A5055334891- Adversarial Robustness in Machine Learning
- Oral Health Pathology and Treatment
- Software Testing and Debugging Techniques
- Oral microbiology and periodontitis research
- Explainable Artificial Intelligence (XAI)
- Head and Neck Cancer Studies
- Anomaly Detection Techniques and Applications
- Oral and Maxillofacial Pathology
- Oral health in cancer treatment
- Software Reliability and Analysis Research
- Formal Methods in Verification
- Salivary Gland Tumors Diagnosis and Treatment
- Software Engineering Research
- Advanced Malware Detection Techniques
- Bone Tumor Diagnosis and Treatments
- Gut microbiota and health
- Cancer Diagnosis and Treatment
- Salivary Gland Disorders and Functions
- Machine Learning and Data Classification
- Machine Learning and Algorithms
- Laser Applications in Dentistry and Medicine
- Tumors and Oncological Cases
- Dental Radiography and Imaging
- Herpesvirus Infections and Treatments
- Dental Research and COVID-19
Ajman University
2022-2025
Ames Research Center
2018-2023
International Medical University
2020-2023
Saveetha University
2022-2023
KBR (United States)
2020-2022
University of Hong Kong
2016-2021
Milton Keynes Hospital
2020-2021
Prince Philip Dental Hospital
2018-2021
General Hospital Ernakulam
2021
Lourdes Hospital
2021
Zoonotic infections by avian influenza viruses occur at the human–poultry interface, but modes of transmission have not been fully investigated. We assessed potential for airborne and fomite live poultry markets in Guangzhou city Hong Kong Special Administrative Region (SAR), China, during 2014 2015. Viral genome infectious A H5N6, H7N9, H9N2 subtypes were detected predominantly from particles larger or equal to 1 μm diameter air sampled with cyclone-based bioaerosol samplers Guangzhou....
We present techniques for automatically inferring formal properties of feed-forward neural networks. observe that a significant part (if not all) the logic feed forward networks is captured in activation status (on or off) its neurons. propose to extract patterns based on neuron decisions as preconditions imply certain desirable output property e.g., prediction being class. input properties, encoding convex predicates space given and layer representing network hidden layers desired behavior....
Abstract Microbial dysbiosis has been implicated in the pathogenesis of oral cancer. We analyzed compositional and metabolic profile bacteriome three specific niches cancer patients along with controls using 16SrRNA sequencing (Illumina Miseq) DADA2 software. found major differences between control subjects. Bacterial communities associated tumor surface deep paired tissue differed significantly. Tumor surfaces carried elevated abundances taxa belonging to genera Porphyromonas,...
Deep Neural Networks (DNN) are increasingly used in a variety of applications, many them with substantial safety and security concerns. This paper introduces DeepCheck, new approach for validating DNNs based on core ideas from program analysis, specifically symbolic execution. The idea is to translate DNN into an imperative program, thereby enabling analysis assist validation. A basic translation however creates programs that very complex analyze. DeepCheck novel techniques lightweight...
Machine Learning is one of the fastest growing areas computer science, with far-reaching applications. In these paper, Supervised tasks most frequently carried out by Intelligent systems. Algorithms as well determines efficient classification algorithm based on data set, number enhances and features. The main goal this paper overall concept SVM KNN algorithms using Techniques.
Objective: While some oral carcinomas appear to arise de novo, others develop within long-standing conditions of the cavity that have malignant potential, now known as potentially disorders (OPMDs). The bacteriome associated with OPMD has been studied a lesser extent than cancer. To characterize association in detail we compared whole mouth fluid (WMF) patients leukoplakia, cancer and healthy controls.Methods: WMF from 20 leukoplakia patients, 31 23 controls were profiled using Illumina...