A5068701943- Network Security and Intrusion Detection
- Advanced Malware Detection Techniques
- Liver Disease Diagnosis and Treatment
- Liver Disease and Transplantation
- Anomaly Detection Techniques and Applications
- Spam and Phishing Detection
- Drug-Induced Hepatotoxicity and Protection
- Big Data and Business Intelligence
- Global Maternal and Child Health
- Data Mining Algorithms and Applications
- Electrolyte and hormonal disorders
- IoT and Edge/Fog Computing
- Clinical Nutrition and Gastroenterology
- Artificial Intelligence in Healthcare
- Blockchain Technology Applications and Security
- Child Nutrition and Water Access
- COVID-19 diagnosis using AI
- Software System Performance and Reliability
- Hyperglycemia and glycemic control in critically ill and hospitalized patients
- Information Technology Governance and Strategy
- Algorithms and Data Compression
- Acute Kidney Injury Research
- Advanced Neural Network Applications
- Demographic Trends and Gender Preferences
- Competitive and Knowledge Intelligence
All India Institute of Medical Sciences
1996-2025
Indian Council of Medical Research
2025
Government of India
2025
Vignana Jyothi Institute of Management
2015-2024
Sri Devaraj Urs Medical College
2024
Institute of Liver and Biliary Sciences
2015-2018
SRM Institute of Science and Technology
2016
SRM University
2006-2016
University Malaya Medical Centre
2007
University of Malaya
2007
Asian Indians are at high risk for the development of atherosclerosis and related complications, possibly initiated by higher body fat (BF). The present study attempted to establish appropriate cut-off levels BMI defining overweight, considering percentage BF in healthy northern India as standard. A total 123 volunteers (eighty-six males aged 18–75 years thirty-seven females 20–69 years) participated study. Clinical examination anthropometric measurements were performed, was calculated. 21·4...
Stability is considered the most important parameter before performing any melanocyte transplantation procedure in vitiligo; however, current criteria rely on history given by patients.This study was undertaken to determine clinical, biochemical and immunological factors determining stability of disease patients with generalized vitiligo facilitate better patient selection for understand mechanisms activity.Thirty-three < 10% body surface area involved were allocated three clinical groups:...
Abstract The problem of anomaly and attack detection in IoT environment is one the prime challenges domain internet things that requires an immediate concern. For example, anomalies attacks such as scan, malicious operation, denial service, spying, data type probing, wrong setup, control can lead to failure system. Datasets generated usually have missing values. presence values makes classifier unsuitable for classification task. This article introduces (a) a novel imputation technique (b)...
In this paper the major objective is to design and analyze suitability of Gaussian similarity measure for intrusion detection. The use as a distance find between any two data samples training set such DARPA Data Set, KDD Set. This metric when applying k-means algorithm. novelty approach making proposed function part algorithm so obtain disjoint clusters. followed by case study, which demonstrates process Intrusion Detection. has fixed upper lower bounds.
Intrusion Detection is one of major threats for organization. The approach intrusion detection using text processing has been research interests which gaining significant importance from researchers. In mining based detection, system calls serve as source and predicting possibility or attack. When an application runs, there might be several are initiated in the background. These form strong basis deciding factor detection. this paper, we mainly discuss by designing a distance measure...
In this paper the major objective is to design and analyze suitability of Gaussian similarity measure for intrusion detection. The use as a distance find between any two data samples training set such DARPA Data Set, KDD Set. This metric when applying k- means algorithm. novelty approach making proposed function part k-means algorithm so obtain disjoint clusters. followed by case study, which demonstrates process Intrusion Detection. has fixed upper lower bounds. satisfies all properties...
Earlier research focus towards anomaly detection has been using classifiers such as kNN, SVM and existing distance measures to perform classification. Traditionally IDSs (Intrusion systems) have developed by applying machine learning techniques adopted single mechanism. This is later extended developing Intrusion Detection Systems adopting multiple mechanisms. Such systems addressed better rates compared Systems. Dimensionality one more serious concern which affects the performance of...
Summary Background Severe alcoholic hepatitis patients have high mortality and limited response to corticosteroids. Microvesicles reflect cellular stress disease conditions. Aims To investigate whether microvesicles are associated with severity, steroid therapy inflammation in severe hepatitis. Methods originating from different cells were studied pre‐therapy 101 patients; (71 responder corticosteroid 30 nonresponders) 20 healthy controls. determined peripheral hepatic vein samples using...
Reducing the processing complexity is main challenge when dealing with intrusion detection systems. The reduced and efficiency increased if we can reduce number of dimensions so that only minimum retained. This work mainly targets on achieving dimensional reduction for using a novel membership function. function used to cluster features in iterative incremental manner obtains representation which retains original distribution process data. A case study discussed explore working proposed model.
Finding intrusion and anomalies in networks is a problem of wide research interest both from academia software industry. This work has three contributions. The first contribution dissimilarity measure for detection. also applied to achieve evolutionary clustering dimensionality reduction system calls. Earlier works used basic Gaussian membership function incrementally cluster by randomly assuming the initial deviation. aims at achieving defining expression choose, deviation eliminating need...
Objective: The objectives of the study are (i) to examine morbidity patterns, (ii) explore treatment-seeking behaviours, and (iii) analyse out-of-pocket expenditures (OOPEs) among Indigenous (referred as tribal in India) elderly Visakhapatnam district Andhra Pradesh, India. Methods: A cross-sectional survey 2,103 households was carried out. Data related socio-demographic, health, healthcare-seeking behaviours were collected for all household members. There 598 people aged 60 years above...
This work discusses the approach for intrusion detection and classification by devising a membership function, inspired from [43] is used in this to carry dimensionality reduction of processes present training set. The reduced process representation then perform prediction detecting intrusion. retains system call distribution same as initial representation.
Pre-natal diagnosis of intra-abdominal pregnancy is difficult. Ultrasound has been the frontline modality to date; however, it gives a diagnostic error 50–90% and its use disappointing. In recent years, MRI emerged as an appealing imaging modality. With good soft tissue contrast non-ionizing property, acts means definitive non-invasive assessment before surgical intervention when ultrasound inconclusive.
Intrusion Detection is one of the major threats for any organization size. The approach intrusion detection using text processing has been research interests among researchers working in area network and information security. In this detection, system calls serve as source mining predicting chance intrusion. When an application runs, there might be several which are initiated background. These form basis deciding factor detection. We perform extensive survey on techniques validate...
Intrusion detection is classified as NP-Hard in the literature even today. Also supervised learning also termed classification, when performed on high dimensional documents has problem from noise or outliers, which make text classification inaccurate and leads to reduced accuracy by classifiers. We discuss feature reduction methods we adopted achieve dimensionality reduction. In Feature Extraction process, are projected onto their corresponding low representation space through using...
The problem of clustering is NP-Complete. existing algorithm in literature the approximate algorithms, which cluster underlying data differently for different datasets. K-Means Clustering suitable frequency but not binary form. When an application runs several system calls are implicitly invoked background. Based on these we can predict normal or abnormal behavior applications. This be done by classification. In this paper tried to perform classification processes running into and states...
In this work, we design and propose an improved fuzzy membership function to detect anomalies intrusions. The objective of the present approach is achieve optimal transformation matrix which can improve classifier accuracies. aimed at mapping original process onto a new space, so that resultant representation free from noise data facilitates overall accuracy also individual class Experimental results show accuracies obtained using our better compared other approaches. particular U2R R2L are...
Computing environment in IoT (Internet of Things) is surrounded with huge amounts heterogeneous data fulfilling many services everyone's daily life. Since, communication process takes place using different devices such as smart phones, sensors, mobile devices, household embedded equipment etc. With the use these variety exchange open internet prone to vulnerabilities. The main cause for vulnerabilities weaknesses design software components and hardware components. Bridging communications...
Imputation of missing data values is an important pre-processing task for mining medical records. Application principles, techniques requires the dataset to be free from values. In this paper, there are two contributions. One imputation measure finding nearest optimal record and another algorithm imputing The proposed function extended by using our previous research in which a similarity temporal pattern named as ASTRA proposed. modified suitably serve measure.
According to Gartner, 60% of organisations are still unable make fruitful decisions due various factors like, data quality issues, lack careful consideration components involved, skill shortage. In this paper, we consider the software process, which is one key and show how a process model may be applied business intelligence (BI) by defining entire BI as two stage component internally involves other components. For purpose, have defined models for process. We also emphasise all...