Testing and Verifying the Security of COVID-19 CT Images Deep Learning System with Adversarial Attack

The Coronavirus disease 2019 (COVID-19) is a pandemic that occurred in December 2019 and spread globally. Most of the current research is on how to apply deep learning to detect COVID-19, but little research has been done on the security of COVID-19 deep learning systems. Therefore, we test and verify the security of COVID-19 CT images deep learning system with adversarial attack. Firstly, we build a deep learning system for recognizing COVID-19 CT images. Secondly, adding imperceptible disturbance to CT images will lead to neural network classification errors. Finally, we discuss the application of formal methods and formal verification to deep learning systems. We hope to draw more attention from researchers to the application of formal methods and formal verification to artificial intelligence. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Global food supply chain resilience assessment: A case in the United Kingdom

As a result of Coronavirus disease (Covid-19), container trade volumes and container port throughputs in the world have both declined over the first half of 2020. Covid - 19 causes unprecedented disruptions to the countries where food supplies heavily rely on shipping, such as the United Kingdom (UK). It is vital to assess the associated food shipping systems to ensure national food supply resilience. This paper aims to assess the national food supply chain (FSC) resilience for the UK by considering food import dependency and shipping transport connectivity. A new national food connectivity index (NFCI) framework is formulated, and supporting data is collected from the United Nations Conference on Trade and Development (UNCTAD), Food and Agriculture Organization of the United Nations (FAO), and the United Nations Commodity Trade (UN Comtrade). NFCI of the UK is calculated and compared with other countries. The formulation and analysis contribute to a newly proposed formal method to assess a nation's FSC resilience and observe and address the shortcomings of its food supply system for food security. © 2021 IEEE.

Modeling and Analysis of Medical Resource Sharing and Scheduling for Public Health Emergencies based on Petri Nets

Medical information systems (MIS) play a vital role in managing and scheduling medical resources to underpin healthcare services, which has become more critically important during major public health emergencies. During the Covid-19 pandemic, MIS is facing significant challenges to cope with the surge in demands of medical resources, resulting in more deaths and wider spreading of the disease. Our research examines how to allocate and utilize the medical resources across hospitals in a more accurate, and effective way to mitigate medical resource shortages and sustain the resource provisions. This paper mainly investigated the hospital's supply-and-demand problems for medical resources under major public health emergencies by analyzing the allocation of medical staff resources. Furthermore, a formal method based on the Colored Petri Nets (CPN) has been proposed to model and characterize the medical business process and resource scheduling tasks. The experiments demonstrate that our approach can correctly and efficiently complete the dynamical scheduling process for surging requests. © 2021 IEEE.

Explainability of radiomics through formal methods.

BACKGROUND AND OBJECTIVE: Artificial Intelligence has proven to be effective in radiomics. The main problem in using Artificial Intelligence is that researchers and practitioners are not able to know how the predictions are generated. This is currently an open issue because results' explainability is advantageous in understanding the reasoning behind the model, both for patients than for implementing a feedback mechanism for medical specialists using decision support systems. METHODS: Addressing transparency issues related to the Artificial Intelligence field, the innovative technique of Formal methods use a mathematical logic reasoning to produce an automatic, quick and reliable diagnosis. In this paper we analyze results given by the adoption of Formal methods for the diagnosis of the Coronavirus disease: specifically, we want to analyse and understand, in a more medical way, the meaning of some radiomic features to connect them with clinical or radiological evidences. RESULTS: In particular, the usage of Formal methods allows the authors to do statistical analysis on the feature value distributions, to do pattern recognition on disease models, to generalize the model of a disease and to reach high performances of results and interpretation of them. A further step for explainability can be accounted by the localization and selection of the most important slices in a multi-slice approach. CONCLUSIONS: In conclusion, we confirmed the clinical significance of some First order features as Skewness and Kurtosis. On the other hand, we suggest to decline the use of the Minimum feature because of its intrinsic connection with the Computational Tomography exam of the lung.

A Lightweight Development of Outbreak Prevention Strategies Built on Formal Methods and xDSLs

Because of the Covid-19 pandemic, several organisations around the world applied social distancing rules with workplace controls. Most of these rules can be automated and supervised using software systems that interact with connected devices such as smart cameras, motion sensors, smart door locks, etc. Given the critical nature of a pandemic prevention application, it seems essential to use techniques such that the possibility of failures is minimised. The integration of formal reasoning within software development is obviously a way to achieve this goal. Unfortunately, often formal methods are deemed too difficult and hence their application is somehow limited. This study builds on real-life pandemic prevention strategies, and shows how a formal method and domain-specific languages can be mixed in a lightweight development process. Our approach extends Meeduse, a language workbench that embeds an animator and model-checker and allows one to define proved executable Domain-Specific Languages (xDSLs) using the B method. In addition to the benefits of using xDSLs together with a formal approach during the development process the originality of this work is two-fold: (1) first, we propose a novel refinement-based approach that allows DSL developers to produce several versions of the application without breaking down the global safety properties;and (2) second, we use the verified specification of the DSL semantics at runtime so that the implementation effort is highly reduced. © 2021 ACM.

Towards Goal-Oriented Software Requirements Elicitation

Correct and unambiguous software requirements are key to the success of any software engineering project. Eliciting such requirements is a daunting task. In this paper, we present a framework that uses goal orientation as its main building blocks. Unlike other frameworks that have been reported in the literature, this framework strives to balance a compromise between formal methods on one hand and natural language on the other hand in specifying operations. A Chabot for covid-19 is presented to illustrate the framework. © 2021 IEEE.

Teaching Software Quality Assurance (SQA) during COVID-19 Using the HyFlex Approach - Course Design, Results, and Experiences

This full paper discusses the HyFlex pedagogical approach to teaching a highly interactive face-to-face Software Quality Assurance (SQA) course during the COVID-19 pandemic. HyFlex, short for hybrid-flexibility, is a teaching model where instructors and students have the option to attend scheduled courses face-to-face (F2F) or remotely. In this teaching model the learning outcomes for the course remain the same for all who enroll regardless of the mode of attendance. Our HyFlex SQA approach consists of lectures (in class, with video recordings), face-to-face activities, as well as group assignments, group projects, and exams all facilitated through an online campus management system. During the lecture period, the instructor delivers content in the form of lecture slides and writing on a whiteboard. This poses significant challenges for the instructor, as the learning outcomes have to be delivered in different modalities, but with the same quality. This is particularly difficult in SQA courses, as these require instruction in formal methods as well as systematic justification of engineering choices, both of which are best facilitated in F2F fashion that implement Think-Pair-Share (TPS) amongst students. TPS is an active learning technique in which students are allocated adequate time to think individually on a task posed by the instructor, followed by pair discussions, and then as a class discussion. The task given by the instructor is of similar complexity to that which is covered as an example during lecture. Online synchronous activities involve students participating in TPS by working collaboratively as a group on tasks that correspond with concepts covered in the slide-based lectures. This way students learn from each other by thinking individually then sharing ideas in the classroom, thus contributing to better understanding of course content. For assignments and projects students are allocated a portion of the class time to meet with group members and discuss their activities. Groups also have the option to ask the instructor questions aloud that will help other groups to complete their assignments and projects successfully. Delivering these in a hybrid format was required during the fall of 2020 due to COVID-19 restrictions. The results show that while performance in projects and homework assignments remained constant, final exam performance was significantly (p < 0.05) lower in 2020 compared to previous course offerings. We also noted a lower enrollment, higher participatory effort on both instructors and students, and a subjectively decreased feeling of collaboration. Nevertheless, students rated their perceived learning experience as high and regard HyFlex learning facilities as adequate. In this paper we adopted a HyFlex teaching model that incorporates reduced F2F seating, educational tools such as Blackboard, Panopto, Zoom, Google docs, and Discord. We conclude by discussing some challenges experienced with HyFlex teaching model and recommendations for adopting the teaching model by other instructors who teach CS courses that involve a considerable amount of group activities. © American Society for Engineering Education, 2021

Increasing Student Self-Reliance and Engagement in Model-Checking Courses

Courses on formal methods focus on two aspects: teaching formalisms and exemplary applications as well as teaching techniques for implementing tools such as model checkers. In this article, we discuss the second aspect and typical shortcomings of corresponding courses. As courses often focus on theoretical results, opportunities for working on real implementations are scarce. In consequence, students are easily overwhelmed with transfer tasks, e.g., when working on existing model checkers during theses or research projects. We present several iterations of our course on model checking, including their goals, course execution as well as feedback from peers and students. Additionally, we discuss how the Covid-19 epidemic impacted our course format and how it was made more suitable for online teaching. Finally, we use these insights to discuss the influence of formality on student engagement, and how to incorporate more practical aspects by introducing inquiry and research-based teaching. © 2021, Springer Nature Switzerland AG.

On the Adoption of Radiomics and Formal Methods for COVID-19 Coronavirus Diagnosis.

Considering the current pandemic, caused by the spreading of the novel Coronavirus disease, there is the urgent need for methods to quickly and automatically diagnose infection. To assist pathologists and radiologists in the detection of the novel coronavirus, in this paper we propose a two-tiered method, based on formal methods (to the best of authors knowledge never previously introduced in this context), aimed to (i) detect whether the patient lungs are healthy or present a generic pulmonary infection; (ii) in the case of the previous tier, a generic pulmonary disease is detected to identify whether the patient under analysis is affected by the novel Coronavirus disease. The proposed approach relies on the extraction of radiomic features from medical images and on the generation of a formal model that can be automatically checked using the model checking technique. We perform an experimental analysis using a set of computed tomography medical images obtained by the authors, achieving an accuracy of higher than 81% in disease detection.