The growth of IoT, edge and mobile Artificial Intelligence (AI) is supporting urban authorities exploit the wealth of information collected by Connected and Autonomous Vehicles (CAV), to drive the development of transformative intelligent transport applications for addressing smart city challenges. A critical challenge is timely and efficient road infrastructure maintenance. This paper proposes an intelligent hierarchical framework for road infrastructure maintenance that exploits the latest developments in 6G communication technologies, deep learning techniques, and mobile edge AI training approaches. The proposed framework abides with the stringent requirements of training efficient machine learning applications for CAV, and is able to exploit the vast numbers of CAVs forecasted to be present on future road networks. At the core of our framework is a novel Convolution Neural Networks (CNN) model which fuses imagery and sensory data to perform pothole detection. Experiments show the proposed model can achieve state of the art performance in comparison to existing approaches while being simple, cost-effective and computationally efficient to deploy. The proposed system can form part of a federated learning framework for facilitating large scale real-time road surface condition monitoring and support adaptive resource allocation for road infrastructure maintenance.

In this paper, we present a data analytics and visualization framework for health-shocks prediction based on large-scale health informatics dataset. The framework is developed using cloud computing services based on Amazon web services (AWS) integrated with geographical information systems (GIS) to facilitate big data capture, storage, index and visualization of data through smart devices for different stakeholders. In order to develop a predictive model for health-shocks, we have collected a unique data from 1000 households, in rural and remotely accessible regions of Pakistan, focusing on factors like health, social, economic, environment and accessibility to healthcare facilities. We have used the collected data to generate a predictive model of health-shock using a fuzzy rule summarization technique, which can provide stakeholders with interpretable linguistic rules to explain the causal factors affecting health-shocks. The evaluation of the proposed system in terms of the interpret-ability and accuracy of the generated data models for classifying health-shock shows promising results. The prediction accuracy of the fuzzy model based on a k-fold cross-validation of the data samples shows above 89% performance in predicting health-shocks based on the given factors. Cloud enabled framework for interactive data visualization, gathering, & analysis.Analysis of iron triangle under socioeconomic, cultural, and geographical norms.First publicly available data-set to understand health-shocks and its causes.

According to a recent Deloitte study, the COVID-19 pandemic continues to place a huge strain on the global health care sector. Covid-19 has also catalysed digital transformation across the sector for improving operational efficiencies. As a result, the amount of digitally stored patient data such as discharge letters, scan images, test results or free text entries by doctors has grown significantly. In 2020, 2314 exabytes of medical data was generated globally. This medical data does not conform to a generic structure and is mostly in the form of unstructured digitally generated or scanned paper documents stored as part of a patient’s medical reports. This unstructured data is digitised using Optical Character Recognition (OCR) process. A key challenge here is that the accuracy of the OCR process varies due to the inability of current OCR engines to correctly transcribe scanned or handwritten documents in which text may be skewed, obscured or illegible. This is compounded by the fact that processed text is comprised of specific medical terminologies that do not necessarily form part of general language lexicons. The proposed work uses a deep neural network based self-supervised pre-training technique: Robustly Optimized Bidirectional Encoder Representations from Transformers (RoBERTa) that can learn to predict hidden (masked) sections of text to fill in the gaps of non-transcribable parts of the documents being processed. Evaluating the proposed method on domain-specific datasets which include real medical documents, shows a significantly reduced word error rate demonstrating the effectiveness of the approach.