SACM - United Kingdom
Permanent URI for this collectionhttps://drepo.sdl.edu.sa/handle/20.500.14154/9667
Browse
2 results
Search Results
Item Restricted Explainable AI Approach for detecting Generative AI Imagery(Aston University, 2024-09-29) Alghamdi, Sara; Barns, ChloeThe rapid advancement of Artificial Intelligence (AI) and machine learning, particularly deep learning models such as Convolutional Neural Networks (CNNs), has revolutionized image classification across diverse fields, including healthcare, autonomous vehicles, and digital forensics. However, the proliferation of AI-generated images, commonly referred to as deepfakes, has introduced significant ethical, societal, and security challenges. Deepfakes leverage AI to create highly realistic yet synthetic media, complicating the ability to differentiate between authentic and manipulated content. This has heightened the need for robust tools capable of accurately detecting and classifying such media to combat the risks of misinformation, fraud, and erosion of public trust. Traditional models, while effective in classification, often lack transparency in their decision-making processes, limiting stakeholder trust. To address this limitation, this study explores the integration of Explainable AI (XAI) techniques, such as SHAP (SHapley Additive exPlanations), with CNNs to enhance interpretability and trust in model predictions. By employing CNNs for high-accuracy classification and XAI methods for feature-level explanations, the research aims to contribute to digital forensics and content moderation, offering both technical reliability and transparency. This study highlights the critical need for trustworthy AI systems in the fight against manipulated media, providing a framework that balances efficacy, transparency, and ethical considerations.48 0Item Restricted Using phages to Treat Urinary Tract Infections: Predicting phage susceptibility using bacterial genome and MALDI-TOF data(Saudi Digital Library, 2023-09-07) Alghamdi, Sara; Clokie, MarthaAMR, and MDR present substantial challenges for individuals and have also become a global concern. This has resulted in these infections, gaining increasing attention. Bacteriophages have become the go-to in dealing with bacteria resistance and decreasing the number of mortalities. For this project, instruments like the bacteria genome sequence and MALDI-TOF data will be used to gain predictions of phage susceptibility and serotypes. A group of 16 phages was collected in the lab with at least one manufactured host. This project obtained 70 clinical strains from the Bristol University Hospital. Two techniques were employed in this project: spot test and plague essay. Both methods seek to measure the concentration of the bacteriophage and evaluate the virus’ effectiveness. The serotypes included in this study are ST131, ST69, ST73, and ST95. The project concluded, the gene pattern of ST131 responds weakly to most phages and all concentrations. ST73_35 was the most sensitive in 108=114, 106=87 104=51. Some strains were more sensitive than the others ST73 and ST95 this is may allow to make predictions in terms of family species or sequencing. On the other hand, ST131 was the most resistance strain and then ST69, this would make more challenging to work for phage predictor. It can be noted that JK08 performing the best with strains. On the other hand, the worst phage UP15 1×104 shows more resistance to strain. In the event that further studies with Whole Genome Sequencing and MALDI-TOF were conducted to confirm this mechanism, so that would be able to predict some genes responsible for susceptibility or resistance. The outcome of this project will demonstrate a platform of a broad collection of E. coli strains that might finds the correlation of sequence types with MALDI-TOF and WGS data so we can make predictions on host range.34 0