A5047076835- Phonocardiography and Auscultation Techniques
- Medical Image Segmentation Techniques
- AI in cancer detection
- Advanced Neural Network Applications
- Digital Imaging for Blood Diseases
- ECG Monitoring and Analysis
- Brain Tumor Detection and Classification
- Cardiac Imaging and Diagnostics
- Advanced X-ray and CT Imaging
- Genital Health and Disease
- Music and Audio Processing
- Diverse Musicological Studies
- Heavy Metal Exposure and Toxicity
- Heavy metals in environment
- Balance, Gait, and Falls Prevention
- Healthcare and Environmental Waste Management
- Digital and Cyber Forensics
- Hair Growth and Disorders
- Advanced Malware Detection Techniques
- Corporate Governance and Management
- Heavy Metals in Plants
- Arsenic contamination and mitigation
- Video Analysis and Summarization
- Gait Recognition and Analysis
- Mercury impact and mitigation studies
Rochester Institute of Technology
2024
National Yang Ming Chiao Tung University
2023-2024
International Islamic University, Islamabad
2023
University of Engineering and Technology Lahore
2021-2023
Pain Clinic of India
2022
University of Engineering and Technology Peshawar
2019-2021
Objective.Cardiovascular diseases (CVDs) are a main cause of deaths all over the world. This research focuses on computer-aided analysis phonocardiogram (PCG) signals based deep learning that can enable improved and timely detection heart abnormalities. The two widely used publicly available PCG datasets from PhysioNet/CinC (2016) PASCAL (2011) challenges. significantly different in terms tools for data acquisition, clinical protocols, digital storages signal qualities, making it challenging...
Abstract Purpose This study focused on two incineration plants in Taiwan, plant Z and CD, utilizing solid recovered fuel (SRF). Methods We assessed emissions of nitrogen oxides (NOx), polycyclic aromatic hydrocarbons (PAHs), polychlorinated dibenzo- p -dioxins (PCDD/Fs), heavy metals (HMs) e.g., (Hg, As, Pb, Cr, Be, Cd) across various environmental media. Results found that co-firing SRF with coal achieved significant reductions: NOx by 25%, PCDD/Fs 70%, Hg 29%. Additionally, particulate...
In this paper, we propose a novel computer vision based system for detecting falls on the basis of monocular camera that can be applied assisting independent living an older adult alone at home. From recorded video data, mask regional convolutional neural network (mask R-CNN) is first to detect and extract human silhouettes. The low-dimensionality representative features extracted silhouettes are then by (CNN), which classification different postures (e.g., sit, bend, stand lie) detection...
Haemoglobin (Hb) electrophoresis is a method of blood testing used to detect thalassaemia. However, the interpretation result test itself complex task. Expert haematologists, specifically in developing countries, are relatively few number and usually overburdened. To assist them with their workload, this paper we present novel for automated assessment thalassaemia using Hb images. Moreover, study compile large image dataset, consisting 103 strips containing 524 images clear consensus on...
In recent decades, heavy metals (HMs) have emerged as a global health concern. Unfortunately, in Pakistan, there is general lack of awareness regarding the potential risks associated with HMs pollution among automobile workers. Herein, we investigated concentration such lead (Pb), cadmium (Cd), and chromium (Cr) workers who were occupationally exposed Mingora City, Khyber Pakhtunkhwa, Pakistan. Three different groups, i.e., battery recyclers, spray painters, mechanics studied detail. A total...
To extract liver from medical images is a challenging task due to similar intensity values of with adjacent organs, various contrast levels, noise associated and irregular shape liver. address these issues, it important preprocess the images, i.e., computerized tomography (CT) magnetic resonance imaging (MRI) data prior analysis quantification. This paper investigates impact permutation preprocessing techniques for CT on automated segmentation using deep learning, U-Net architecture. The...
Android is one of the leading operating systems for smart phones in terms market share and usage. Unfortunately, it also an appealing target attackers to compromise its security through malicious applications. To tackle this issue, domain experts researchers are trying different techniques stop such attacks. All attempts securing platform somewhat successful. However, existing detection have severe shortcomings, including cumbersome process feature engineering. Designing representative...
Autonomous vehicles and human drivers are prone to line-of-sight limitations. Road-side mounted 3D sensors like LiDARs can augment a vehicle's on-board perception. However, this entails fusing frames at low latency high accuracy. vehicle captured from different viewpoints. This adversely affects alignment accuracy be computationally expensive. To end, VRF optimizes for both by decoupling the process into indirect direct alignments. First, indirectly aligns aligning them common reference...
Improper execution of modern code‐designed structures in many developing countries have resulted significant deficient building stock; low strength concrete, reduced reinforcement, inappropriate detailing beam‐column members, and lack lateral ties joint panels. Observations based on earthquake‐induced damages experimental studies conducted such buildings revealed vulnerability joints bare moment‐resisting frame structures. Shake table tests were selected three 1 : 4 reduced‐scale three‐story...
This paper presents an improved classification technique for automated of phonocardiogram (PCG) signals. In the light presented literature study, a number representative multidomain time and frequency features are suggested heart signal analysis with comparatively large imbalanced dataset. Machine learning algorithms such as support vector machines (SVM), k-nearest neighbor (KNN), Decision Tree (DT) TreeBagger (TB) tested sound (HS) classification. For performance evaluation metrics as,...
Heart diseases are serious and must be detected early using an auscultation examination. To explore diagnose heart problems, several signal processing machine learning approaches used. From a Phonocardiogram (PCG) signal, the sound (HS) can categorized into normal abnormal. This paper presents improvedcomputer-aidedtechniquefor classification of HS long short-term memory (LSTM)deployed withdifferent time frequency domain features, i.e., discrete wavelet transform (DWT) Mel-frequency cepstral...
The fundamental working of background subtraction is to identify the moving region by taking pixel-wise difference current frame from previous one. proposed study presents comparison and implementation different techniques i.e., frame-difference method, mixture Gaussian model 2 (MOG2) k-nearest neighbor (KNN) for subtraction. For all three techniques, prior segmentation, modeling then features extraction steps are performed. It investigated that on same dataset, technique outperforms both...
This paper investigates the performance of machine learning techniques for automated assessment lymphocytes. A total four algorithms, i.e., support vector (SVM), deep java (DL4J), multi-layer perceptron (MLP) and K* are applied to lymphocytes dataset. To ensure robustness, all algorithms evaluated with both preprocessed data without prior pre-processing. With pre-processed data, MLP outperforms other achieve an accuracy 98.64%. SVM, DL4J accuracies 97.97%, 96.62% 97.29%, respectively.
Over the past few decades there was a steady decline in number of sexually transmitted diseases (STDs). However with rapid emergence human immunodeficiency virus (HIV) which has now become pandemic these STDs are once again focus. But classical presentation is no longer rule; and such scenario clinical suspicion to be confirmed by specific investigations for correct identification thus management diseases. This article tries enlist that can done case STD attending clinic. The list means...