Papers by Ghanashyama Prabhu
F1000Research, Jun 12, 2023
In this current era of communications and networking, The Internet of things plays the main role ... more In this current era of communications and networking, The Internet of things plays the main role in the making of smart communication and networking. In this article, we have focused on the literature survey on wireless sensor networks which are energy efficient. Various standard protocols are reviewed along with some enhanced protocols which makes the network energy efficient. The comparison of the standard and enhanced protocols with respect to various applications in wireless sensor networks is thoroughly done in this article. The outcomes of the enhanced protocols are also briefly discussed. For easier analysis to future researchers, a comparative table which lists the enhanced protocols which are compared with standard counterparts along with the factors for energy efficiency of the protocols. This article also comments on the issues and challenges of the protocols which can be further analyzed for making the wireless sensor network more energy efficient. Keywords Energy efficient (EE), Internet of Things (IoT), Wireless sensor networks (WSNs) This article is included in the Manipal Academy of Higher Education gateway. Open Peer Review Approval Status AWAITING PEER REVIEW Any reports and responses or comments on the article can be found at the end of the article.
Osteoporosis Detection Using Deep Learning on X-Ray images of Human Spine
Journal of Physics: Conference Series
Osteoporosis is a condition, which results in weakness of bones, and its effects grow detrimental... more Osteoporosis is a condition, which results in weakness of bones, and its effects grow detrimental as the person ages. Currently, there is no proven cure for Osteoporosis, so early detection is the most prudent way to slow down its effects. The work implements classification models based on CNN and other deep learning models to detect and classify osteoporosis using human spine X-ray image datasets made publicly available by kaggle. This paper presents various deep learning models incorporating transfer learning on hybrid combinations to train and test for better results. Result conveyed through accuracy, Confusion Matrix, Precision, Recall, F1 score, AUC, and observes considerable improvement when compared to previous works.
Comparison of Classification Models Based on Deep learning on COVID-19 Chest X-Rays
Journal of Physics: Conference Series
COVID -19, is a deadly, dangerous and contagious disease caused by the novel corona virus. It is ... more COVID -19, is a deadly, dangerous and contagious disease caused by the novel corona virus. It is very important to detect COVID-19 infection accurately as quickly as possible to avoid the spreading. Deep learning methods can significantly improve the efficiency and accuracy of reading Chest X-Rays (CXRs). The existing Deep learning models with further fine tune provide cost effective, rapid, and better classification results. This paper tries to deploy well studied AI tools with modification on X-ray images to classify COVID 19. This research performs five experiments to classify COVID-19 CXRs from Normal and Viral Pneumonia CXRs using Convolutional Neural Networks (CNN). Four experiments were performed on state-of-the-art pre-trained models using transfer learning and one experiment was performed using a CNN designed from scratch. Dataset used for the experiments consists of chest X-Ray images from the Kaggle dataset and other publicly accessible sources. The data was split into th...
Performance of Network Coding
With network coding the Intermediate node forwards packets obtained by linear combination of the ... more With network coding the Intermediate node forwards packets obtained by linear combination of the previously received incoming packets. These packets can be referred to as coded packets. Network Coding offers many advantages like bandwidth efficiency, energy efficiency and Throughput. The error performance of Network coding at an intermediate node is studied over an additive white guassian channel with hard decoding.
An activity recognition framework based on machine learning to automatically recognize LME exerci... more An activity recognition framework based on machine learning to automatically recognize LME exercises and to count the repetitions using a wrist-worn inertial sensor is proposed. Fourteen binary classifiers are trained using optimized SVM models [1, 3] to recognize individual LME exercises, achieving overall accuracy of more than 98%.
Rehabilitation from cardiovascular disease (CVD) usually requires lifestyle changes, especially a... more Rehabilitation from cardiovascular disease (CVD) usually requires lifestyle changes, especially an increase in exercise and physical activity. However, uptake and adherence to exercise is low for community-based programmes. We propose a mobile application that allows users to choose the type of exercise and compete it at a convenient time in the comfort of their own home. Grounded in a behaviour change framework, the application provides feedback and encouragement to continue exercising and to improve on previous results. The application also utilizes wearable wireless technologies in order to provide highly personalized feedback. The application can accurately detect if a specific exercise is being done, and count the associated number of repetitions utilizing accelerometer or gyroscope signals Machine learning models are employed to recognize individual local muscular endurance (LME) exercises, achieving overall accuracy of more than 98%. This technology allows providing a near re...
A Deep Learning Model for Exercise-Based Rehabilitation Using Multi-channel Time-Series Data from a Single Wearable Sensor
Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering
Digital Image Watermarking: Algorithms, Attacks, Benchmarking and Applications
This paper gives an introduction to digital image watermarking. Algorithms, distortions, attacks ... more This paper gives an introduction to digital image watermarking. Algorithms, distortions, attacks and benchmarking are described in some detail. Finally, we discuss some of the scenarios where watermarking is being already used as well as other potential applications.
In this paper, we propose an algorithmic approach for a motion analysis framework to automaticall... more In this paper, we propose an algorithmic approach for a motion analysis framework to automatically recognize local muscular endurance (LME) exercises and to count their repetitions using a wrist-worn inertial sensor. LME exercises are prescribed for cardiovascular disease rehabilitation. As a technical solution, we propose activity recognition based on machine learning. We developed an algorithm to automatically segment the captured data from all participants. Relevant time and frequency domain features were extracted using a sliding window technique. Principal component analysis (PCA) was applied for dimensionality reduction of the extracted features. We trained 15 binary classifiers using support vector machine (SVM) to recognize individual LME exercises, achieving overall accuracy of more than 98%. We applied grid search technique to obtain the optimal SVM hyperplane parameters. The learning curves (mean ± stdev) for each model is investigated to verify that the models were not o...
Feature Extraction using PCA on Wearable Multimodal Wireless Sensor Data in Human Activity Recognition (HAR)
The feature extraction and classification is an important stage in human activity recognition (HA... more The feature extraction and classification is an important stage in human activity recognition (HAR). In this paper, we discuss human activity classification using wearable multimodal wireless sensors in healthcare, especially in individuals with cardiovascular disease (CVD). We use majorly principle component analysis (PCA) on data collected using accelerometers and gyroscope data from subjects for 15 Local Muscular Endurance (LME) exercises. Well-known time domain and frequency-domain signal characteristic features are extracted and classification of best features is carried out with PCA. Supervised learning algorithms based with support vector machines (SVM) are used further for recognition of movement patterns.
Proceedings of the 2nd International Workshop on Multimedia for Personal Health and Health Care, 2017
The third phase of the recovery from cardiovascular disease (CVD) is an exercise-based rehabilita... more The third phase of the recovery from cardiovascular disease (CVD) is an exercise-based rehabilitation programme. However, adherence to an exercise regime is typically not maintained by the patient for a variety of reasons such as lack of time, financial constraints, etc. In order to facilitate patients to perform their exercises from the comfort of their home and at their own convenience, we have developed a mobile application, termed MedFit. It provides access to a tailored suite of exercises along with easy to understand guidance from audio and video instructions. Two types of wearable sensors are utilized to provide motivational feedback. Fitbit, a commercially available activity and fitness tracker, is used to provide in-depth feedback for self-monitoring over longer periods of time (e.g. day, week, month), whereas the Shimmer wireless sensing platform provides the data for near real-time feedback on the quality of the exercises performed. MedFit is a simple and intuitive mobile...
The third phase of the recovery from cardiovascular disease (CVD) is an exercise-based rehabilita... more The third phase of the recovery from cardiovascular disease (CVD) is an exercise-based rehabilitation programme. However, adherence to an exercise regime is typically not maintained by the patient for a variety of reasons such as lack of time, financial constraints, etc. In order to facilitate patients to perform their exercises from the comfort of their home and at their own convenience, we have developed a mobile application, termed MedFit. It provides access to a tailored suite of exercises along with easy to understand guidance from audio and video instructions. Two types of wearable sensors are utilized to allow motivational feedback to be provided to the user for self monitoring and to provide near real-time feedback. Fitbit, a commercially available activity and fitness tracker, is used to provide in-depth feedback for self-monitoring over longer periods of time (e.g. day, week, month), whereas the Shimmer wireless sensing platform provides the data for near real-time feedbac...
Recognition and Repetition Counting for LME Exercises in Exercise-Based CVD Rehabilitation: A Comparative Study Using Artificial Intelligence Models
Exercise-based cardiac rehabilitation requires patients to perform a set of certain prescribed ex... more Exercise-based cardiac rehabilitation requires patients to perform a set of certain prescribed exercises a specific number of times. Local muscular endurance (LME) exercises are an important part of the rehabilitation program. Automatic exercise recognition and repetition counting, from wearable sensor data is an important technology to enable patients to perform exercises independently in remote settings, e.g. their own home. In this paper we first report on a comparison of traditional approaches to exercise recognition and repetition counting, corresponding to supervised machine learning and peak detection from inertial sensing signals respectively, with more recent machine learning approaches, specifically Convolutional Neural Networks (CNNs). We investigated two different types of CNN: one using the AlexNet architecture, the other using time-series array. We found that the performance of CNN based approaches were better than the traditional approaches. For exercise recognition t...
Sensors
Exercise-based cardiac rehabilitation requires patients to perform a set of certain prescribed ex... more Exercise-based cardiac rehabilitation requires patients to perform a set of certain prescribed exercises a specific number of times. Local muscular endurance exercises are an important part of the rehabilitation program. Automatic exercise recognition and repetition counting, from wearable sensor data, is an important technology to enable patients to perform exercises independently in remote settings, e.g., their own home. In this paper, we first report on a comparison of traditional approaches to exercise recognition and repetition counting (supervised ML and peak detection) with Convolutional Neural Networks (CNNs). We investigated CNN models based on the AlexNet architecture and found that the performance was better than the traditional approaches, for exercise recognition (overall F1-score of 97.18%) and repetition counting (±1 error among 90% observed sets). To the best of our knowledge, our approach of using a single CNN method for both recognition and repetition counting is n...
International Journal of Future Computer and Communication, 2013
Network coding is a new concept of combining the information at intermediate nodes, which can inc... more Network coding is a new concept of combining the information at intermediate nodes, which can increase the throughput of the network. In this paper we apply network coding techniques to Wireless LAN (WLAN). Network coding is applied at the access node of a 7-cluster WLAN and the probability of network coding feasibility is arrived at. Further queuing theory is applied to analytically obtain the WLAN throughput both in the presence as well as absence of physical channel induced transmission errors. From the analysis it is observed that application of network coding to WLAN increases the throughput by 5 to 7% as compared to traditional store and forward routing.
Uploads
Papers by Ghanashyama Prabhu