Ivor D. Addo

A Reference Architecture for Improving Security and Privacy in Internet of Things Applications

As the promise of the Internet of Things (IoT) materializes in our everyday lives, we are often challenged with a number of concerns regarding the efficacy of the current data privacy solutions that support the pervasive components at play in IoT. The privacy and security concerns surrounding IoT often manifests themselves as a treat to end-user adoption and negatively impacts trust among end-users in these solutions. In this paper, we present a reference software architecture for building cloud-enabled IoT applications in support of collaborative pervasive systems aimed at achieving trustworthiness among end-users in IoT scenarios. We present a case study that leverages this reference architecture to protect sensitive user data in an IoT application implementation and evaluate the response of an end-user study accomplished through a survey.

A Reference Architecture for High-Availability Automatic Failover between PaaS Cloud Providers

As the adoption rate of Cloud Computing continues to clamber on among various application archetypes, there is a growing concern for identifying reliable automatic failover solutions between various cloud providers in an attempt to minimize the effect of recent cloud provider outages among diverse always-on and mission-critical applications in healthcare, e-Commerce and ancillary settings. Automatic failover between cloud providers stands out as a solution for course-plotting application reliability requirements in support of high-availability, disaster recovery and high-performance scenarios. Using a case study involving Microsofts Windows Azure cloud and the Google App Engine cloud solution, we investigate some of the key characteristics in this area of concern and present a reference architecture for automatic failover between multiple Platform-as-a-Service (PaaS) cloud delivery providers in a bid to maximize the delivery of architecturally significant quality attributes pertaining to High-Availability, Performance and Disaster Recovery in a mission-critical application prototype.

Toward Collective Intelligence for Fighting Obesity

The emergent prevalence of childhood and adolescent obesity remains one of the most significant health care challenges facing the United States today. On the other hand, breakthroughs in Human-Robot Interaction (HRI) research and the diminishing cost of personal robots and virtual agents along with the ever-increasing use of smart personal devices, suggests that there is room for harnessing the power of ubiquitous intelligent systems that can work in partnership to solve some of our most difficult challenges in the very near future. In this paper, we present the design and prototype implementation of a collective intelligence approach aimed at employing machine learning algorithms that work in concert to facilitate the personalization of a humanoid robot Health Coach with a focus on childhood obesity intervention through Child-Robot Interactions and other adaptive Ubiquitous Computing (UbiComp) solutions.

Applying affective feedback to reinforcement learning in ZOEI, a comic humanoid robot

As robotic technologies of varying shapes and forms continue to make their way into our everyday lives, the significance of a humanoid robots ability to make a human interaction feel natural, engaging and entertaining becomes an area of keen interest in sociable robotics. In this paper, we present our findings on how affective feedback can be used to drive reinforcement learning in human-robot interactions (HRI) and other dialogue systems. We implemented a system where a humanoid robot, named ZOEI, acts as a standup comedian by entertaining a human audience in a bid to generate humor and positively influence the emotional state of the humans. The mood rating of the audience is recorded prior to the interaction session. Using a survey, the eventual emotional state of the human participant is captured after the HRI session. For each audience member, we capture feedback regarding how funny each joke was. We present the implementation of the content selection framework. We share our findings to substantiate the idea that by using expressive behaviors of the humanoid to influence the delivery of content (in this case, jokes) as well as employing reinforcement learning techniques for driving targeted content selection, the robot was able to improve the human mood score progressively across the 16 people who engaged in the study.

Toward an mHealth Intervention for Smoking Cessation

The prevalence of tobacco dependence in the United States (US) remains alarming. Invariably, smoke-related health problems are the leading preventable causes of death in the US. Research has shown that a culturally tailored cessation counseling program can help reduce smoking and other tobacco usage. In this paper, we present a mobile health (mHealth) solution that leverages the Short Message Service (SMS) or text messaging feature of mobile devices to motivate behavior change among tobacco users. Our approach implements the Theory of Planned Behavior (TPB) and a phase-based framework. We make contributions to improving previous mHealth intervention approaches by delivering personalized and evidence-based motivational SMS messages to participants. Our proposed solution implements machine learning algorithms that take the participants demographic profile and previous smoking behavior into account. We discuss our preliminary evaluation of the system against a couple of pseudo-scenarios and our observation of the systems performance.

PriSN: a privacy protection framework for healthcare social networking sites

A new class of patient driven healthcare web applications are emerging to supplement and extend traditional healthcare delivery models and empower patient self-care. Patient-driven healthcare can be characterized as having an increased level of information flow, transparency, customization, collaboration, patient choice and responsibility-taking, as well as quantitative, predictive and preventive aspects. Social networking applications or sites are usually dedicated to fostering interaction between users. Healthcare Social Networking (HSN) sites constitute virtual communities where users connect with each other around common health issues and share relevant health data. HSNs have become very popular and broadly adopted by various medical professionals and patients. The growing use of HSNs has prompted public concerns about the underlying risks that such online data-sharing platforms pose to the privacy and security of Personal Health Information (PHI). This paper presents a set of privacy risks introduced by social networking applications in healthcare scenarios. The main contribution of this paper is the introduction of a privacy preserving framework, PriSN, which seeks to preserve the privacy of sensitive healthcare data of end-user in HSNs. PriSN safeguards a users privacy by generalizing the contextual PHI collected in the HSN applications and shared with a given end-users peers. To support multiple levels of granularity in the contextual PHI, the proposed obfuscation procedure establishes an ontological description stating the granularity of object instances.

SPTP: A Trust Management Protocol for Online and Ubiquitous Systems

With the recent proliferation of ubiquitous, mobile and cloud-based systems, security, privacy and trust concerns surrounding the use of emerging technologies in the ensuing wake of the Internet of Things (IoT) continues to mount. In most instances, trust and privacy concerns continuously surface as a key deterrent to the adoption of these emergent technologies. The ensuing literature presents a Secure, Private and Trustworthy protocol (named SPTP) that was prototyped for addressing critical security, privacy and trust concerns surrounding mobile, pervasive and cloud services in Collective Intelligence (CI) scenarios. The efficacy of the protocol and its associated characteristics are evaluated in CI-related scenarios including multimodal monitoring of Elderly people in smart home environments, Online Advertisement targeting in Computational Advertising settings, and affective state monitoring through game play as an intervention for Autism among Children. We present our evaluation criteria for the proposed protocol, our initial results and future work.

Privacy Challenges and Goals in mHealth Systems

Abstract The global phenomena of mobile technology have encouraged collaborations between national governments and diverse international stakeholders in applying mobile-based health (mHealth) solutions as a powerful opportunity for improving health and development in rural and remote areas. A significant impact offered by modern mHealth technologies includes the potential to transform various aspects of healthcare, improving accessibility, quality, and affordability. Over the years, mHealth has become important in the field of healthcare information technology as patients begin to use mobile-based medical sensors to record their daily activities and vital signs. The rapid expansion of mobile information and communications technologies within health service delivery and public health systems has created a range of new opportunities to deliver new forms of interactive health services to patients, clinicians, and caregivers alike. The scope and scale of mHealth interventions range from simple direct-to-individual consumer and interactive patient-provider communications to more complex computer-based systems facilitating coordinated patient care and management.

Privacy in Healthcare

In recent years, the field of healthcare has seen an increased prevalence of electronic healthcare systems. Some of these systems seek to help patients make more informed decisions about their own health, while others may assist users in receiving proper care no matter where they are. Despite these positive impacts, the systems bring with them new risks. In particular, electronic healthcare systems have a variety of privacy concerns surrounding their use due to the personal nature of the data collected. In this chapter, we introduce several electronic healthcare systems that are currently in use and explore the different privacy challenges surrounding some of these systems. Finally, we highlight a few methods to address these privacy concerns and, thereby, improve privacy protection in healthcare systems.

Reality Versus Grant Application Research “Plans”

This article describes the implementation of the American Indian mHealth Smoking Dependence Study focusing on the differences between what was written in the grant application compared to what happened in reality. The study was designed to evaluate a multicomponent intervention involving 256 participants randomly assigned to one of 15 groups. Participants received either a minimal or an intense level of four intervention components: (1) nicotine replacement therapy, (2) precessation counseling, (3) cessation counseling, and (4) mHealth text messaging. The project team met via biweekly webinars as well as one to two in-person meetings per year throughout the study. The project team openly shared progress and challenges and collaborated to find proactive solutions to address challenges as compared to what was planned in the original grant application. The project team used multiple strategies to overcome unanticipated intervention issues: (1) cell phone challenges, (2) making difficult staffing decisions, (3) survey lessons, (4) nicotine replacement therapy, (5) mHealth text messages, (6) motivational interviewing counseling sessions, and (7) use of e-cigarettes. Smoking cessation studies should be designed based on the grant plans. However, on the ground reality issues needed to be addressed to assure the scientific rigor and innovativeness of this study.