A5083737327- Complex Systems and Time Series Analysis
- Earthquake Detection and Analysis
- Image and Signal Denoising Methods
- Energy Load and Power Forecasting
- Stock Market Forecasting Methods
- Radiomics and Machine Learning in Medical Imaging
- Fractal and DNA sequence analysis
- Anomaly Detection Techniques and Applications
- COVID-19 diagnosis using AI
- EEG and Brain-Computer Interfaces
- AI in cancer detection
- Radioactivity and Radon Measurements
- Artificial Intelligence in Healthcare
- Complex Network Analysis Techniques
- Non-Invasive Vital Sign Monitoring
- Radioactive Decay and Measurement Techniques
- Digital Imaging for Blood Diseases
- Sleep and Wakefulness Research
- ECG Monitoring and Analysis
- Neural dynamics and brain function
- Climate variability and models
- Seismology and Earthquake Studies
- Machine Learning in Bioinformatics
- Traffic Prediction and Management Techniques
- Geochemistry and Geologic Mapping
University of Azad Jammu and Kashmir
2020-2023
Mirpur University of Science and Technology
2020
Abstract Background The large volume and suboptimal image quality of portable chest X-rays (CXRs) as a result the COVID-19 pandemic could post significant challenges for radiologists frontline physicians. Deep-learning artificial intelligent (AI) methods have potential to help improve diagnostic efficiency accuracy reading CXRs. Purpose study aimed at developing an AI imaging analysis tool classify lung infection based on Materials Public datasets ( N = 130), bacterial pneumonia 145),...
This study used machine learning classification of texture features from MRI breast tumor and peri-tumor at multiple treatment time points in conjunction with molecular subtypes to predict eventual pathological complete response (PCR) neoadjuvant chemotherapy.This employed a subset patients (N = 166) PCR data the I-SPY-1 TRIAL (2002-2006). cohort consisted stage 2 or 3 cancer that underwent anthracycline-cyclophosphamide taxane treatment. Magnetic resonance imaging (MRI) was acquired...
An efficient management and better scheduling by the power companies are of great significance for accurate electrical load forecasting. There exists a high level uncertainties in time series, which is challenging to make short-term forecast (STLF), medium-term (MTLF), long-term (LTLF). To extract local trends capture same patterns short, medium forecasting we proposed long memory (LSTM), Multilayer perceptron, convolutional neural network (CNN) learn relationship series. These models...
In this study, we aim to provide an efficient load prediction system projected for different local feeders predict the Medium- and Long-term Load Forecasting. This model improves future requirements expansions, equipment retailing or staff recruiting electric utility company. We aimed improve ahead forecasting by using hybrid approach optimizing parameters of our models. used Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), Multilayer perceptron (MLP) methods. Root Mean...
The incidence of congestive heart failure (CHF) is approximately 10 per 1000 for Americans over the age 65 years. dynamics CHF are highly complex, nonlinear, and temporal dynamics. Based on these characteristics, we extracted multimodal features from normal sinus rhythm (NSR) signals. We performed synthetic minority over-sampling technique (SMOTE) to increase number subjects balance our train data. classification between with original data SMOTE was using machine learning classifiers such as...
Non-linear processes evolving within earth crust and in atmosphere gives complex time series extracting meaningful physical information from such data is not easy without development practice of modern computational techniques. Detrended fluctuation analysis (DFA), detrended cross correlation (DCCA), are used to explore long-range correlations characterization correlated more than one non-stationary series. We present results DFA, DCCA techniques applied on radon, thoron, temperature...
In the whole world, approximately two billion people are using smartphones for their better life The smartphone's sensor data consist of high dimensions which increase complexity training model and degrade overall performance a classifier. We investigated analyzed different machine learning algorithms with various dimensionality reduction techniques on publicly available smartphone-based recognition human activities postural transitions dataset. highest accuracy was obtained at set 1 support...
Abstract Load Forecasting is an approach that implemented to foresee the future load demand projected on some physical parameters such as loading lines, temperature, losses, pressure, and weather conditions etc. This study specifically aimed optimize of deep convolutional neural networks (CNN) improve short-term forecasting (STLF) Medium-term (MTLF) i.e. one day, week, month three months. The models were tested based real-world case by conducting detailed experiments validate their stability...
Sleep is regulated autonomously with circadian behavior. The sleep disorders greatly impact major disturbances in patients suffered from Parkinson's disease (PD) and epilepsy affecting at night increased abnormality of muscles tones during NREM stage. aim this research to quantify the dynamics different pathologies by applying threshold- dependent symbolic entropy. threshold-dependent entropy applied distinguish healthy subjects such as narcolepsy, behavior disorder (RBD), disordered...
Abstract Continuous exposure to environmental radiation, whether it derives from natural or artificial sources, is thought pose a substantial risk public health. In addition the health effects associated with prolonged radiations, long-term measurements of these radiations can be used for variety beneficial purposes, such as forecasting impending earthquakes. Signal processing an important application purpose forecasting. Wavelets, being signal-processing tools, are helpful in many...
Deep learning artificial intelligent (AI) methods have potential to help improve diagnostic efficiency and accuracy for reading portable CXRs. Public datasets taken from SIRM Kaggle repositories comprised of COVID-19 (N = 130, 975), normal 138, 1525), bacterial pneumonia 145, 2521), non-COVID-19 viral 1342) respectively CXRs were analyzed. On the first dataset, we extracted 2048 features last pooling layer Residual Network 101 (ResNet101) which fed into selected classifiers. The three-class...
Epilepsy is a neurological disease caused by excessive neuron excitability in the brain. Feature-extracting strategies are necessary for EEG detection-dependent epilepsy to perform well. In this paper, we proposed Chi-square feature-ranking method on 22 multimodal features extracted from epileptic seizure based time and frequency domain characteristics, entropy-based complexity-based features, wavelet-based few statistical features. The were temporal variation spectral changes nonlinear...