A5012106402- Pulsars and Gravitational Waves Research
- Gamma-ray bursts and supernovae
- COVID-19 diagnosis using AI
- Astronomy and Astrophysical Research
- Astrophysical Phenomena and Observations
- Radiomics and Machine Learning in Medical Imaging
- Astronomical Observations and Instrumentation
- Model Reduction and Neural Networks
- Hip disorders and treatments
- Pelvic and Acetabular Injuries
- Lung Cancer Diagnosis and Treatment
- AI in cancer detection
- Anomaly Detection Techniques and Applications
- Energy Harvesting in Wireless Networks
- Scientific Computing and Data Management
- Radio Astronomy Observations and Technology
- Innovative Energy Harvesting Technologies
- Cloud Computing and Resource Management
- Hip and Femur Fractures
- Computational and Text Analysis Methods
- Hematological disorders and diagnostics
- Adversarial Robustness in Machine Learning
- Wireless Signal Modulation Classification
- Skin and Cellular Biology Research
- Galaxies: Formation, Evolution, Phenomena
University of Illinois Urbana-Champaign
2018-2025
National University of Sciences and Technology
2017-2025
Baba Raghav Das Medical College
2025
Sarojini Naidu Medical College
2023-2024
Lahore General Hospital
2023
Argonne National Laboratory
2022
National Center for Supercomputing Applications
2019-2022
Intel (United States)
2015-2020
Abstract Significant investments to upgrade and construct large-scale scientific facilities demand commensurate in R&D design algorithms computing approaches enable engineering breakthroughs the big data era. Innovative Artificial Intelligence (AI) applications have powered transformational solutions for challenges industry technology that now drive a multi-billion dollar industry, which play an ever increasing role shaping human social patterns. As AI continues evolve into paradigm...
We introduce the use of deep learning ensembles for real-time, gravitational wave detection spinning binary black hole mergers. This analysis consists training independent neural networks that simultaneously process strain data from multiple detectors. The output these is then combined and processed to identify significant noise triggers. have applied this methodology in O2 O3 finding clearly mergers open source available at Gravitational-Wave Open Science Center. also benchmarked...
The scale of ongoing and future electromagnetic surveys pose formidable challenges to classify astronomical objects. Pioneering efforts on this front include citizen science campaigns adopted by the Sloan Digital Sky Survey (SDSS). SDSS datasets have been recently used train neural network models galaxies in Dark Energy (DES) that overlap footprint both surveys. Herein, we demonstrate knowledge from deep learning algorithms, pre-trained with real-object images, can be transferred DES...
We introduce an ensemble of artificial intelligence models for gravitational wave detection that we trained in the Summit supercomputer using 32 nodes, equivalent to 192 NVIDIA V100 GPUs, within 2 h. Once fully trained, optimized these accelerated inference TensorRT. deployed our inference-optimized AI ThetaGPU at Argonne Leadership Computer Facility conduct distributed inference. Using entire supercomputer, consisting 20 nodes each which has 8 A100 Tensor Core GPUs and AMD Rome CPUs,...
Following its initial identification on December 31, 2019, COVID-19 quickly spread around the world as a pandemic claiming more than six million lives. An early diagnosis with appropriate intervention can help prevent deaths and serious illness distinguishing symptoms that set apart from pneumonia influenza frequently don't show up until after patient has already suffered significant damage. A chest X-ray (CXR), one of many imaging modalities are useful for detection most used, offers...
Human beings tend to incrementally learn from the rapidly changing environment without comprising or forgetting already learned representations. Although deep learning also has potential mimic such human behaviors some extent, it suffers catastrophic due which its performance on tasks drastically decreases while about newer knowledge. Many researchers have proposed promising solutions eliminate during knowledge distillation process. However, our best knowledge, there is no literature...
We present a deep-learning artificial intelligence model (AI) that is capable of learning and forecasting the late-inspiral, merger ringdown numerical relativity waveforms describe quasicircular, spinning, nonprecessing binary black hole mergers. used theNRHybSur3dq8 surrogate to produce train, validation test sets $\ensuremath{\ell}=|m|=2$ cover parameter space mergers with mass ratios $q\ensuremath{\le}8$ individual spins $|{s}_{{1,2}}^{z}|\ensuremath{\le}0.8$. These time range...
Melorheostosis is an uncommon benign tumour of mesodermal origin belonging to group sclerotic bone dysplasia. Leri and joanny in 1922 described it for the rst time as candle wax hyperostosis (1). derives from Greek word melos = limb rheos ow due classic radiological appearance 'owing hyperosteosis' resembling hardened that has dripped down side a candle. Its incidence low 0.9 per million population (2) No hereditary predisposition been seen. Both men women are equally affected (3). The...
The spin distribution of binary black hole mergers contains key information concerning the formation channels these objects, and astrophysical environments where they form, evolve coalesce. To quantify suitability deep learning to characterize signal manifold quasi-circular, spinning, non-precessing mergers, we introduce a modified version WaveNet trained with novel optimization scheme that incorporates general relativistic constraints properties holes. neural network model is trained,...
Technology nowadays is revolving around intelligent systems that mimic the learning capability of human brain. These learn, adapt, act and make decisions autonomously instead just executing predefined programmed instructions. This paper presents such computer-aided diagnostic (CAD) system, named as RID network, learns how to distinguish between normal Tuberculosis (TB) infected radiograph. Such can help reducing TB epidemic it a curable disease early diagnosis critical step towards its...
This report provides an overview of recent work that harnesses the Big Data Revolution and Large Scale Computing to address grand computational challenges in Multi-Messenger Astrophysics, with a particular emphasis on real-time discovery campaigns. Acknowledging transdisciplinary nature this document has been prepared by members physics, astronomy, computer science, data software cyberinfrastructure communities who attended NSF-, DOE- NVIDIA-funded "Deep Learning for Astrophysics: Real-time...
In the last couple of decades, Convolution Neural Network (CNN) emerged as most active field research. There are a number applications CNN, and its architectures used for improvement accuracy efficiency in various fields. this paper, we aim to use CNN order generate fusion visible thermal camera images detect persons present those reliable surveillance application. kinds image methods achieve multi-sensor, multi-modal, multi-focus multi-view fusion. Our proposed methodology includes...
We use artificial intelligence (AI) to learn and infer the physics of higher order gravitational wave modes quasi-circular, spinning, non precessing binary black hole mergers. trained AI models using 14 million waveforms, produced with surrogate model NRHybSur3dq8, that include up ℓ≤4 (5,5), except for (4,0) (4,1), describe binaries mass-ratios q≤8, individual spins s{1,2}z∈[−0.8,0.8], inclination angle θ∈[0,π]. Our probabilistic surrogates can accurately constrain mass-ratio, spins,...
Abstract Finding new ways to use artificial intelligence (AI) accelerate the analysis of gravitational wave data, and ensuring developed models are easily reusable promises unlock opportunities in multi-messenger astrophysics (MMA), enable wider use, rigorous validation, sharing by community. In this work, we demonstrate how connecting recently deployed DOE NSF-sponsored cyberinfrastructure allows for publish models, subsequently deploy these into applications using computing platforms...
The functional simulator Simics provides a co-simulation integration path with SystemC simulation environment to create Virtual Platforms. With increasing complexity of the models, this platform suffers from performance degradation due single threaded nature integrated Platform. In paper, we present multi-threaded solution that significantly improves over existing solution. two schedulers run independently, only communicating in thread safe manner through message interface. based logging and...
COVID-19 is a novel virus which originated from Wuhan, city in China. By March 2021, World Health Organization has confirmed the increased number of infections to over 117 million cases globally. In this scenario increasing Corona infected patients, most hospitals are lagging availability test kits. Owing lack precise automated toolkits, auxiliary diagnostic tools high demand. Therefore, it becomes necessary enforce AI-based automatic detection techniques. It can also address issue...
Abstract Significant investments to upgrade and construct large-scale scientific facilities demand commensurate in R\&D design algorithms computing approaches enable engineering breakthroughs the big data era. Innovative Artificial Intelligence (AI) applications have powered transformational solutions for challenges industry technology that now drive a multi-billion dollar industry, which play an ever increasing role shaping human social patterns. As AI continues evolve into paradigm...
Daily life of thousands individuals around the globe suffers due to physical or mental disability related limb movement. The quality for such can be made better by use assistive applications and systems. In scenario, mapping actions from movement a computer aided application lead way solution. Surface Electromyography (sEMG) presents non-invasive mechanism through which we translate signals classification in applications. this paper, propose machine learning based framework 4 actions. looks...