Muhammad Usman Akhtar

Optical coherence tomography (OCT) is a non-invasive imaging technique that allows for real-time visualization of interior tissue architecture with micrometer-level resolution. It is extensively used in eye care because it is good at looking at the shape of the retina and helping to diagnose vision-threatening disorders including Age-related Macular Degeneration and Diabetic Macular Edema. Automated OCT image processing is very important to clinical workflows since it make diagnosis easier and helps find diseases earlier. This chapter proposes a transformer-based classification framework for OCT image analysis that addresses robustness and accuracy. The method uses a Vision Transformer architecture that has already been trained and self-attention mechanisms. Using transfer learning and adversarial training together makes feature representations more stable and better able to deal with noise from enemies. This study improves OCT-based diagnostic systems and shows how well transformer models work for safe and accurate medical image classification.

The evolving technology in AI and distributed systems requires ethical concepts of how sensitive data can be verified without breach of privacy. Conventional AI systems present the following critical concerns: exposure of data, breach of privacy, and ethical issues concerning transparent but confidential computation. This chapter is a full-fledged cryptographic proof, Zero-Knowledge Proofs (ZKPs), which makes it possible to deploy AI ethically by verifying privacy. The framework is supported by mathematical underpinnings to enable model validation and training verification, as well as federated learning without the underlying datasets or parameters of the models. The chapter shows that ZKPs can be used to meet ethical AI without compromising privacy. It can be used in healthcare, finance, and voting systems where ethical concerns require verification and confidentiality. This chapter offers a new method of dealing with core ethical dilemmas in AI systems and safeguarding privacy and security in algorithmic decision-making exercises.

Infectious disease network dismantling is crucial for controlling disease spread in social and biological systems. One effective strategy is dismantling the network of disease transmission by identifying and targeting key individuals or behavioral hubs. Existing approaches often assume complete and accurate network data. This data is rarely available in real-world scenarios. This lack of data poses challenges in mitigating infectious disease spread. In, addition, these approaches overlook the effects of missing or hidden links, and consequently, many non-essential individuals are unnecessarily targeted, leading to suboptimal control of the epidemic spread. In this chapter, we introduce the Infectious Disease Network Dismantling Algorithm (IDND), which leverages missing link prediction and node centrality context to identify minimal critical nodes for efficient dismantling. Specifically, this work explores how to mitigate the spread of a pandemic by focusing interventions on a small subset of individuals, despite having partial knowledge of the network.

Advances in the healthcare sector need to secure efficient and decentralized systems to handle the large number of images and test reports. Traditional cloud-based healthcare systems pose crucial challenges, such as higher maintenance costs, significant storage demand, and, most importantly, privacy concerns. This chapter introduced the comprehensive patient-centric system for decentralized storage, secure medical data sharing, integrating blockchain technology and distributed file systems. The healthcare system leverages the IPFS and Ethereum blockchain to empowering scalable privacy protection of globally available storage of medical metadata. The framework enhances the accessibility and usability of incorporating a mobile offering for the patients, hospitals, and other stakeholders to interact with the decentralized system through the blockchain-integrated mobile application.This chapter provides an innovative approach to addressing the healthcare centralized data management limitations while ensuring patient privacy and data security.

Real world complex networks are indirect representation of complex systems. They grow over time. These networks are fragmented and raucous in practice. An important concern about complex network is link prediction. Link prediction aims to determine the possibility of probable edges. The link prediction demand is often spotted in social networks for recommending new friends, and, in recommender systems for recommending new items (movies, gadgets etc) based on earlier shopping history. In this work, we propose a new link prediction algorithm namely “Common Neighbor and Centrality based Parameterized Algorithm” (CCPA) to suggest the formation of new links in complex networks. Using AUC (Area Under the receiver operating characteristic Curve) as evaluation criterion, we perform an extensive experimental evaluation of our proposed algorithm on eight real world data sets, and against eight benchmark algorithms. The results validate the improved performance of our proposed algorithm.

As GPS signals are weak broad casted signal over wireless channels therefore they are vulnerable to in-band interference. Even a low power interference can spoof the GPS, which can cause destruction. Because of the above discussed reasons spoofing and anti-spoofing techniques are the emerging issues of the field of GPS. Generally, the spoofer try to generate or mimic the actual GPS signals to mislead a GPS receiver and spoofing works since the targeted receiver is not aware of it. Spoofing is becoming easier and cheaper because of advancement in SDR (software defined receiver) technology.

A successful development of UWSN in different areas is going on where a continuous and remote examining, detection and monitoring required under the water. Ample of research papers have been available that emphasis on different parameters such as energy consumption, throughput, stability period, end-to-end delay in UWSNs. In our research, we focus on the decrease of endto-delay, increase its stability period and throughput as well as load balancing of motes. In order to attain our objective we propose a scheme called DAE with major goal to implement the technique of anycasting. Finally we compared DAE with two existing protocols DBR and AMCTD. The performance of DAE is much efficient than these schemes in various parameters. Outcomes shows that the throughput and stability period of DAE are much better than DBR as well as AMCTD and these results are achieved with minimum delay.