- EEG and Brain-Computer Interfaces
- Video Surveillance and Tracking Methods
- Anomaly Detection Techniques and Applications
- Human Pose and Action Recognition
- Medical Imaging Techniques and Applications
- Energy Efficient Wireless Sensor Networks
- Advanced Neural Network Applications
- Protein Structure and Dynamics
- Image Enhancement Techniques
- Advanced Image and Video Retrieval Techniques
- Advanced Vision and Imaging
- Video Analysis and Summarization
- Machine Learning in Bioinformatics
- Radiomics and Machine Learning in Medical Imaging
- Fire Detection and Safety Systems
- Generative Adversarial Networks and Image Synthesis
- Non-Invasive Vital Sign Monitoring
- IoT and Edge/Fog Computing
- Remote-Sensing Image Classification
- Heart Rate Variability and Autonomic Control
- Epilepsy research and treatment
- Advanced MRI Techniques and Applications
- Neuroscience and Neural Engineering
- Time Series Analysis and Forecasting
- Enzyme Structure and Function
Tata Consultancy Services (India)
2016-2025
Embedded Systems (United States)
2018-2022
Centre for Artificial Intelligence and Robotics
2020
The University of Melbourne
2007-2016
Increasing population density in urban centers demands adequate provision of services and infrastructure to meet the needs city inhabitants, encompassing residents, workers, visitors. The utilization information communications technologies achieve this objective presents an opportunity for development smart cities, where management citizens are given access a wealth real-time about environment upon which base decisions, actions, future planning. This paper framework realization cities...
Poincaré plot is one of the important techniques used for visually representing heart rate variability. It valuable due to its ability display nonlinear aspects data sequence. However, problem lies in capturing temporal information quantitatively. The standard descriptors quantifying (SD1, SD2) measure gross variability time series data. Determination advanced methods properties pose a significant challenge. In this paper, we propose novel descriptor "Complex Correlation Measure (CCM)"...
Obstructive sleep apnea or hypopnea causes a pause reduction in airflow with continuous breathing effort. The aim of this study is to identify individual and events from normal using wavelet-based features 5-s ECG signals (sampling rate = 250 Hz) estimate the surrogate index (AI)/hypopnea (HI) (AHI). Total 82,535 epochs (each duration) during sleep, 1638 689 events, 3151 1862 were collected 17 patients training set. Two-staged feedforward neural network model was trained...
The emerging Internet of Things (IoT) that effectively integrates cyber-physical space to create smart environments will undoubtedly have a plethora applications in the near future. Meanwhile, it is also key technological enabler cities, which provide great benefits our society. In this paper, four different IoT network architectures spanning various city are presented and their corresponding Quality Service (QoS) requirements defined. Furthermore, as beneficiary city, we responsibility...
Good water quality is essential for the health of our aquatic ecosystems. Continuous monitoring an important tool catchment management authorities, providing real-time data environmental protection and tracking pollution sources; however, continuous at high temporal spatial resolution remains prohibitively expensive. An affordable wireless system will enable cost-effective collection, assisting managers to maintain In this paper, a low-cost physiochemistry sensing presented. The results...
Real-time environment monitoring and analysis is an important research area of Internet Things (IoT). Understanding the behavior complex ecosystem requires detailed observations over a range different conditions. One such example in urban areas includes study tree canopy cover microclimate using heterogeneous sensor data. There are several challenges that need to be addressed, as obtaining reliable area, detecting unusual events from data, visualizing real-time way easily understandable by...
Infrastructure detection and monitoring is a difficult task. Due to the advances in unmanned vehicles image analytics, it possible decrease human effort achieve consistent results infrastructure assessments using aerial processing. Reliable integrity checking of power including conductor lines, pylons insulators diverse background most challenging task drone based automatic monitoring. Most techniques literature use first principle approach that tries represent as features interest. This...
Analyzing crowd events in a video is key to understanding the behavioral characteristics of people (humans). Detecting videos challenging because articulated human movements and occlusions. The aim this paper detect probabilistic framework for automatically interpreting visual behavior. In paper, event detection classification optical flow manifolds (OFMs) are addressed. A new algorithm walking running has been proposed, which uses vector lengths OFMs. Furthermore, merging splitting...
This paper presents a lip reading technique to classify the discrete utterances without evaluating acoustic signals. The reported analysis video data of motions by computing optical flow (OF). statistical properties vertical OF component were used form feature vectors for training support vector machines (SVM) classifier. impact variation in speed/velocity speaking on performance system was minimized removing zero energy frames and normalizing number interpolation. resulting is an efficient...
Summary Purpose A definite diagnosis of psychogenic nonepileptic seizures ( PNES ) usually requires in‐patient video–electroencephalography EEG monitoring. Previous research has shown that convulsive demonstrate a characteristic pattern rhythmic movement artifact on the . Herein we sought to examine potential for time‐frequency mapping data from movement‐recording device (accelerometer) worn wrist as diagnostic tool differentiate between epileptic and Methods Time‐frequency was performed...
The asymmetry in heart rate variability is a visibly obvious phenomenon the Poincaré plot of normal sinus rhythm. It shows unevenness distribution points above and below line identity, which indicates instantaneous changes beat to rate. major limitation existing definition that it considers only rather than pattern (increase/decrease). In this paper, novel proposed considering geometry 2D plot. Based on definition, traditional indices—Guzik's index (GI), Porta's (PI) Ehlers' (EI)—have been...
Epilepsy is one of the most common neurological disorders and patients suffer from unprovoked seizures. In contrast, psychogenic nonepileptic seizures (PNES) are another class that involuntary events not caused by abnormal electrical discharges but a manifestation psychological distress. The similarity these two types poses diagnostic challenges often leads in delayed diagnosis PNES. Further, PNES involves high-cost hospital admission monitoring using video-electroencephalogram machines. A...
The growth in pervasive network infrastructure called the Internet of Things (IoT) enables a wide range physical objects and environments to be monitored fine spatial temporal detail. detailed, dynamic data that are collected large quantities from sensor devices provide basis for variety applications. Automatic interpretation these evolving is required timely detection interesting events. This article develops exemplifies two new relatives visual assessment tendency (VAT) improved (iVAT)...
Abstract Background Stroke is one of the major causes morbidity and mortality. Its recovery treatment depends on close clinical monitoring by a clinician especially during first few hours after onset stroke. Patients who do not exhibit early motor post thrombolysis may benefit from more aggressive treatment. Method A novel approach for stroke using wireless accelerometer based activity system developed. It monitors measuring acceleration arms in three axes. In presented proof concept study,...
Ubiquitous sensing enabled by Wireless Sensor Network (WSN) technologies cuts across many areas of modern day living. This offers the ability to measure, infer and understand environmental indicators, from delicate ecologies natural resources urban environments. The proliferation these devices in a communicating-actuating network creates Internet Things (IoT), wherein, sensors actuators blend seamlessly with environment around us, information is shared platforms order develop common...
The use of drones in infrastructure monitoring aims at decreasing the human effort and achieving consistency. Accurate aerial image analysis is key block to achieve same. Reliable detection integrity checking power line conductors a diverse background are most challenging drone based automatic monitoring. Most techniques literature first principle approach that tries represent as features interest. This paper proposes machine learning for detection. A new deep architecture proposed with very...
Vision is one of the most advanced and important sensory input in humans. However, many people have vision problems due to birth defects, uncorrected errors, work nature, accidents, aging. The white cane guide dog are widely used means navigation for vision-impaired. With advancements technology, electronic devices been created using different sensors technologies help navigate blind. Electronic Travel Aids (ETAs) assist navigating a person by collecting information about environment...
Internet of Things (IoT) is denned as interconnection sensing and actuating devices providing the ability to share information across platforms through a unified framework, developing common operating picture for enabling innovative applications. As world urban population set cross unprecedented levels, adequate provision services infrastructure poses huge challenges. The emerging IoT that offers ubiquitous actuation can be utilized effectively managing environments. In this paper, new...