals, Mammogram Analysis, Epilepsy Seizure Prediction and Speech disorders
3. Two-day National Workshop on Image Processing for Biomedical Applications (Coordinator) Amrita School of Engineering, Coimbatore June 2015 Insight on image processing techniques for diagnosis of breast cancer and retinopathy, Metaheuristic techniques for image processing such as Genetic Algorithm and Particle Swarm Optimization, Registration and reconstruction of medical images, OpenCV library for image processing.
4. Three-day National Workshop on Biomedical Signal Acquisition and Conditioning (Organizing team) Amrita School of Engineering, Coimbatore December 2015 Insight on hardware circuits for acquiring Biomedical Signals
5. Two-day National Workshop on Image Processing for Biomedical Applications (Organizing team) Amrita School of Engineering, Coimbatore December 2016 Insight on Computer Aided Diagnosis, CAD techniques for Retinopathy, CAD techniques for Lung Cancer, Hands on experience
6. INDICON 2018 (Publication Committee) Amrita School of Engineering, Coimbatore June 2015 Exchange of recent research trends in circuits & systems, signal & image processing, power electronics & drives, power systems, control & instrumentation, computer science and communication & networks

Academic Research – PhD Guidance

SNo Name of the Scholar Specialization / Title Registration Milestone
1. Ms.DeviVijayan Biomedical Image Processing/Computer-aided breast cancer diagnosis June 2015 Open Seminar
2. Ms.G.Suguna Biomedical Image Processing/ Automated Techniques for Glaucoma Diagnosis Jan 2016 Qualifying Exam
3. Mr.M.Ganesan Multivariate Signal Processing / Fault Detection in Satellite Power Systems using Deep Learning Jan 2009 Open Seminar
4. Mr.A.Gandhimathinathan Multivariate Signal Processing / Early Diagnosis of Faults in Safety Critical Systems Jan 2019 Course work

Academic Research – PG Projects

SNo Name of the Student Programme Specialization Duration Status
1. B.Saranya M.Tech VLSI Design 2008-2010 Completed -Paper published
2. Sreediya R M.Tech VLSI Design 2010-2012 Completed - Paper published
3. Shilpa Gopalakrishnan M.Tech VLSI Design 2012-2014 Completed -Paper published
4. Dhadma Balachandran M.Tech Biomedical Engineering 2014-2016 Completed -Paper published and indexed in SCOPUS
5. Dharani V M.Tech Biomedical Engineering 2015-2017 Completed -Paper published and indexed in SCOPUS
6. Amala C Nair M.Tech Biomedical Engineering 2016-2018 Completed -Paper published and indexed in SCOPUS
7. Praveena S M.Tech Biomedical Engineering 2017-2019 Completed -Paper published and indexed in SCOPUS
8. Srinithi V M.Tech Biomedical Engineering 2017-2019 Completed -Paper published and indexed in SCOPUS
9. Shanchita B M.Tech Biomedical Engineering 2018-2020 Completed -Paper published and indexed in SCOPUS
10. Ahalya Krishnan M.Tech Biomedical Engineering 2018-2020 Completed -Paper published and indexed in SCOPUS
11. Raveenthini M M.Tech Biomedical Engineering 2019-2021 Ongoing
12 Srinivashini N M.Tech Biomedical Engineering 2019-2021 Ongoing

Sponsored Research

SNo Title Agency Amount Date of Submission Status
1. Predictive Diagnosis in Nuclear Power Plants using Deep Learning Techniques BRNS (DAE) Rs. 24,70,050 3 years Sanctioned in December 2019

Collaborations

  • The sponsored project “Predictive Fault Diagnosis in Nuclear Power Plant using Deep Learning Techniques” is carried out in collaboration with Indira Gandhi Centre for Atomic Research (IGCAR), Kalpakkam.

Publications

Publication Type: Journal Article

Year of Publication Title

2021

Dr. Lavanya R., Devi, N., and Ganesan, M., “Fault detection in satellite power system using convolutional neural network”, Telecommunication Systems, vol. 76, pp. 1-7, 2021.[Abstract]


Satellite failures account for heavy, irreparable damages, especially when associated with the Power System which is the heart of a satellite. Anomalies in Satellite Power System (SPS) can lead to complete failure of the mission. This demands the need to understand the causes of power system related failures. Huge number of sensors installed in a satellite system conveys information regarding the health of the system. The conventional manual level checking of sensors can be augmented with data driven fault diagnosis approach to reduce the false alarm and burden on operating personnel. The latter has the advantage of exploiting the interrelationship between sensor measurements for fault diagnosis. In this work, Convolutional Neural Network (CNN) is trained on satellite telemetry data for sensor fault detection in SPS. Various processing schemes in time and frequency domains were explored to process the input data to CNN. Promising results were obtained with combination of Stockwell transform (S-transform) and CNN for data processing and classification, respectively. Advanced Diagnostics and Prognostics Testbed (ADAPT), a publicly-available dataset was analysed and used for validating the proposed algorithm, yielding an accuracy as high as 96.7%, precisison of 0.9, F1 score of 0.95 and AUC equal to 0.976.

More »»

2021

Ganesan and Dr. Lavanya R., “A Deep Learning Approach to Fault Detection in Satellite Power System using Gramian Angular Field”, International Journal of Engineering Systems Modelling and Simulations (Accepted), 2021.

2021

D. Vijayan and Dr. Lavanya R., “Ensemble of density-specific experts for mass characterization in mammograms”, Signal, Image and Video Processing, pp. 1-9, 2021.[Abstract]


Breast cancer is considered the most serious cancer in women, among other prevalent cancer types. Chances of survival can be significantly increased through early detection, for which mammogram is considered to be the gold standard. This work addresses the development of a computer-aided diagnosis (CAD) system for analysis of mammographic masses. However, different mammographic tissue densities exhibit different characteristics and present abnormalities diversely. This renders a unified CAD framework for evaluation of masses, ineffective. To this end, we propose an ensemble framework for mass characterization, comprising different experts each specialized for a particular tissue category. Specifically, three segmentation-free feature descriptors including local binary pattern (LBP), scale-invariant feature transform (SIFT) and gray-level co-occurrence matrix (GLCM) are extracted from the regions of interest (ROIs), followed by individual classification with each feature descriptor using four different learning models, viz. support vector machine (SVM), artificial neural network (ANN), random forest (RF) and extreme grading boosting machine (XG Boost). All these 12 combinations are explored for each of the four breast density categories separately, to determine the best feature–classifier combination for a given category. The proposed ensemble scheme was validated on 1057 suspicious regions from digital database for screening mammograms (DDSM), demonstrating an improved performance when compared to state-of-the-art single learning framework modeled on all density categories collectively.

More »»

2020

G. Suguna and Dr. Lavanya R., “Performance Assessment of EyeNet Model in Glaucoma Diagnosis”, Pattern Recognition and Image Analysis, Springer. (Accepted), 2020.

2020

D. Vijayan and Dr. Lavanya R., “Hybrid Local Descriptor for Improved Detection of Masses in Mammographic CAD Systems”, International Journal of Advanced Intelligence Paradigms, Inderscience (Accepted), 2020.

2020

A. Krishnan and Dr. Lavanya R., “Model for predictive maintenance strategy in medical equipment’s”, Journal of Advanced Research in Dynamical and Control Systems, vol. 12, pp. 1067-1071, 2020.[Abstract]


In healthcare sectors, device management and maintenance systems are designed for providing improved patient care, reducing adverse incidents and to increase the utilization time of the devices. Predictive maintenance helps in reducing equipment downtime by reducing delays in corrective action process. Reliability engineering can used to anticipate which component would fail and when it is about to fail, taking into consideration the working environment, usage, and the age of the device. Most of the maintenance strategies primarily focuses on predicting the remaining useful time of any component in isolation. A major limitation to be overcome is to develop a framework that can track the degradation of multiple interacting components. This paper presents a neural network (NN) model to predict the number of parts that might fail in future and needs to be stocked. Validation is performed by predicting failures over 7 years (2013–2019) based on a simulated learning algorithm, using failure event dataset. The efficiency of the model is evident through the low values of root mean square and absolute errors.

More »»

2019

M. Sushil, Suguna, G., Dr. Lavanya R., and M. Devi, N., “Performance comparison of pre-trained deep neural networks for automated glaucoma detection”, Lecture Notes in Computational Vision and Biomechanics, vol. 30, pp. 631-637, 2019.[Abstract]


This paper addresses automated glaucoma detection system using pre-trained convolutional neural networks (CNNs). CNNs, a class of deep neural networks (DNNs), extract features of high-level abstractions from the fundus images, thereby eliminating the need for hand-crafted features which are prone to inaccuracies in segmenting landmark regions and require excessive involvement of experts for annotating these landmarks. This work investigates the applicability of pre-trained CNNs for glaucoma diagnosis, which is preferred when the dataset size is small. Further, pre-trained networks have the advantage of the quick model building. The proposed system has been validated on the High-Resolution (HRF), which is a publicly available benchmark database. Results demonstrate that among other pre-trained CNNs, VGG16 network is more suitable for glaucoma diagnosis. © Springer Nature Switzerland AG 2019.

More »»

2018

A. C. Nair and Dr. Lavanya R., “Enhanced Empirical Wavelet Transform for Denoising of Fundus Images”, Communications in Computer and Information Science, vol. 837, pp. 116-124, 2018.[Abstract]


Glaucoma is an ophthalmic pathology caused by increased fluid pressure in the eye, which leads to vision impairment. The evaluation of the Optic Nerve Head (ONH) using fundus photographs is a common and cost effective means of diagnosing glaucoma. In addition to the existing clinical methods, automated method of diagnosis can be used to achieve better results. Recently, Empirical Wavelet Transform (EWT) has gained importance in image analysis. In this work, the effectiveness of EWT and its extension called Enhanced Empirical Wavelet Transform (EEWT) in denoising fundus images was analyzed. Around 30 images from High Resolution Fundus (HRF) image database were used for validation. It was observed that EEWT demonstrates good denoising performance when compared to EWT for different noise levels. The mean Peak Signal to Noise Ratio (PSNR) improvement achieved by EEWT was as high as 67% when compared to EWT.

More »»

2015

Dr. Lavanya R., Nagarajan, N., and Devi, N., “Computer-aided Diagnosis of Breast Cancer by Hybrid Fusion of Ultrasound and Mammogram Features”, Advances in Intelligent Systems and Computing, vol. 325, pp. 403-409, 2015.[Abstract]


Ultrasound images are increasingly being used as an important adjunct to X-ray mammograms for diagnosis of breast cancer. In this paper, a computer-aided diagnosis system that utilizes a hybrid fusion strategy based on canonical correlation analysis (CCA) is proposed for discriminating benign and malignant masses. The system combines information from three different sources, i.e., ultrasound and two views of mammogram, namely, mediolateral oblique (MLO) and craniocaudal (CC) views. CCA is employed on ultrasound-MLO and ultrasound-CC feature pairs to explore the hidden correlations between ultrasound and mammographic view. The two pairs of canonical variates are fused at the feature level and given as input to support vector machine (SVM) classifiers. Finally, decisions of the two classifiers are fused. Results show that the proposed system outperforms unimodal systems and state-of-the-art fusion strategies.

More »»

2015

R. M. Kirubaa, Dr. Lavanya R., Kotwal, N. P., and Vijayan, D., “Change Detection In Mammogram Images Using Fuzzy C- Means Clustering”, International Journal of Applied Engineering Research, vol. 10, no. 11, pp. 29825-29834 , 2015.[Abstract]


Experts have estimated that breast cancer is diagnosed in about one out of every eight women. At present mammography is the most efficient tool for the screening of breast cancer and studies show that misinterpretation is an important cause of missing breast cancer. In this paper we propose a computer aided detection system to identify changes in temporal mammographic images which would aid radiologists in the early and accurate detection of mammographic lesions. This system involves pre-processing, registration, generation of difference image and the analysis of difference image to obtain the changed and unchanged regions of the lesion. The novelty of this research work is to effectively find changes in mammogram images obtained from consecutive screening rounds using fuzzy c-means (FCM) clustering. The efficiency of FCM is compared with K-means clustering using overall error (OE) and kappa coefficient (KC). Experimental results show that the proposed method is a better alternative to the K-means clustering method. These techniques have been tested on mammogram images obtained from a private hospital. More »»

2015

G. Menon, Dr. Palanisamy T., and Dr. Lavanya R., “Hardware Architecture for Variational Mode Decomposition for Breast Cancer Feature Extraction on Ultrasound Images”, International Journal of Applied Engineering Research, vol. 10, no. 7, pp. 16343-16354, 2015.[Abstract]


Ultrasound (US) imaging proved to be less harmful than the traditional mammography is used for diagnosing breast cancer and this has helped reduce the number of unnecessary biopsies.The most important feature of malignant breast lesion is its infiltrative nature in US images.This infiltrative nature having composed of the frequency components that are adjacent to the lower frequency band contains the local variances that are characterized by Variational Mode Decomposition (VMD).On comparison with the existing decomposition models such as Empirical Mode Decomposition (EMD) and Wavelet Transform (WT) which are known for their limitations like sensitivity to noise and sampling which could only partially be addressed by more mathematical attempts to this decomposition problem, like synchrosqueezing, empirical wavelets or recursive variational decomposition.To overcome these limitations, a non-recursive VMD was selected.In this paper, we have presented an algorithmbased on VMD and a suitable architectureto obtain the infiltrative nature of the malignant breast lesion from the US image.

More »»

2014

Dr. Lavanya R., Nagarajan, N., and M. Nirmala Devi, “Multiview mammographic CADx system based on correlation analysis”, International Journal of Graphics and Image Processing, vol. 4, 2014.

2014

Dr. Lavanya R. and Nagarajan, N., “Comparison of fusion schemes for two-view analysis of breast cancer using mammograms”, International Journal of Advanced Computer Technology, vol. 3, 2014.

2014

Dr. Lavanya R., Nagarajan, N., and M. Devi, N., “False positive reduction in computer aided detection of mammographic masses using canonical correlation analysis”, Journal of Theoretical and Applied Information Technology, vol. 59, pp. 139-145, 2014.

2014

Dr. Lavanya R. and Nagarajan, N., “Information fusion in CAD systems for breast cancer diagnosisusing mammography and ultrasound imaging: A survey”, Journal of Artificial Intelligence, vol. 7, pp. 113-122, 2014.[Abstract]


Breast cancer is the highest incident cancer in women and a serious threat to a woman's life. Early detection and treatment of breast cancer can reduce the mortality rate. Currently, mammography is widely employed for routine screening of breast cancer. Ultrasound imaging is used as an important adjunct to mammography, especially in the post-screening (diagnostic) phase. Irrespective of the imaging modality, several factors including the level of radiologists' expertise affect the accuracy of breast cancer detection and diagnosis. Computer Aided Detection/Diagnosis (CAD) systems are objective in nature as opposed to the subjective analysis made by radiologists. Many studies show that the use of a CAD system as a second reader has the potential to improve the accuracy of breast cancer detection and diagnosis. Recently, integration of information from multiple sources is gaining wide popularity in data analysis. Information fusion in CAD systems would serve to mimic the radiologist's practice of combining information from multiple mammographic views and from multiple imaging modalities like ultrasound imaging and mammography to arrive at better diagnostic decisions. This study reviews the literature on such CAD systems based on mammograms and ultrasound images for breast cancer detection and diagnosis.

More »»

2013

B. .A.Sabarish and Dr. Lavanya R., “Modified Leach Protocol for Wireless Sensor Network”, International Journal of Computer Applications, vol. 62, pp. 1-5, 2013.

Publication Type: Conference Proceedings

Year of Publication Title

2020

B. Ashwini, Ganesan, M., and Dr. Lavanya R., “A survey on the advancement of ECG classification using deep convolutional neural network”, Journal of Advanced Research in Dynamical and Control Systems, vol. 12, 5 vol. pp. 1072–1078, 2020.[Abstract]


Classification of Electrocardiograph (ECG) signal has an ultimate position in clinical analysis of heart diseases. In this paper, we have made a survey on advanced way of classifying ECG. It is based on multivariate time series classification. Here combining Dynamic Time Warping (DTW) and Nearest Neighbor (k-NN) classification was found to give a much desired result. Usually as the data point set increases, the time consumption of DTW and 1-NN is very much costly and it is less effective for feature- based classification methods. So, here they are modifying the procedure of outmoded feature established method by a new feature learning techniques. Here they have also claimed about the Multi Channels Deep Convolutional Neural Networks (MCDCNN), as a novel deep learning framework for classifying the recorded ECG beats. In the proposed system, initially every channel learns the features of all specific single variate time series information which will then be combined to give a feature representation to the final outcome layer. Then, these learnt features are given to the Multilayer Perceptron (MLP) for further classification. As a result on carrying out the experiments based on real time data, this model has proven it is more efficient and accurate than all the previous proven models.

More »»

2018

M. Ganesan, Dr. Lavanya R., and Sumesh, E. P., “A Survey on Ballistocardiogram to study the Mechanical Activity of Heart”, Proceedings of the 2017 IEEE International Conference on Communication and Signal Processing, ICCSP 2017. Institute of Electrical and Electronics Engineers Inc., pp. 0557-0561, 2018.[Abstract]


Ballistocardiogram (BCG) refers to mechanical activity of the heart. It is a non-invasive technique that provides information on cardiovascular forces. Few studies have demonstrated the correlation that BCG exhibits with the cardiac output and pathological condition of the heart. However there is no clear proof of the relationship between cardiac cycles and BCG wave characteristics. Different BCG acquisition techniques exist, with differing sensor positions that include arm, feet and spine. To use BCG for clinical applications, the differences in the parameter of signals using different methods have to be reduced and a global template as that of Electrocardiogram (ECG) has to be obtained. So there is a need to understand the way BCG is captured and the nature of the signal. In this paper we have focused on the survey of various techniques for acquiring BCG, denoising methods and classification techniques. Recent developments in BCG are discussed along with comparative studies. The purpose of this paper is to investigate the feasibility of BCG in the field of medical diagnosis.

More »»

2017

Dr. Lavanya R., Swathi, O. N., and Ganesan, M., “Peak Detection and Feature Extraction for the Diagnosis of Heart Diseases”, Proceedings in IEEE International Conference in Computing, Communications and Informatics. 2017.[Abstract]


Patient monitors with arrhythmia detection will enhance the quality of living of human by aiding in prediction of diseases in much early stage. In this work we have developed an algorithm for classifying the ECG signals into normal and arrhythmic signal. Here we have detected the R peaks from denoised ECG signal with an accuracy of 97.56%. Extracted features from the signal in both time and frequency domain and the signals are classified into normal and abnormal signals using support vector algorithm. The accuracy of the algorithm is tested by applying on MIT-BIH arrhythmia database and we obtained an overall 80% classifier accuracy.

More »»

2014

ShilpaGopalakrishnan and Dr. Lavanya R., “Bilateral analysis of Mammograms for False Positive Reduction using Eigen ROI Approach”, International conference on Trends in Technology for Convergence. 2014.

2014

D. Pramod, DasariVenkatesh,, VenkataKarthik, J., Raghunathan, R., Kiran, C. S. Sai, and Dr. Lavanya R., “Identification of Corresponding Lesions from Ipsilateral Mammograms using eigen ROI approach”, International Journal of Computational Intelligence and Informatics. 2014.

2014

Y. Anjana, .Vinutna, G., Kalimatha, N., Namrata, B., Abinaya, A., and Dr. Lavanya R., “Handling Missing Data In Ipsilateral Mammograms For Computer Aided Breast Cancer Diagnosis”, International Journal of Scientific & Engineering Research, vol. 5. 2014.[Abstract]


Breast cancer is the most fatal among the cancers detected in women. Mammography is the most common and efficient tool for early detection of breast cancer. In mammography images of breast are captured in two standard views namely Mediolateral oblique (MLO) view and Craniocaudal (CC) view. Radiologists generally analyze both the views during diagnosis. In recent years as the number of cases to be diagnosed is increasing significantly, computer aided diagnosis (CAD) systems were developed. In the datasets used by these systems, often there is a chance for the presence of missing values (MVs) in any one of mammographic views due to various reasons. Some of these include obscuration of mass by a dense breast tissue, region of interest being out of frame during the image acquisition etc. This results in the use of data from single view alone for diagnosis. But the diagnostic performance of CAD systems is better when multi-view data is used. In this paper the use of Iterative Singular Value Decomposition (ISVD) imputation is proposed to handle missing values in order to preserve the advantage of using multi-view data during the diagnosis. Classification accuracy and Kappa statistics are the metrics used to assess the performance of ISVD scheme for different percentages of MVs ranging from 1-15%. Experimental results demonstrated that diagnosis using multi-view following ISVD performed at least as well as and most of the time better than the systems using single view.

More »»

2012

Dr. Lavanya R., Sunith, G. N., Pradeep, S., DivyaBonagiri,, Anusha, A., and AbhipsaSamantra, “Computer Aided Detection and Diagnosis of Masses in Dense Tissue Breasts”, Proceedings of International Conference on Information Technology, Electronics & Communications. 2012.[Abstract]


Mammogram is the widely accepted imaging method used for screening breast cancer. Computer aided detection (CAD) systems intend to provide assistance to the radiologists in detecting breast cancer. However the detection of masses by CAD system is prone to a high rate of false positives. Radiologists often use bilateral analysis ie, symmetry between left and right breast to improve detection performance. In this work, a computer aided bilateral image analysis based on eigen region of interest (ROI) approach is used in order to reduce the false positive rate. It has been found that the proposed method outperforms the conventional false positive reduction techniques.

More »»

2010

Dr. Lavanya R. and .Saranya, B., “High Speed, Low Complexity, Folded, Polymorphic Wavelet Architecture Reconfigurable Hardware”, International Journal of Advanced Science and Technology, vol. 18. 2010.[Abstract]


The main aim of this paper is to design and implement a high speed, low complexity and polymorphic architecture for reconfigurable folded wavelet filters. 5/3 wavelet results are incorporated into the 9/7 data path which reduces the number of adders compared to other solutions and also allows on the fly switching between the filters. The proposed work is to improve the speed of this reconfigurable architecture. This is accomplished by scheduling. A weight based scheduling algorithm has been used in this paper. This is an analysis method to improve inter task communication as well as data dependencies among tasks which will reduce the overall communication overhead and processing time.

More »»

2009

P. .L.Chidambaram, .B.Ganeshkumar, R., .Naren, M., .Raghuram, K., .Youvarajah, K., and Dr. Lavanya R., “Design of Multiple Description Image Coding System Applying Directional Lifting Based Transform”. 2009.

Publication Type: Conference Paper

Year of Publication Title

2020

B. Bhardwaj, S. R., D., Anjali, S., Ganesan, M., and Dr. Lavanya R., “Brain on a-board implementation of Neural Networks”, in 2020 6th International Conference on Advanced Computing and Communication Systems (ICACCS), Coimbatore, India, 2020.[Abstract]


In an era of digital computing and software supremacy, analogue computing and electronics often take a back seat. And it is only fair to say that current computers, general applications digital devices, are relied on because of their efficient software rather than hardware. The current flaws in combination with new age computing such as machine learning, and the limitations become glaring especially on devices on the lower end. To overcome this, it becomes necessary to rely on analogue techniques to allow efficient computation and fulfilling the need for dedicated Machine Learning devices capable of similar accuracy as software models. In this paper, with the goal of developing analogue neural networks, the accuracy obtained for on board sensor data was 77% same as a software approach.

More »»

2020

S. B and Dr. Lavanya R., “Root Cause Analysis on the Nature of Breakdown for Computerized Maintenance Management System”, in 2020 5th International Conference on Communication and Electronics Systems (ICCES), Coimbatore, India, 2020.[Abstract]


Equipment failures are triggered by initiating events called root causes, the identification of which can aid in preventing the failures. However, unpredictable nature of certain failures renders the determination of root causes, difficult. In this paper, a knowledge-based approach that uses a combination of Six Sigma technique, Cause and Effect technique, and Pareto Analysis technique is implemented to assist and investigate the breakdown of assets through Root Cause Analysis (RCA). The foremost objective of this RCA method is to reduce adverse events from recurrence, in a proactive manner. The proposed method has been validated using ventilator breakdown data collected from Narayana Health (NH) over seven years, encompassing failures and associated operating parameters.

More »»

2019

V. Srinithi and Dr. Lavanya R., “Novel Colour Derivative Based Approach for CDR Estimation in Glaucoma Diagnosis”, in 2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI), Tirunelveli, India, 2019.[Abstract]


Glaucoma is a chronic eye disease and a leading cause of blindness, after cataract. It can be diagnosed using Cup-to-Disc ratio (CDR) which can be assessed using a few imaging modalities. Digital Fundus Photography (DFP) is a widely employed for imaging the interior surface of the eye and hence popular in glaucoma diagnosis. Computer Aided Diagnosis (CAD) can help ophthalmologists in diagnostic decision making, reducing their burden and the cost and time involved. However, automatic extraction of CDR depends on localization and segmentation of landmarks such as Optic Nerve Head (ONH) localization, optic disc (OD), optic cup (OC). The segmentation procedure is affected by many factors including blood vessel occlusion, non-uniform illumination and poor contrast. The error in segmentation will propagate to rest of the stages and reflect in poor diagnosis and the use of colour information is not fully exploited in state-of-art techniques for glaucoma diagnosis. This has motivated the need for new techniques for clinical feature extraction that reduces segmentation dependent errors. In this work, clinical feature is extracted using vector based colour derivative approach in CIE L*a*b* colour space using CIE94 and CIE2000 colour distance. A comparative analysis between the two colour metrics and the resulting classification accuracies using bagged tree classifier are compared. Publicly available benchmark databases (RIM ONE, HRF) will be used for validating the algorithm.

More »»

2019

S. Praveena and Dr. Lavanya R., “Superpixel based Segmentation for Multilesion Detection in Diabetic Retinopathy”, in 2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI), Tirunelveli, India, 2019.[Abstract]


Diabetic Retinopathy (DR) is a progressive chronic vision-threatening disease of retinal microvasculature associated with prolonged hyperglycaemia, hypertension and other conditions associated with diabetes. Indicators of DR include different kinds of lesions appearing on the retinal surface that are visible in a Digital Fundus Photograph (DFP). Localization of lesions and visual perception is essential to aid physicians in understanding the severity of the condition and to plan an appropriate treatment procedure for the patient. Segmentation inaccuracies due to factors like subtle nature of abnormalities and interference of blood vessels reflect in reduced classification accuracy in case of Feature Based Machine Learning (FML). While Pixel Based Machine Learning (PML) can overcome these issues, they require high computational capabilities and are redundant. Non-segmentation approaches like deep learning have been employed as an alternative for DR diagnosis. However, these techniques directly grade the image through classification and do not allow for visual perception. Thus we have used an intermediate approach called Super-pixel based segmentation that can overcome the problems in FML and PML while retaining the advantages of both. They are consistent with human visual perception and also overcome the data insufficiency problem. In this paper, we have compared the results of multilesion detection associated with DR using super-pixels segmented from three different algorithms namely, Compacted Watershed (CWS), Simple Linear Iterative Clustering (SLIC) & Linear Spectral Clustering (LSC) under a single unified framework. Experimental results show that LSC over-performs both SLIC and CWS quantitatively and qualitatively.

More »»

2019

A. Rajkumar, Ganesan, M., and Dr. Lavanya R., “Arrhythmia classification on ECG using Deep Learning”, in 2019 5th International Conference on Advanced Computing Communication Systems (ICACCS), Coimbatore, India, 2019.[Abstract]


In this paper, an intellectual based electrocardiogram (ECG) signal classification approach utilizing Deep Learning (DL) is being developed. ECG plays important role in diagnosing various Cardiac ailments. The ECG signal with irregular rhythm is known as Arrhythmia such as Atrial Fibrillation, Ventricular Tachycardia, Ventricular Fibrillation, and so on. The main aspire of this task is to screen and distinguish the patient with various cardio vascular arrhythmia. This examination encourages us to recognize diverse kinds of arrhythmia utilizing Deep Learning algorithm. Here we use Convolutional Neural Network (CNN) a DL algorithm which is efficient in classifying signals. Utilizing CNN, features are learned Automatically from the time domain ECG signals which are acquired from MIT-BIH Database from Physiobank.com. The feature adapted specifically replaces manually extracted features and this analysis will help the Cardiologists in screening the patient with Cardiac illness effectively. The CNN is trained, tested using ECG Dataset obtained from MIT-BIH Database and from the signal 7 of arrhythmia were classified. The proposed system is compared for Various Activation function by varying the number of epochs. From the result obtained we came to know that ELU activation function gives better result with an accuracy of 93.6% and with a loss of 0.2.

More »»

2019

K. Kiruthika, Vijayan, D., and Dr. Lavanya R., “Retrieval driven classification for mammographic masses”, in Proceedings of the 2019 IEEE International Conference on Communication and Signal Processing, ICCSP 2019, 2019, pp. 725-729.[Abstract]


Accurate diagnosis is pivotal for successful treatment for breast cancer. High chances of survival are possible, if malignancy is detected at an early stage. Mammography is the most efficient and widely accepted modality for screening breast cancer. In this paper we propose a decision support system based on image retrieval which retrieves similar pathology based mammographic images to serve the physician in the diagnosis of breast cancer. The work explores how to use the retrieved similar cases as references to improve the classification performance. The rationale is that by incorporating the closeness information for decision making improves classifier performance rather than making decision from whole database. Experiments were carried out on DDSM database utilizing 4300 images of breast cancer. The results demonstrated the effectiveness of proposed system and show the vitality for clinical applications. © 2019 IEEE.

More »»

2019

S. Dheepadharshani, Anandh, S., Bhavinaya, K. B., and Dr. Lavanya R., “Multivariate time-series classification for automated fault detection in satellite power systems”, in Proceedings of the 2019 IEEE International Conference on Communication and Signal Processing, ICCSP 2019, 2019, pp. 814-817.[Abstract]


Data driven techniques have become prominent in big data analysis. In this paper multisensory time series data is analyzed using Kernel Principal Component Analysis (KPCA) and Multilayer Perceptron (MLP) for fault detection in satellite power system. NASA's ADAPT dataset is used for validating the proposed algorithm. The proposed work differs from conventional time series techniques by considering each instantaneous measurement of multiple sensors as a data sample. This varied form of data augmentation results in improved fault diagnosis performance when compared to the conventional time series analysis. © 2019 IEEE.

More »»

2018

V. Sharathappriyaa, Gautham, S., and Dr. Lavanya R., “Auto-encoder Based Automated Epilepsy Diagnosis”, in 2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI), Bangalore, India, 2018.[Abstract]


This work proposes the design of an automated system based on Auto-encoders to detect epilepsy using Electroencephalogram (EEG) signal. Harmonic Wavelet Packet Transform (HWPT) and Fractal Analysis are employed to extract the initial feature vector. HWPT is used to capture spectral information using non-overlapping octave bands, thereby avoiding recursive calculations. Fractal dimension (FD) based on Katz technique is performed on windowed segments of the signal to capture spatial information of the EEG signal. Features comprising HWPT and FD are fed to an Auto-encoder, which is used to compress the high dimensional feature vector to a more discriminative and lower dimensional feature vector, for efficient classification. Finally, a Softmax classifier is used for the binary classification problem of discriminating epilepsy signals from normal cases.

More »»

2018

V. Dharani and Dr. Lavanya R., “Improved Microaneurysm Detection in Fundus Images for Diagnosis of Diabetic Retinopathy”, in Advances in Signal Processing and Intelligent Recognition Systems, Cham, 2018.[Abstract]


This paper addresses the development of a computer-aided diagnosis (CAD) system for early detection of diabetic retinopathy (DR), a sight threatening disease, using digital fundus photography (DFP). More specifically, the proposed CAD system is intended for detection of microaneurysms (MA) which are the earliest indicators of DR; CAD systems for MA detection involve two stages: coarse segmentation for candidate MA detection and fine segmentation for false positive elimination. The system addresses the common challenges in candidate MA detection, which includes detection of subtle MAs and MAs close to each other and those close to blood vessels which leads to low sensitivity. The system employs four major steps. The first step involves preprocessing of the fundus images, which comprises of shade correction, denoising and intensity normalization. The second step aims at the segmentation of candidate MAs using bottom hat transform, thresholding and hit or miss transformation. The use of modified morphological contrast enhancement and multiple structuring elements (SEs) in the hit or miss transform has improved the detection rate of MAs. The proposed method has been validated using a set of 20 fundus images from the DIARETDB1 database. The Free Response Operating Characteristics (FROC) curve demonstrates that many MAs that are otherwise missed out are detected by the proposed CAD system.

More »»

2018

S. N. Kumar, Dinesh, D., Siddharth, T., Ramkumar, S., Nikhill, S., and Dr. Lavanya R., “Selection of Features Using Particle Swarm Optimization for Microaneurysm Detection in Fundus Images”, in Proceedings of the 2017 International Conference on Wireless Communications, Signal Processing and Networking, WiSPNET 2017, 2018, vol. 2018-January, pp. 140-144.[Abstract]


Diabetic Retinopathy (DR) is a predominant factor leading to blindness for decades. The main reason for loss of vision is the damage to vasculature in the retina. The earliest signs of DR are microaneurysms which show up as tiny red spots on the retina. Early detection of this indicator helps the ophthalmologists to detect DR, which helps in preventing blindness. In this work, a series of image processing algorithms including pre-processing, and coarse segmentation using mathematical morphology are employed to detect initial candidates for microaneurysms. This is followed by fine segmentation for which a set of optimal features is estimated using Particle Swarm Optimization (PSO). Classification performance of naïve Bayes and Support Vector Machine (SVM) are compared. The set of 19 features selected using PSO-SVM has led to an accuracy of 99.92% compared to the PSO-naïve Bayes with 22 features and an accuracy of 93.31%. The proposed system could be employed for accurate and fast detection of microaneurysms and thereby would considerably lower the workload and time spent by ophthalmologists.

More »»

2017

S. Simon, Dr. Lavanya R., and Vijayan, D., “PSO Based Density Classifier for Mammograms”, in The 16th International Conference on Biomedical Engineering, Singapore, 2017.[Abstract]


Breast cancer is the major cancer diagnosed in both, developed and developing countries. Early detection and treatment of breast cancer is necessary to moderate the associated fatality rates. Mammography is the widely accepted modality for screening breast cancer. Breast density is considered one of the major risk indicators for Breast cancer. Nevertheless, low contrast and subtle nature of abnormalities reduces the sensitivity of mammograms, especially in dense breast. In this paper we present an automatic method for breast density classification based on two level cascaded support vector machine (SVM) classifiers. Particle Swarm Optimization (PSO) has been employed for SVM parameter optimization that resulted in a low set up time for building the system. The proposed system was tested on mini-MIAS database, and an overall classification accuracy of 82% was achieved. Also the system could prompt the radiologists on high-risk cases, thereby gaining more attention from them for diagnosis of such cases.

More »»

2016

D. Balachandran and Dr. Lavanya R., “Mass Characterization in Mammograms using an Optimal Ensemble Classifier”, in 2016 IEEE Region 10 Conference (TENCON), 2016.[Abstract]


Mass is the most common indicator in mammograms, especially in the early stages of breast cancer. Due to subtle nature of the masses, there is a considerable overlap between the malignant and benign mass characteristics. In this work, a Computer Aided Diagnosis (CADx) system that employs ensemble classifier has been proposed to improve mass characterization. Genetic algorithm (GA), an optimization technique, was employed to select the optimal ensemble. Multicollinearity among classifiers has to be resolved while forming the ensemble. Combining the classifiers that are highly correlated will not guarantee an improved performance when compared to individual classifiers. Variation Inflation Factor (VIF) analysis is incorporated in this work for detecting multicollinearity among classifiers.

More »»

2015

Dr. Lavanya R., Nagarajan, N., and M. Devi, N., “Computer-aided Diagnosis of Breast Cancer by Hybrid Fusion of Ultrasound and Mammogram Features”, in Artificial Intelligence and Evolutionary Algorithms in Engineering Systems, New Delhi, 2015, vol. 325, pp. 403–409.[Abstract]


Ultrasound images are increasingly being used as an important adjunct to X-ray mammograms for diagnosis of breast cancer. In this paper, a computer-aided diagnosis system that utilizes a hybrid fusion strategy based on canonical correlation analysis (CCA) is proposed for discriminating benign and malignant masses. The system combines information from three different sources, i.e., ultrasound and two views of mammogram, namely, mediolateral oblique (MLO) and craniocaudal (CC) views. CCA is employed on ultrasound-MLO and ultrasound-CC feature pairs to explore the hidden correlations between ultrasound and mammographic view. The two pairs of canonical variates are fused at the feature level and given as input to support vector machine (SVM) classifiers. Finally, decisions of the two classifiers are fused. Results show that the proposed system outperforms unimodal systems and state-of-the-art fusion strategies.

More »»