High-speed download cost-effective version control compression on cloud
Traditional techniques for image retrieval are not supported for the ever-expansive image database. These downsides can be removed by utilizing contents of the image for image retrieval. This sort of image retrieval can be accomplished with the help of cross batch redundancy detection (CBRD). When the cross batch redundancy detection (CBRD) is combinedly work with Bandwidth Energy Efficient sharing (BEES) has very precise result other than any commonly known feature detector or traditional method that are usually utilized in vision programming. In this paper we mainly focus in texture, colour ,shape and resolution of an image matching with superior accuracy and we also review some of the futuristic tool developed on CBRD
Published by: Hari Krishnan M., Jayaraman N., Mahendran P., Dr. N. Kanya
Author: Hari Krishnan M.
Paper ID: V7I3-2117
Paper Status: published
Published: June 25, 2021
An authentication & data sharing of the medical report using blockchain security
The health care services industry is always showing payments. Blockchain 2.0 focused on decentralizing the entire market and is employed to transform assets through smart signs of change and supporting new advancements and advances. One of the predominant requirements in today’s health care systems is to protect the patient's medical report against potential attackers. Hence, it is basic to have secure information that can just approve people can get to the patient's medical report. So, we have proposed Blockchain technology as a disbursed approach to grant security in accessing the medical report of a patient. It’s composed of three phases Authentication, Encryption, and Data Retrieval using BlockChain technology. For authentication – Quantum Cryptography, for Encryption – AES, and for Data Retrieval– SHA algorithms are used to resist the frequent attacks. This proposed framework may likewise ensure the protection of the patients and moreover keeps up the security and trustworthiness of the health care system.
Published by: Komal Suresh Raut
Author: Komal Suresh Raut
Paper ID: V7I3-2116
Paper Status: published
Published: June 25, 2021
Network intrusion detection system
With the fast development of PC utilization and PC network, the security of the PC framework has turned out to be very significant. Businesses are looking for consistently new sorts of assaults. As the danger turns into a genuine chapter year by year, interruption discovery advancements are essential for organization and PC security. An assortment of interruption recognition approaches be available to determine this severe condition; in any case, the fundamental problem is execution. It is imperative to increment the discovery rates and lessen bogus alert rates in the space of interruption recognition. To recognize the interruption, different methodologies have been created and proposed over the most recent decade. This paper effectively compares the accuracy of different classification algorithms, like algorithms like Support Vector Machine (SVM), Naive Bayes, KNN, Decision Tree[15]. This study aims to perform a comparative analysis of these different machine learning algorithms on datasets available to predict which model best suits intrusion detection.
Published by: Kj Bhoomika, Impana N., Priya D.
Author: Kj Bhoomika
Paper ID: V7I3-2102
Paper Status: published
Published: June 25, 2021
A review on Deep Learning for content-based image retrieval
Learning effective feature representations and similarity measures are crucial to the retrieval performance of a content-based image retrieval (CBIR) system. Despite extensive research efforts for decades, it remains one of the most challenging open problems that considerably hinders the successes of real-world CBIR systems. The key challenge has been attributed to the well-known “semantic gap” issue that exists between low-level image pixels captured by machines and high-level semantic concepts perceived by human. Among various techniques, machine learning has been actively investigated as a possible direction to bridge the semantic gap in the long term. Inspired by recent successes of deep learning techniques for computer vision and other applications, in this paper, we attempt to address an open problem: if deep learning is a hope for bridging the semantic gap in CBIR and how much improvements in CBIR tasks can be achieved by exploring the state-of-the-art deep learning techniques for learning feature representations and similarity measures. Specifically, we investigate a framework of deep learning with application to CBIR tasks with an extensive set of empirical studies by examining a state-of-the-art deep learning method (Convolution Neural Networks) for CBIR tasks under varied settings. From our empirical studies, we find some encouraging results and summarize some important insights for future research.
Published by: Nikhil Nandkishor Kela, S. R. Tayde
Author: Nikhil Nandkishor Kela
Paper ID: V7I3-2050
Paper Status: published
Published: June 25, 2021
A study on analyzing fake news through Neural Network Models
Fake news is defined as a made-up story with an intention to deceive or to mislead. In this paper we present the solution to the task of fake news detection by using Deep Learning architectures. Gartner research [1] predicts that “By 2022, most people in mature economies will consume more false information than true information”. The exponential increase in production and distribution of inaccurate news presents an immediate need for automatically tagging and detecting such twisted news articles. However, automated detection of fake news is a hard task to accomplish as it requires the model to understand nuances in natural language. Moreover, majority of the existing fake news detection models treat the problem at hand as a binary classification task, which limits model’s ability to understand how related or unrelated the reported news is when compared to the real news. To address these gaps, we present neural network architecture to accurately predict the stance between a given pair of headline and article body. Our model outperforms existing model architectures by 2.5% and we are able to achieve an accuracy of 94.21% on test data.
Published by: Ashwini Durgadas Kalpande, Sachin A. Vyawhare
Author: Ashwini Durgadas Kalpande
Paper ID: V7I3-2041
Paper Status: published
Published: June 25, 2021
The PICC-related complication in Hemato-Oncological patients for the administration of chemotherapeutic drugs
Background Majority of malignant diseases require therapy with chemotherapeutic agents for medium to long term time duration. Duration, feasibility and ease of placement, comfort, and rate of complications in patients are of utmost importance in choosing the type of catheter placement. In light of this peripherally inserted central venous catheters (PICC) have added advantage over conventional central venous catheters. The aim of this study was to evaluate the rate of infection related to PICC, the duration and outcome of PICC in oncological patients. Patients and Methods: A longitudinal study was conducted to look at the PICC (Peripherally Inserted Central Catheter)-related complication rates which occurred in inpatient and outpatient settings on patients who had a PICC line inserted for the administration of chemotherapeutic drugs, between 2015 to 2019, a period of five years. A total of 865 patients with PICC lines were analyzed. Pertinent patient demographics, as well as catheter-related factors, were collected. The data was analyzed by using excel and the association between duration and complication is compared using the chi-square test. p<0.05 is considered as significant.to identify catheter-related complications and the outcome of the PICC line in relation to the removal of the line. The Retrospective data analyzed so waiver of consent is taken on the day of PICC line insertion and the institutional ethical clearance has been taken for this study. As the data is taken retrospectively from the patient records with anonymity being maintained, no actual consent is required in this study. Approval from the hospital ethical committee got. Also during PICC insertion, informed consent is taken from all subjects. Results:s The PICCs placed for 865 patients, each between the duration of 3 months to 6 months was analyzed. The most suitable vein for the insertion was the basilic vein (85%). Conclusions: Our study suggests, PICC is an excellent option for various diagnostic and therapeutic interventions and offers clinicians and nurses a safe and effective option for central access. The data of this study shows PICC to be more cost-effective in terms of longer duration of use and have lower complication rates than the conventional CICCs and hence has to be promoted more in patients for chemotherapy drug administration.
Published by: Sampa Mandal, Leema Chandravadhana, Punitha Rani Singh, Raju C.
Author: Sampa Mandal
Paper ID: V7I3-1931
Paper Status: published
Published: June 25, 2021
