Ivan Poliakov

Exploring the acceptability of google glass as an everyday assistive device for people with parkinson's

We describe a qualitative study investigating the acceptability of the Google Glass eyewear computer to people with Parkinsons disease (PD). We held a workshop with 5 PD patients and 2 carers exploring perceptions of Glass. This was followed by 5-day field trials of Glass with 4 PD patients, where participants wore the device during everyday activities at home and in public. We report generally positive responses to Glass as a device to instil confidence and safety for this potentially vulnerable group. We also raise concerns related to the potential for Glass to reaffirm dependency on others and stigmatise wearers.

Workcraft: a static data flow structure editing, visualisation and analysis tool

Reliable high-level modeling constructs are crucial to the design of efficient asynchronous circuits. Concepts such as static data flow structures (SDFS) considerably facilitate the design process by separating the circuit structure and functionality from the lower-level implementation details. Aside from providing a more abstract, higher level view, SDFS allow for efficient circuit analysis that is done by converting it to a Petri Net preserving behavioural equivalence. Once the equivalent Petri Net is obtained, existing theoretical and tool base can be applied to perform the model verification. However, recent advances in SDFS design were largely theoretical. There are no practical software tools available which would allow working with different SDFS models in a consistent way and provide means for their analysis and comparison. This paper presents a tool which aims to provide a common, cross-platform environment to assist with aforementioned tasks. The tool offers a GUI-based framework for visual editing, real-time simulation, animation and extendable analysis features for different SDFS types. The models themselves, as well as the supporting tools, are implemented as plug-ins.

Automated Verification of Asynchronous Circuits Using Circuit Petri Nets

To detect problematic circuit behaviour, such as potential hazards and deadlocks, in a reasonable amount of time a technique is required which would avoid exhaustive exploration of the state space of the system. Many of the existing methods rely on symbolic traversal of the state space, with the use of binary decision diagrams (BDDs) and associated software packages. This paper presents an alternative approach of using a special type of Petri nets to represent circuits. An algorithm for automatic conversion of a circuit netlist into a behaviourally equivalent Petri net is proposed. Once the circuit Petri net is constructed and composed with the provided environment specification, the presence and reachability of troublesome states is verified by using methods based on finite prefixes of Petri net unfoldings. The shortest trace leading to a deadlock or a hazard in the circuit Petri net is mapped back onto the gate-level representation of the circuit, thus assisting a designer in solving the problem. The method has been automated and compared against previously existing circuit verification tools.

Comparison of INTAKE24 (an Online 24-h Dietary Recall Tool) with Interviewer-Led 24-h Recall in 11–24 Year-Old

Online dietary assessment tools offer a convenient, low cost alternative to traditional dietary assessment methods such as weighed records and face-to-face interviewer-led 24-h recalls. INTAKE24 is an online multiple pass 24-h recall tool developed for use with 11–24 year-old. The aim of the study was to undertake a comparison of INTAKE24 (the test method) with interviewer-led multiple pass 24-h recalls (the comparison method) in 180 people aged 11–24 years. Each participant completed both an INTAKE24 24-h recall and an interviewer-led 24-h recall on the same day on four occasions over a one-month period. The daily energy and nutrient intakes reported in INTAKE24 were compared to those reported in the interviewer-led recall. Mean intakes reported using INTAKE24 were similar to the intakes reported in the interviewer-led recall for energy and macronutrients. INTAKE24 was found to underestimate energy intake by 1% on average compared to the interviewer-led recall with the limits of agreement ranging from minus 49% to plus 93%. Mean intakes of all macronutrients and micronutrients (except non-milk extrinsic sugars) were within 4% of the interviewer-led recall. Dietary assessment that utilises technology may offer a viable alternative and be more engaging than paper based methods, particularly for children and young adults.

Workcraft --- A Framework for Interpreted Graph Models

A large number of models that are employed in the field of concurrent systems design, such as Petri Nets, gate-level circuits, Static Data Flow Structures and Conditional Partial Order Graphs have an underlying static graph structure. Their semantics, however, is defined using additional entities, e.g. tokens or node/arc states, which in turn form the overall state of the system. We jointly refer to such formalisms as Interpreted Graph Models . The similarities in notation allow for links between different models to be created, such as interfaces between different formalisms or conversion from one model type into another, which greatly extend the range of applicable analysis techniques. This paper presents the new version of the Workcraft tool designed to provide a flexible common framework for development of Interpreted Graph Models, including visual editing, (co-)simulation and analysis. The latter can be carried out either directly or by mapping a model into a behaviourally equivalent model of a different type (usually a Petri Net). Hence the user can design a system using the most appropriate formalism (or even different formalisms for the subsystems), while still utilising the power of Petri Net analysis techniques. The tool is platform-independent, highly customisable by means of plug-ins, and is freely available for academic use.

LApp: A Speech Loudness Application for People with Parkinson's on Google Glass

Reduced vocal volume in Parkinsons is extremely common and can have significant social and emotional impact. We describe the development and evaluation of LApp--an application for Google Glass to help people with Parkinsons (PwP) monitor their speech volume and cue themselves to speak louder when necessary. Our findings highlight enthusiasm for using the application both at home as a volume training tool and in public social settings as a situated cueing device. We contribute insights to the literature on how eyewear technologies can provide assistance to people with health conditions and offer insights for the design of future self-monitoring and management applications on Google Glass.

Asynchronous Data Path Models

A token-based model for asynchronous data path is formally defined and three token game semantics, spread token, antitoken and counterflow, are introduced. These semantics are studied and their advantages and drawbacks are highlighted. For analysis and comparison a software tool is developed which integrates these models into a consistent framework. The models are verified by mapping them into Petri nets and employing the existing model checking tools.

Iterative Development of an Online Dietary Recall Tool: INTAKE24

Collecting large-scale population data on dietary intake is challenging, particularly when resources and funding are constrained. Technology offers the potential to develop novel ways of collecting large amounts of dietary information while making it easier, more convenient, intuitive, and engaging for users. INTAKE24 is an online multiple pass 24 h dietary recall tool developed for use in national food and nutrition surveys. The development of INTAKE24 was a four-stage iterative process of user interaction and evaluation with the intended end users, 11–24 years old. A total of 80 11–24 years old took part in the evaluation, 20 at each stage. Several methods were used to elicit feedback from the users including, ‘think aloud’, ‘eye tracking’, semi-structured interviews, and a system usability scale. Each participant completed an interviewer led recall post system completion. Key system developments generated from the user feedback included a ‘flat’ interface, which uses only a single interface screen shared between all of the various activities (e.g., free text entry, looking up foods in the database, portion size estimation). Improvements to the text entry, search functionality, and navigation around the system were also influenced through feedback from users at each stage. The time to complete a recall using INTAKE24 almost halved from the initial prototype to the end system, while the agreement with an interviewer led recall improved. Further developments include testing the use of INTAKE24 with older adults and translation into other languages for international use. Our future aim is to validate the system with recovery biomarkers.

Recommender system based on pairwise association rules

Abstract Recommender systems based on methods such as collaborative and content-based filtering rely on extensive user profiles and item descriptors as well as on an extensive history of user preferences. Such methods face a number of challenges; including the cold-start problem in systems characterized by irregular usage, privacy concerns, and contexts where the range of indicators representing user interests is limited. We describe a recommender algorithm that builds a model of collective preferences independently of personal user interests and does not require a complex system of ratings. The performance of the algorithm is analyzed on a large transactional data set generated by a real-world dietary intake recall system.