Suzanne Kieffer

Feature extraction and selection for objective gait analysis and fall risk assessment by accelerometry

BackgroundFalls in the elderly is nowadays a major concern because of their consequences on elderly general health and moral states. Moreover, the aging of the population and the increasing life expectancy make the prediction of falls more and more important. The analysis presented in this article makes a first step in this direction providing a way to analyze gait and classify hospitalized elderly fallers and non-faller. This tool, based on an accelerometer network and signal processing, gives objective informations about the gait and does not need any special gait laboratory as optical analysis do. The tool is also simple to use by a non expert and can therefore be widely used on a large set of patients.MethodA population of 20 hospitalized elderlies was asked to execute several classical clinical tests evaluating their risk of falling. They were also asked if they experienced any fall in the last 12 months. The accelerations of the limbs were recorded during the clinical tests with an accelerometer network distributed on the body. A total of 67 features were extracted from the accelerometric signal recorded during a simple 25 m walking test at comfort speed. A feature selection algorithm was used to select those able to classify subjects at risk and not at risk for several classification algorithms types.ResultsThe results showed that several classification algorithms were able to discriminate people from the two groups of interest: fallers and non-fallers hospitalized elderlies. The classification performances of the used algorithms were compared. Moreover a subset of the 67 features was considered to be significantly different between the two groups using a t-test.ConclusionsThis study gives a method to classify a population of hospitalized elderlies in two groups: at risk of falling or not at risk based on accelerometric data. This is a first step to design a risk of falling assessment system that could be used to provide the right treatment as soon as possible before the fall and its consequences. This tool could also be used to evaluate the risk several times during the revalidation procedure.

Multi-fidelity prototyping of user interfaces

Multi-fidelity prototyping combines within a single user interface prototype several elements whose representations are reproduced with different levels of fidelity with respect to the final user interface: no fidelity at all, low fidelity, medium fidelity, and high fidelity. In order to effectively and efficiently support multi-fidelity, an original set of techniques is defined and discussed: multiple representation manipulation by sketching and gesture recognition, smooth transition between any representation at any time, prototype reusability, multi-platform support, and multiple user interface prototyping facilities. The full implementation of these techniques in prototyping software provides designers and developers with a unique environment for exploring multiple designs with unprecedented support for quickly designing interfaces from scratch or from previously existing design templates. An experimental study reveals that the multiple representation manipulation together with smooth transition represents a valuable advantage for naturally designing user interfaces. The prototyping software supports several aspects involved in the user interface development life cycle and is convenient for non-WIMP user interfaces.

Effects of display layout on gaze activity during visual search

We report an experimental study that aims at investigating the influence of spatial layout on visual search efficiency and comfort. 4 layouts were used for displaying 120 scenes comprising 30 realistic colour photos each: random, elliptic, radial and matrix-like. Scenes (30 per structure) were presented to 5 participants who had to select a pre-viewed photo in each scene using the mouse. Eye-tracking data indicate that elliptic layouts provided better visual comfort than any of the other layouts (shortest scan paths), and proved to be more efficient than matrix layouts (shorter search times). These results are statistically significant (paired t-tests).

User interface design by sketching: a complexity analysis of widget representations

User interface design by sketching, as well as other sketching activities, typically involves sketching objects through representations that should combine meaningfulness for the end users and easiness for the recognition engines. To investigate this relationship, a multi-platform user interface design tool has been developed that enables designers to sketch design ideas in multiple levels of fidelity with multi-stroke gestures supporting widget representations and commands. A usability analysis of these activities, as they are submitted to a recognition engine, suggests that the level of fidelity, the amount of constraints imposed on the representations, and the visual difference of representations positively impact the sketching activity as a whole. Implications for further sketch representations in user interface design and beyond are provided based on usability guidelines.

User-Centered Design and Fast Prototyping of an Ambient Assisted Living System for Elderly People

This paper presents the Keep-In-Touch project, which aims at developing an integrated Ambient Assisted Living (AAL) solution assisting and monitoring elderly people in their daily-life activities, supporting personal autonomy and well-being, and maintaining social cohesion. The focus of this paper is the integration of interactive user modeling, the benefit of the combination of user-centered development method and fast prototyping implementation in order to develop a solution which fits the end-user. The key elements to achieve this goal are: multimodality, accessibility, adaptability to user profile and changes, and usability. We will show how our approach addresses these needs.

Usability evaluation of virtual reality interaction techniques for positioning and manoeuvring in reduced, manipulation-oriented environments

This paper introduces some novel interaction techniques based on the concepts of composite positioning and composite manoeuvring (described in the paper). In contrast with other previous proposals, these techniques have been designed and evaluated in the context of a user centred process. The results of this evaluation and some relevant findings for the field of human computer interaction are also described.

STRATUS: a questionnaire for strategic usability assessment

Strategic usability is widely recognized as a challenge, as moving from scattered to strategic usability requires organizations to operate significant changes. The current solutions are hardly accessible and complicated to implement. STRATUS is a questionnaire for cost-effectively assessing strategic usability. A companion model enables to easily analyze and describe the current state of strategic usability according to five dimensions and related indicators. A pilot STRATUS survey compared the responses between novices and experts in usability. It highlighted dependencies between experience and usability awareness, as well as positive correlations between the presence of usability staff and strategical use of usability. The questionnaire needs now to be validated by means of a longitudinal study. Another promising track towards its validation consists in further investigating the causations between the dimensions of strategic usability.

Interactive breast cancer segmentation based on relevance feedback: from user-centered design to evaluation

Computer systems play an important role in medical imaging industry since radiologists depend on it for visualization, interpretation, communication and archiving. In particular, computer-aided diagnosis (CAD) systems help in lesion detection tasks. This paper presents the design and the development of an interactive segmentation tool for breast cancer screening and diagnosis. The tool conception is based upon a user-centered approach in order to ensure that the application is of real benefit to radiologists. The analysis of user expectations, workflow and decision-making practices give rise to the need for an interactive reporting system based on the BIRADS, that would not only include the numerical features extracted from the segmentation of the findings in a structured manner, but also support human relevance feedback as well. This way, the numerical results from segmentation can be either validated by end-users or enhanced thanks to domain-experts subjective interpretation. Such a domain-expert centered system requires the segmentation to be sufficiently accurate and locally adapted, and the features to be carefully selected in order to best suit users knowledge and to be of use in enhancing segmentation. Improving segmentation accuracy with relevance feedback and providing radiologists with a user-friendly interface to support image analysis are the contributions of this work. The preliminary result is first the tool conception, and second the improvement of the segmentation precision.

How really effective are multimodal hints in enhancing visual target spotting? Some evidence from a usability study

The main aim of the work presented here is to contribute to computer science advances in the multimodal usability area, in-asmuch as it addresses one of the major issues relating to the generation of effective oral system messages: how to design messages which effectively help users to locate specific graphical objects in information visualisations? An experimental study was carried out to determine whether oral messages including coarse information on the locations of graphical objects on the current display may facilitate target detection tasks sufficiently for making it worth while to integrate such messages in GUIs. The display spatial layout varied in order to test the influence of visual presentation structure on the contribution of these messages to facilitating visual search on crowded displays. Finally, three levels of task difficulty were defined, based mainly on the target visual complexity and the number of distractors in the scene. The findings suggest that spatial information messages improve participants’ visual search performances significantly; they are more appropriate to radial structures than to matrix, random and elleptic structures; and, they are particularly useful for performing difficult visual search tasks.

Towards task analysis tool support

This paper discusses challenges in contextual task analysis and the need of tools that support analysts to collect such information in context. Specifically we argue that the analysis of collaborative and distributed tasks can be supported by ambulatory assessment tools. We illustrate how contextual task analysis can be supported by TEMPEST, a platform originally created for experience sampling and more generally, longitudinal ambulatory assessment studies. We present a case study that illustrates the extent to which this tool meets the needs of real-world task analysis, describing the gains in efficiency it can provide but also directions for the development of tool support for task analysis.