10 results on '"Syed Adil Hussain Shah"'
Search Results
2. Automated facial characterization and image retrieval by convolutional neural networks
- Author
-
Syed Taimoor Hussain Shah, Syed Adil Hussain Shah, Shahzad Ahmad Qureshi, Angelo Di Terlizzi, and Marco Agostino Deriu
- Subjects
oriented gradient-based algorithm ,convolutional neural networks ,GoogLeNet ,AlexNet ,KNN ,computer vision ,Electronic computers. Computer science ,QA75.5-76.95 - Abstract
IntroductionDeveloping efficient methods to infer relations among different faces consisting of numerous expressions or on the same face at different times (e.g., disease progression) is an open issue in imaging related research. In this study, we present a novel method for facial feature extraction, characterization, and identification based on classical computer vision coupled with deep learning and, more specifically, convolutional neural networks.MethodsWe describe the hybrid face characterization system named FRetrAIval (FRAI), which is a hybrid of the GoogleNet and the AlexNet Neural Network (NN) models. Images analyzed by the FRAI network are preprocessed by computer vision techniques such as the oriented gradient-based algorithm that can extract only the face region from any kind of picture. The Aligned Face dataset (AFD) was used to train and test the FRAI solution for extracting image features. The Labeled Faces in the Wild (LFW) holdout dataset has been used for external validation.Results and discussionOverall, in comparison to previous techniques, our methodology has shown much better results on k-Nearest Neighbors (KNN) by yielding the maximum precision, recall, F1, and F2 score values (92.00, 92.66, 92.33, and 92.52%, respectively) for AFD and (95.00% for each variable) for LFW dataset, which were used as training and testing datasets. The FRAI model may be potentially used in healthcare and criminology as well as many other applications where it is important to quickly identify face features such as fingerprint for a specific identification target.
- Published
- 2023
- Full Text
- View/download PDF
3. Hyperglycemia-associated Alzheimer’s-like symptoms and other behavioral effects attenuated by Plumeria obtusa L. Extract in alloxan-induced diabetic rats
- Author
-
Sumeera Naz, Imran Imran, Muhammad Asad Farooq, Syed Adil Hussain Shah, Iqra Ajmal, Zartash Zahra, Aqsa Aslam, Muhammad Irfan Sarwar, Jaffer Shah, and Ambreen Aleem
- Subjects
Alzheimer ,anxiolytic ,anti-depressant ,learning ,memory ,anti-diabetic ,Therapeutics. Pharmacology ,RM1-950 - Abstract
Diabetes mellitus is a chronic metabolic complaint with numerous short- and long-term complications that harm a person’s physical and psychological health. Plumeria obtusa L. is a traditional medicine used in the treatment of diabetes to reduce complications related to behavior. Plumeria is a genus with antipsychotic activities. The objective of this study was to examine the effects of a methanolic extract of Plumeria obtusa L. in the attenuation of diabetes, on symptoms of Alzheimer disease, and on other associated behavioral aspects. A single dose of alloxan was administered to an experimental group of rats to induce development of diabetes (150 mg/kg, intraperitoneal) and the rats were then administered selected doses of methanolic extract of Plumeria obtusa L. (Po.Cr) or glibenclamide (0.6 mg/kg) for 45 consecutive days. Behavioral effects were evaluated using three validated assays of anxiety-related behavior: the open field test, the light and dark test, and the elevated plus maze. Anti-depressant effects of Plumeria obtusa L. were evaluated using the forced swim test (FST) and memory and learning were assessed using the Morris water maze (MWM) task. Po.Cr was also evaluated for phytochemicals using total phenolic content (TPC), total flavonoid content (TFC), and high-performance liquid chromatography assays, and antioxidant capability was assessed through assays of DPPH radical scavenging, total oxidation capacity, and total reducing capacity. In the alloxan-induced model of diabetes, the administration of Po.Cr and glibenclamide for 45 days produced a marked decrease (p < 0.001) in hyperglycemia compared to control animals. Po.Cr treatment also resulted in improvement in indicators, such as body weight and lipid profile (p < 0.05), as well as restoration of normal levels of alanine transaminase (ALT) (p < 0.001), a biomarker of liver function. Diabetic rats presented more Alzheimer-like symptoms, with greater impairment of memory and learning, and increased anxiety and depression compared to non-diabetic normal rats, whereas treated diabetic rats showed significant improvements in memory and behavioral outcomes. These results demonstrate that Po.Cr reversed alloxan-induced hyperglycemia and ameliorated Alzheimer-related behavioral changes, which supports additional study and assessment of conventional use of the plant to treat diabetes and associated behavioral complications.
- Published
- 2022
- Full Text
- View/download PDF
4. A Novel Hybrid Learning System Using Modified Breaking Ties Algorithm and Multinomial Logistic Regression for Classification and Segmentation of Hyperspectral Images
- Author
-
Syed Taimoor Hussain Shah, Shahzad Ahmad Qureshi, Aziz ul Rehman, Syed Adil Hussain Shah, Arslan Amjad, Adil Aslam Mir, Amal Alqahtani, David A. Bradley, Mayeen Uddin Khandaker, Mohammad Rashed Iqbal Faruque, and Muhammad Rafique
- Subjects
active learning ,hyperspectral imaging system ,multinomial logistic regression ,segmentation framework ,machine learning ,Markov random fields ,Technology ,Engineering (General). Civil engineering (General) ,TA1-2040 ,Biology (General) ,QH301-705.5 ,Physics ,QC1-999 ,Chemistry ,QD1-999 - Abstract
A new methodology, the hybrid learning system (HLS), based upon semi-supervised learning is proposed. HLS categorizes hyperspectral images into segmented regions with discriminative features using reduced training size. The technique utilizes the modified breaking ties (MBT) algorithm for active learning and unsupervised learning-based regressors, viz. multinomial logistic regression, for hyperspectral image categorization. The probabilities estimated by multinomial logistic regression for each sample helps towards improved segregation. The high dimensionality leads to a curse of dimensionality, which ultimately deteriorates the performance of remote sensing data classification, and the problem aggravates further if labeled training samples are limited. Many studies have tried to address the problem and have employed different methodologies for remote sensing data classification, such as kernelized methods, because of insensitiveness towards the utilization of large dataset information and active learning (AL) approaches (breaking ties as a representative) to choose only prominent samples for training data. The HLS methodology proposed in the current study is a combination of supervised and unsupervised training with generalized composite kernels generating posterior class probabilities for classification. In order to retrieve the best segmentation labels, we employed Markov random fields, which make use of prior labels from the output of the multinomial logistic regression. The comparison of HLS was carried out with known methodologies, using benchmark hyperspectral imaging (HI) datasets, namely “Indian Pines” and “Pavia University”. Findings of this study show that the HLS yields the overall accuracy of {99.93% and 99.98%}Indian Pines and {99.14% and 99.42%}Pavia University for classification and segmentation, respectively.
- Published
- 2021
- Full Text
- View/download PDF
5. Intelligent system development to monitor the neonatal behaviour: A review.
- Author
-
Syed Adil Hussain Shah, Angelo di Terlizzi, and Marco Agostino Deriu
- Published
- 2022
6. Classification and Segmentation Models for Hyperspectral Imaging - An Overview.
- Author
-
Syed Taimoor Hussain Shah, Shahzad Ahmad Qureshi, Aziz ul Rehman, Syed Adil Hussain Shah, and Jamal Hussain
- Published
- 2020
- Full Text
- View/download PDF
7. Investigations of plausible pharmacodynamics supporting the antispasmodic, bronchodilator, and antidiarrheal activities of Berberis lycium Royle. Via in silico, in vitro, and in vivo studies
- Author
-
Syed Adil Hussain Shah and Ambreen Aleem
- Subjects
Pharmacology ,Drug Discovery - Abstract
Berberis lycium Royle, a member of the Berberidaceae family, is a high-value medicinal plant with a documented history of usage in traditional medicine and has demonstrated significant therapeutic results among local populations throughout the globe. It is used traditionally in many parts of Pakistan to treat diarrhea, abdominal spasms, coughs, and chest problems.To investigate the antispasmodic, bronchodilator, and antidiarrheal effects of B. lycium and its possible underlying mechanisms through in silico, in vitro, and in vivo studies.LC ESI-MS/MS analysis was used to identify bioactive components within the hydromethanolic extract of B. lycium. In silico studies, including network pharmacology and molecular docking, were utilized to investigate the antispasmodic and bronchodilator properties of the extract's bioactive components. In vitro pharmacological studies were conducted using isolated rabbit jejunum, trachea, urinary bladder, and rat ileum preparations. In vivo antidiarrheal activities were conducted in mice, including castor oil-induced diarrhea, intestinal transit, and castor oil-induced enteropooling.The LC ESI-MS/MS analysis of the hydromethanolic extract of B. lycium identified 38 bioactive compounds. Network pharmacology study demonstrated that the mechanism of BLR for the treatment of diarrhea might involve IL1B, TLR4, PIK3R1, TNF, PTPRC, IL2, PIK3CD, and ABCB1, whereas, for respiratory ailments, it may involve PIK3CG, TRPV1, STAT3, ICAM1, ACE, PTGER2, PTGS2, TNF, MMP9, NOS2, IL2, CCR5, HRH1, and VDR. Molecular docking research revealed that chlorogenic acid, epigallocatechin, isorhamnetin, quinic acid, gallic acid, camptothecin, formononetin-7-O-glucoside, velutin, caffeic acid, and (S)-luteanine exhibited a higher docking score than dicyclomine with validated proteins of smooth muscle contractions such as CACB2_HUMAN, ACM3_HUMAN, MYLK_HUMAN, and PLCG1_HUMAN. In vitro investigations demonstrated that Blr.Cr, Blr.EtOAc, and Blr.Aq relaxed spontaneously contracting jejunum preparations; carbachol (1 μM)-induced and KThe dual blocking mechanism of muscarinic receptors and Ca
- Published
- 2023
8. Classification and Segmentation Models for Hyperspectral Imaging - An Overview
- Author
-
Syed Taimoor Hussain Shah, Aziz ul Rehman, Syed Adil Hussain Shah, Shahzad Ahmad Qureshi, and Jamal Hussain
- Subjects
Computer science ,Active learning (machine learning) ,business.industry ,Hyperspectral imaging ,Pattern recognition ,Segmentation ,Artificial intelligence ,Belief propagation ,business ,Feature learning ,Multinomial logistic regression - Abstract
An advancement in Hyperspectral Imaging (HI) technology is creating important attraction among the researchers to develop better classification techniques. This technology is well known for its high spatial and spectral information due to which the discrimination of materials is much more accurate and efficient. The useful information is extracted in Hyperspectral Imaging technology after applying it in agriculture, biomedical, and disaster management studies. A review comparison has been carried out for air borne images using hyperspectral acquisition hardware for classification as well as segmentation purpose. Numerous approaches that have been focused for implementation namely semi-supervised technique used for hyperspectral imaging using active learning and multinomial logistic regression, Generalized Composite Kernels (GCKs) classification framework, classification of spectral-spatial based data on loopy belief propagation (LBP), multiple feature learning of HI classification, and semi-supervised GCKs with classification accuracy on AVIRIS dataset (59.97%, 92.89%, 81.45%, 75.84%, and 95.50) and segmentation accuracies using α-expansion method as (73.27%, 93.57%, 92.86%, 91.73% and 98.31), respectively.
- Published
- 2021
9. A Novel Hybrid Learning System Using Modified Breaking Ties Algorithm and Multinomial Logistic Regression for Classification and Segmentation of Hyperspectral Images
- Author
-
Mayeen Uddin Khandaker, Arslan Amjad, Syed Adil Hussain Shah, Syed Taimoor Hussain Shah, Amal Alqahtani, Aziz ul Rehman, Shahzad Ahmad Qureshi, Adil Aslam Mir, Muhammad Rafique, D.A. Bradley, and Mohammad Rashed Iqbal Faruque
- Subjects
semi-supervised learning ,Technology ,QH301-705.5 ,Computer science ,Active learning (machine learning) ,QC1-999 ,Data classification ,Semi-supervised learning ,Markov random fields ,Discriminative model ,active learning ,General Materials Science ,Segmentation ,Biology (General) ,QD1-999 ,Instrumentation ,hyperspectral imaging system ,Multinomial logistic regression ,Fluid Flow and Transfer Processes ,Physics ,Process Chemistry and Technology ,General Engineering ,Hyperspectral imaging ,Engineering (General). Civil engineering (General) ,Computer Science Applications ,Chemistry ,ComputingMethodologies_PATTERNRECOGNITION ,machine learning ,Unsupervised learning ,TA1-2040 ,multinomial logistic regression ,Algorithm ,segmentation framework - Abstract
A new methodology, the hybrid learning system (HLS), based upon semi-supervised learning is proposed. HLS categorizes hyperspectral images into segmented regions with discriminative features using reduced training size. The technique utilizes the modified breaking ties (MBT) algorithm for active learning and unsupervised learning-based regressors, viz. multinomial logistic regression, for hyperspectral image categorization. The probabilities estimated by multinomial logistic regression for each sample helps towards improved segregation. The high dimensionality leads to a curse of dimensionality, which ultimately deteriorates the performance of remote sensing data classification, and the problem aggravates further if labeled training samples are limited. Many studies have tried to address the problem and have employed different methodologies for remote sensing data classification, such as kernelized methods, because of insensitiveness towards the utilization of large dataset information and active learning (AL) approaches (breaking ties as a representative) to choose only prominent samples for training data. The HLS methodology proposed in the current study is a combination of supervised and unsupervised training with generalized composite kernels generating posterior class probabilities for classification. In order to retrieve the best segmentation labels, we employed Markov random fields, which make use of prior labels from the output of the multinomial logistic regression. The comparison of HLS was carried out with known methodologies, using benchmark hyperspectral imaging (HI) datasets, namely “Indian Pines” and “Pavia University”. Findings of this study show that the HLS yields the overall accuracy of {99.93% and 99.98%}Indian Pines and {99.14% and 99.42%}Pavia University for classification and segmentation, respectively.
- Published
- 2021
10. Novel Classification Technique for Hyperspectral Imaging using Multinomial Logistic Regression and Morphological Profiles with Composite Kernels
- Author
-
Shahzad Ahmad Qureshi, Syed Taimoor Hussain Shah, Syed Gibran Javed, Abdul Majid, and Syed Adil Hussain Shah
- Subjects
Flowchart ,010504 meteorology & atmospheric sciences ,Computer science ,business.industry ,0211 other engineering and technologies ,Hyperspectral imaging ,Pattern recognition ,02 engineering and technology ,Image segmentation ,01 natural sciences ,law.invention ,Statistical classification ,Data point ,law ,Segmentation ,Artificial intelligence ,Developing regions ,business ,021101 geological & geomatics engineering ,0105 earth and related environmental sciences ,Multinomial logistic regression - Abstract
Hyperspectral imaging (HI) is getting much more attention among researchers in different fields like agriculture, defense, medical, and geographical surveys. In this work, we have proposed a novel automated system for the classification and segmentation of landscapes using hyperspectral images. The proposed semi-supervised based approach has improved the extraction of spatial characteristics of the scene that has employed an extended multi-attribute profile (EMAP) by stacking of several attributes. The unlabeled data points located near the classifier boundaries are selected on the basis of entropy related to the corresponding class labels. In the next segmentation phase, MLR probabilities are computed against the output of classifier. Finally, maximum-a-posteriori segmentation is carried out on the multilevel logistic prior labels. The simulated results have obtained classification accuracy of 95.50% by comparing predicted labels with original ones. The segmentation accuracy, after developing regions on the output of classification, is 98.31%. A performance comparison of the proposed approach with several approaches has also been carried out.
- Published
- 2019
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.