A5072928978- Radiomics and Machine Learning in Medical Imaging
- COVID-19 diagnosis using AI
- Machine Learning in Materials Science
- Lung Cancer Diagnosis and Treatment
- Topic Modeling
- Advanced Text Analysis Techniques
- Computational and Text Analysis Methods
- Advanced Chemical Physics Studies
- Advanced DC-DC Converters
- AI in cancer detection
- Advanced Neural Network Applications
- Superconducting Materials and Applications
- Multilevel Inverters and Converters
- Video Surveillance and Tracking Methods
- Electromagnetic Launch and Propulsion Technology
- Advanced Control Systems Optimization
- Electrostatic Discharge in Electronics
- Synthesis and Properties of Aromatic Compounds
- Bioinformatics and Genomic Networks
- Sentiment Analysis and Opinion Mining
- Ethics in Clinical Research
- Stock Market Forecasting Methods
- VLSI and FPGA Design Techniques
- Artificial Intelligence in Healthcare and Education
- Ethics and Social Impacts of AI
United Kingdom Atomic Energy Authority
2024
ETH Zurich
2022-2024
Bhabha Atomic Research Center Hospital
2024
Bhabha Atomic Research Centre
2024
Indian Institute of Technology Kharagpur
2012-2022
Adopting effective techniques to automatically detect and identify small drones is a very compelling need for number of different stakeholders in both the public private sectors. This work presents three original approaches that competed grand challenge on “Drone vs. Bird” detection problem. The goal one or more appearing at some time point video sequences where birds other distractor objects may be also present, together with motion background foreground. Algorithms should raise an alarm...
Success of the UK’s Spherical Tokamak for Energy Production (STEP) programme requires a robust plasma control system. This system has to guide from initiation burning phase, maintain it there, produce desired fusion power duration and then terminate safely. be done in challenging environment with limited sensors without overloading plasma-facing components. The parameters operational regime STEP prototype will very different tokamaks, which are presently operation. During burn,...
Particle Swarm Optimization (PSO), a population based technique for stochastic search in multidimensional space, has so far been employed successfully solving variety of optimization problems including many multifaceted problems, where other popular methods like steepest descent, gradient conjugate gradient, Newton method, etc. do not give satisfactory results. Herein, we propose modified PSO algorithm unbiased global minima by integrating with density functional theory which turns out to be...
Abstract In 2019 the UK launched Spherical Tokamak for Energy Production (STEP) programme to design and build a prototype electricity producing nuclear fusion power plant, aiming start operation around 2040. The plant should lay foundation development of commercial plants. is based on spherical tokamak principle, which opens route high pressure, steady state, operation. While facilitating state operation, introduces some specific plasma control challenges: (i) All current during burn phase...
Rheumatoid arthritis (RA) is a chronic inflammatory and long-term autoimmune disease that can lead to joint bone erosion. This patients’ disability if not treated in timely manner. Early detection of RA settings such as primary care (as the first contact with patients) have an important role on treatment disease. We aim develop web-based Decision Support System (DSS) provide proper assistance for providers early patients. Using Sparse Fuzzy Cognitive Maps, well quantum-learning algorithm, we...
This study presents a particle swarm optimisation (PSO)-based approach to optimise node count and path length of the binary decision diagram (BDD) representation Boolean function. The is achieved by identifying good ordering input variables affects structure resulting BDD. Both longest shared BDDs using identified are found be much superior existing results. improvements more prominent for larger benchmarks. PSO parameters have been tuned suitably explore large search space within reasonable...
Chest radiographs are primarily employed for the screening of pulmonary and cardio-/thoracic conditions. Being undertaken at primary healthcare centers, they require presence an on-premise reporting Radiologist, which is a challenge in low middle income countries. This has inspired development machine learning based automation process. While recent efforts demonstrate performance benchmark using ensemble deep convolutional neural networks (CNN), our systematic search over multiple standard...
Monitoring a fleet of robots requires stable long-term tracking with re-identification, which is yet an unsolved challenge in many scenarios. One application this the analysis autonomous robotic soccer games at RoboCup. Tracking these handling identically looking players, strong occlusions, and non-professional video recordings, but also offers state information estimated by robots. In order to make effective use coming from robot sensors, we propose robust identification pipeline. It fuses...
The number of increased social media users has led to a lot people misusing these platforms spread offensive content and use hate speech. Manual tracking the vast amount posts is impractical so it necessary devise automated methods identify them quickly. Large language models are trained on data they also make contextual embeddings. We fine-tune large help in our task. quite unbalanced; we used modified cross-entropy loss tackle issue. observed that using model which fine-tuned hindi corpora...
The paper describes the work that team submitted to FinCausal 2020 Shared Task. This is associated with first sub-task of identifying causality in sentences. various models used experiments tried obtain a latent space representation for each Linear regression was performed on these representations classify whether sentence causal or not. have shown BERT (Large) best, giving F1 score 0.958, task detecting sentences financial texts and reports. class imbalance dealt modified loss function give...
Chest radiographs are primarily employed for the screening of pulmonary and cardio-/thoracic conditions. Being undertaken at primary healthcare centers, they require presence an on-premise reporting Radiologist, which is a challenge in low middle income countries. This has inspired development machine learning based automation process. While recent efforts demonstrate performance benchmark using ensemble deep convolutional neural networks (CNN), our systematic search over multiple standard...
Legal proceedings take plenty of time and also cost a lot. The lawyers have to do lot work in order identify the different sections prior cases statutes. paper tries solve first tasks AILA2021 (Artificial Intelligence for Assistance) that will be held FIRE2021 (Forum Information Retrieval Evaluation). task is semantically segment document into assigned one 7 predefined labels or "rhetorical roles." uses BERT obtain sentence embeddings from sentence, then linear classifier used output final...
The paper describes the work that has been submitted to 5th workshop on Challenges and Applications of Automated Extraction socio-political events from text (CASE 2022). is associated with Subtask 1 Shared Task 3 aims detect causality in protest news corpus. authors used different large language models customized cross-entropy loss functions exploit annotation information. experiments showed bert-based-uncased refined outperformed others, achieving a F1 score 0.8501 Causal News Corpus dataset.
The paper describes the work that has been submitted to 5th workshop on Challenges and Applications of Automated Extraction socio-political events from text (CASE 2022). is associated with Subtask 1 Shared Task 3 aims detect causality in protest news corpus. authors used different large language models customized cross-entropy loss functions exploit annotation information. experiments showed bert-based-uncased refined outperformed others, achieving a F1 score 0.8501 Causal News Corpus dataset.