AI-assisted ethics? considerations of AI simulation for the ethical assessment and design of assistive technologies

Current ethical debates on the use of artificial intelligence (AI) in healthcare treat AI as a product of technology in three ways. First, by assessing risks and potential benefits of currently developed AI-enabled products with ethical checklists; second, by proposing ex ante lists of ethical values seen as relevant for the design and development of assistive technology, and third, by promoting AI technology to use moral reasoning as part of the automation process. The dominance of these three perspectives in the discourse is demonstrated by a brief summary of the literature. Subsequently, we propose a fourth approach to AI, namely, as a methodological tool to assist ethical reflection. We provide a concept of an AI-simulation informed by three separate elements: 1) stochastic human behavior models based on behavioral data for simulating realistic settings, 2) qualitative empirical data on value statements regarding internal policy, and 3) visualization components that aid in understanding the impact of changes in these variables. The potential of this approach is to inform an interdisciplinary field about anticipated ethical challenges or ethical trade-offs in concrete settings and, hence, to spark a re-evaluation of design and implementation plans. This may be particularly useful for applications that deal with extremely complex values and behavior or with limitations on the communication resources of affected persons (e.g., persons with dementia care or for care of persons with cognitive impairment). Simulation does not replace ethical reflection but does allow for detailed, context-sensitive analysis during the design process and prior to implementation. Finally, we discuss the inherently quantitative methods of analysis afforded by stochastic simulations as well as the potential for ethical discussions and how simulations with AI can improve traditional forms of thought experiments and future-oriented technology assessment.

At Crossroads in a Virtual City: Effect of Spatial Disorientation on Gait Variability and Psychophysiological Response among Healthy Older Adults

<b><i>Introduction:</i></b> Aging has been associated with a decline in cognitive and motor performance, often expressed in multitasking situations, which could include wayfinding. A major challenge to successful wayfinding is spatial disorientation, occurring mostly at crossings. Although gait changes have been observed in various dual-task conditions, little is known about the effect of disorientation on gait and psychophysiological response among older adults during wayfinding. The study aimed at identifying the effect of spatial disorientation on gait variability and psychophysiological response among healthy older adults during wayfinding in a controlled environment. <b><i>Method:</i></b> We analyzed data of 28 participants (age 70.8 ± 4.6, 18 female), 14 experimental and 14 controls. Participants performed a wayfinding task consisting of 14 major decision points (7 intersections) within a virtual environment (VE) projected on a 180° screen while walking on a self-paced treadmill equipped with a marker-based optical motion-capture system. The VE was held constant for the controls and manipulated for the experimental participants. Disorientation was identified based on a customized annotation scheme. Variability in gait, including the coefficient of variation (CV), was measured as the primary endpoint. Psychophysiological response measures, including heart rate variability (RMSSD) and skin conductance response (SCR), were continuously monitored as secondary endpoints and estimates of cognitive effort. Linear Mixed Effects models were applied to hypothesis-driven outcome measures extracted from decision points. <b><i>Results:</i></b> Walking speed and step length decreased when disoriented (<i>p</i> &#x3c; 0.05), while stride time, stance time, walking speed CV, stance time CV, SCR amplitude, and SCR count increased when disoriented (<i>p</i> &#x3c; 0.05). A higher RMSSD was associated with being disoriented at crossings (<i>p</i> &#x3c; 0.05). SCR count was greater in the older experimental group (<i>p</i> &#x3c; 0.001), including when disoriented (<i>p</i> &#x3c; 0.001). <b><i>Discussion/Conclusion:</i></b> The results provide evidence for the impact of spatial disorientation on changes in gait pattern and psychophysiological response among older adults during wayfinding. Location also had implications for the effect of disorientation on gait and cognitive effort. This gives further insight into the substrates of real-world navigation challenges among older adults, with an emphasis on viable features for designing situation-adaptive interventional devices aiding independent mobility.

Automatic Generation of Personalised and Context-Dependent Textual Interventions During Neuro-rehabilitation

In this paper we present our system that synthesises personalised and context dependent texts during robot guided exercises for neuro-rehabilitation. This system is used to generate texts for the communication between a care robot and patients. We present requirements that a system in such a medical domain has to meet. Afterwards the results of a systematic literature review are presented. We present our solution based on the RosaeNLG system. It supports different language levels and referring expressions in a real-time text generation system, so that generated texts can be adapted to the reader in the best possible way. We evaluate our system with respect to the requirements. The contribution of the paper is twofold: We present a set of requirements for Natural Language Generation (NLG) in medical domains and we show how to extend RosaeNLG with an external dialogue memory to handle complex referring expressions in medical real time settings.

Prediction of Disorientation by Accelerometric and Gait Features in Young and Older Adults Navigating in a Virtually Enriched Environment

Objective

To determine whether gait and accelerometric features can predict disorientation events in young and older adults.

Methods

Cognitively healthy younger (18–40 years, n = 25) and older (60–85 years, n = 28) participants navigated on a treadmill through a virtual representation of the city of Rostock featured within the Gait Real-Time Analysis Interactive Lab (GRAIL) system. We conducted Bayesian Poisson regression to determine the association of navigation performance with domain-specific cognitive functions. We determined associations of gait and accelerometric features with disorientation events in real-time data using Bayesian generalized mixed effect models. The accuracy of gait and accelerometric features to predict disorientation events was determined using cross-validated support vector machines (SVM) and Hidden Markov models (HMM).

Results

Bayesian analysis revealed strong evidence for the effect of gait and accelerometric features on disorientation. The evidence supported a relationship between executive functions but not visuospatial abilities and perspective taking with navigation performance. Despite these effects, the cross-validated percentage of correctly assigned instances of disorientation was only 72% in the SVM and 63% in the HMM analysis using gait and accelerometric features as predictors.

Conclusion

Disorientation is reflected in spatiotemporal gait features and the accelerometric signal as a potentially more easily accessible surrogate for gait features. At the same time, such measurements probably need to be enriched with other parameters to be sufficiently accurate for individual prediction of disorientation events.

Lifted Bayesian Filtering in Multiset Rewriting Systems

We present a model for Bayesian filtering (BF) in discrete dynamic systems where multiple entities (inter)-act, i.e. where the system dynamics is naturally described by a Multiset rewriting system (MRS). Typically, BF in such situations is computationally expensive due to the high number of discrete states that need to be maintained explicitly. We devise a lifted state representation, based on a suitable decomposition of multiset states, such that some factors of the distribution are exchangeable and thus afford an efficient representation. Intuitively, this representation groups together similar entities whose properties follow an exchangeable joint distribution. Subsequently, we introduce a BF algorithm that works directly on lifted states, without resorting to the original, much larger ground representation. This algorithm directly lends itself to approximate versions by limiting the number of explicitly represented lifted states in the posterior. We show empirically that the lifted representation can lead to a factorial reduction in the representational complexity of the distribution, and in the approximate cases can lead to a lower variance of the estimate and a lower estimation error compared to the original, ground representation.

Effect of Spatial Disorientation in a Virtual Environment on Gait and Vital Features in Patients with Dementia: Pilot Single-Blind Randomized Control Trial

Background

Orientation deficits are among the most devastating consequences of early dementia. Digital navigation devices could overcome these deficits if adaptable to the user’s needs (ie, provide situation-aware, proactive navigation assistance). To fulfill this task, systems need to automatically detect spatial disorientation from sensors in real time. Ideally, this would require field studies consisting of real-world navigation. However, such field studies can be challenging and are not guaranteed to cover sufficient instances of disorientation due to the large variability of real-world settings and a lack of control over the environment.

Objective

Extending a foregoing field study, we aim to evaluate the feasibility of using a sophisticated virtual reality (VR) setup, which allows a more controlled observation of disorientation states and accompanying behavioral and physiological parameters in cognitively healthy older people and people with dementia.

Methods

In this feasibility study, we described the experimental design and pilot outcomes of an ongoing study aimed at investigating the effect of disorientation on gait and selected physiological features in a virtual laboratory. We transferred a real-world navigation task to a treadmill-based virtual system for gait analysis. Disorientation was induced by deliberately manipulating landmarks in the VR projection. Associated responses in motion behavior and physiological parameters were recorded by sensors. Primary outcomes were variations in motion and physiological parameters, frequency of disorientation, and questionnaire-derived usability estimates (immersion and perceived control of the gait system) for our population of interest. At this time, the included participants were 9 cognitively healthy older participants [5/9 women, 4/9 men; mean age 70 years, SD 4.40; Mini–Mental State Examination (MMSE) mean 29, SD 0.70) and 4 participants with dementia (2/4 women, 2/4 men; mean age 78 years, SD 2.30 years; MMSE mean 20.50, SD 7.54). Recruitment is ongoing, with the aim of including 30 cognitively healthy older participants and 20 participants with dementia.

Results

All 13 participants completed the experiment. Patients’ route was adapted by shortening it relative to the original route. Average instances of disorientation were 21.40, 36.50, and 37.50 for the cognitively healthy older control, cognitively healthy older experimental participants, and participants with dementia, respectively. Questionnaire outcomes indicated that participants experienced adequate usability and immersion; 4.30 for presence, 3.73 for involvement, and 3.85 for realism of 7 possible points, indicating a good overall ability to cope with the experiment. Variations were also observed in motion and physiological parameters during instances of disorientation.

Conclusions

This study presents the first feasibility outcomes of a study investigating the viability of using a sophisticated VR setup, based on an earlier real-world navigation study, to study spatial disorientation among cognitively healthy older people and people with dementia. Preliminary outcomes give confidence to the notion that our setup can be used to assess motion and physiological markers of disorientation, even in people with cognitive decline.

Trial Registration

ClinicalTrials.gov; https://clinicaltrials.gov/ct2/show/NCT04134806

Gaussian Graphical Models Reveal Inter-Modal and Inter-Regional Conditional Dependencies of Brain Alterations in Alzheimer's Disease

Alzheimer's disease (AD) is characterized by a sequence of pathological changes, which are commonly assessed in vivo using various brain imaging modalities such as magnetic resonance imaging (MRI) and positron emission tomography (PET). Currently, the most approaches to analyze statistical associations between regions and imaging modalities rely on Pearson correlation or linear regression models. However, these models are prone to spurious correlations arising from uninformative shared variance and multicollinearity. Notably, there are no appropriate multivariate statistical models available that can easily integrate dozens of multicollinear variables derived from such data, being able to utilize the additional information provided from the combination of data sources. Gaussian graphical models (GGMs) can estimate the conditional dependency from given data, which is conceptually expected to closely reflect the underlying causal relationships between various variables. Hence, we applied GGMs to assess multimodal regional brain alterations in AD. We obtained data from N = 972 subjects from the Alzheimer's Disease Neuroimaging Initiative. The mean amyloid load (AV45-PET), glucose metabolism (FDG-PET), and gray matter volume (MRI) were calculated for each of the 108 cortical and subcortical brain regions. GGMs were estimated using a Bayesian framework for the combined multimodal data and the resulted conditional dependency networks were compared to classical covariance networks based on Pearson correlation. Additionally, graph-theoretical network statistics were calculated to determine network alterations associated with disease status. The resulting conditional dependency matrices were much sparser (≈10% density) than Pearson correlation matrices (≈50% density). Within imaging modalities, conditional dependency networks yielded clusters connecting anatomically adjacent regions. For the associations between different modalities, only few region-specific connections were detected. Network measures such as small-world coefficient were significantly altered across diagnostic groups, with a biphasic u-shape trajectory, i.e., increased small-world coefficient in early mild cognitive impairment (MCI), similar values in late MCI, and decreased values in AD dementia patients compared to cognitively normal controls. In conclusion, GGMs removed commonly shared variance among multimodal measures of regional brain alterations in MCI and AD, and yielded sparser matrices compared to correlation networks based on the Pearson coefficient. Therefore, GGMs may be used as alternative to thresholding-approaches typically applied to correlation networks to obtain the most informative relations between variables.

Automated sensor-based detection of challenging behaviors in advanced stages of dementia in nursing homes

INTRODUCTION

Sensor-based assessment of challenging behaviors in dementia may be useful to support caregivers. Here, we investigated accelerometry as tool for identification and prediction of challenging behaviors.

METHODS

We set up a complex data recording study in two nursing homes with 17 persons in advanced stages of dementia. Study included four-week observation of behaviors. In parallel, subjects wore sensors 24 h/7 d. Participants underwent neuropsychological assessment including MiniMental State Examination and Cohen-Mansfield Agitation Inventory.

RESULTS

We calculated the accelerometric motion score (AMS) from accelerometers. The AMS was associated with several types of agitated behaviors and could predict subject's Cohen-Mansfield Agitation Inventory values. Beyond the mechanistic association between AMS and behavior on the group level, the AMS provided an added value for prediction of behaviors on an individual level.

DISCUSSION

We confirm that accelerometry can provide relevant information about challenging behaviors. We extended previous studies by differentiating various types of agitated behaviors and applying long-term measurements in a real-world setting.

Real-Time Detection of Spatial Disorientation in Persons with Mild Cognitive Impairment and Dementia

BACKGROUND

Detecting manifestations of spatial disorientation in real time is a key requirement for adaptive assistive navigation systems for people with dementia.

OBJECTIVE

To identify predictive patterns of spatial disorientation in cognitively impaired people during unconstrained locomotion behavior in an urban environment.

METHODS

Accelerometric data and GPS records were gathered during a wayfinding task along a route of about 1 km in 15 people with amnestic mild cognitive impairment or clinically probable Alzheimer's disease dementia (13 completers). We calculated a set of 48 statistical features for each 10-s segment of the acceleration sensor signal to characterize the physical motion. We used different classifiers with the wrapper method and leave-one-out cross-validation for feature selection and for determining accuracy of disorientation detection.

RESULTS

Linear discriminant analysis using three features showed the best classification results, with a cross-validated ROC AUC of 0.75, detecting 65% of all scenes of spatial disorientation in real time. Consideration of an additional feature that informed about a person's distance to the next traffic junction did not provide an additional information gain.

CONCLUSIONS

Accelerometric data are able to capture the uniformity and activity of a person's walking, which are identified as the most informative locomotion features of spatially disoriented behavior. This serves as an important basis for real-time navigation assistance. To improve the required accuracy of real-time disorientation prediction, as a next step we will analyze whether location-based behavior is able to inform about person-centered habitual factors of orientation.

A Tablet App- and Sensor-Based Assistive Technology Intervention for Informal Caregivers to Manage the Challenging Behavior of People With Dementia (the insideDEM Study): Protocol for a Feasibility Study

Background

Despite the enormous number of assistive technologies (ATs) in dementia care, the management of challenging behavior (CB) of persons with dementia (PwD) by informal caregivers in home care is widely disregarded. The first-line strategy to manage CB is to support the understanding of the underlying causes of CB to formulate individualized nonpharmacological interventions. App- and sensor-based approaches combining multimodal sensors (actimetry and other modalities) and caregiver information are innovative ways to support the understanding of CB for family caregivers.

Objective

The main aim of this study is to describe the design of a feasibility study consisting of an outcome and a process evaluation of a newly developed app- and sensor-based intervention to manage CB of PwD for family caregivers at home.

Methods

In this feasibility study, we perform an outcome and a process evaluation with a pre-post descriptive design over an 8-week intervention period. The Medical Research Council framework guides the design of this feasibility study. The data on 20 dyads (primary caregiver and PwD) are gathered through standardized questionnaires, protocols, and log files as well as semistructured qualitative interviews. The outcome measures (neuropsychiatric inventory and Cohen-Mansfield agitation inventory) are analyzed by using descriptive statistics and statistical tests relevant to the individual assessments (eg, chi-square test and Wilcoxon signed-rank test). For the analysis of the process data, the Unified Theory of Acceptance and Use of Technology is used. Log files are analyzed by using descriptive statistics, protocols are analyzed by using documentary analysis, and semistructured interviews are analyzed deductively using content analysis.

Results

The newly developed app- and sensor-based AT has been developed and was evaluated until July in 2018. The recruitment of dyads started in September 2017 and was concluded in March 2018. The data collection was completed at the end of July 2018.

Conclusions

This study presents the protocol of the first feasibility study to encompass an outcome and process evaluation to assess a complex app- and sensor-based AT combining multimodal actimetry sensors for informal caregivers to manage CB. The feasibility study will provide in-depth information about the study procedure and on how to optimize the design of the intervention and its delivery.

International Registered Report Identifier (IRRID)

DERR1-10.2196/11630

Analysing Cooking Behaviour in Home Settings: Towards Health Monitoring

Wellbeing is often affected by health-related conditions. Among them are nutrition-related health conditions, which can significantly decrease the quality of life. We envision a system that monitors the kitchen activities of patients and that based on the detected eating behaviour could provide clinicians with indicators for improving a patient’s health. To be successful, such system has to reason about the person’s actions and goals. To address this problem, we introduce a symbolic behaviour recognition approach, called Computational Causal Behaviour Models (CCBM). CCBM combines symbolic representation of person’s behaviour with probabilistic inference to reason about one’s actions, the type of meal being prepared, and its potential health impact. To evaluate the approach, we use a cooking dataset of unscripted kitchen activities, which contains data from various sensors in a real kitchen. The results show that the approach is able to reason about the person’s cooking actions. It is also able to recognise the goal in terms of type of prepared meal and whether it is healthy. Furthermore, we compare CCBM to state-of-the-art approaches such as Hidden Markov Models (HMM) and decision trees (DT). The results show that our approach performs comparable to the HMM and DT when used for activity recognition. It outperformed the HMM for goal recognition of the type of meal with median accuracy of 1 compared to median accuracy of 0.12 when applying the HMM. Our approach also outperformed the HMM for recognising whether a meal is healthy with a median accuracy of 1 compared to median accuracy of 0.5 with the HMM.

State-Space Abstractions for Probabilistic Inference: A Systematic Review

Tasks such as social network analysis, human behavior recognition, or modeling biochemical reactions, can be solved elegantly by using the probabilistic inference framework. However, standard probabilistic inference algorithms work at a propositional level, and thus cannot capture the symmetries and redundancies that are present in these tasks. Algorithms that exploit those symmetries have been devised in different research fields, for example by the lifted inference-, multiple object tracking-, and modeling and simulation-communities. The common idea, that we call state space abstraction, is to perform inference over compact representations of sets of symmetric states. Although they are concerned with a similar topic, the relationship between these approaches has not been investigated systematically. This survey provides the following contributions. We perform a systematic literature review to outline the state of the art in probabilistic inference methods exploiting symmetries. From an initial set of more than 4,000 papers, we identify 116 relevant papers. Furthermore, we provide new high-level categories that classify the approaches, based on common properties of the approaches. The research areas underlying each of the categories are introduced concisely. Researchers from different fields that are confronted with a state space explosion problem in a probabilistic system can use this classification to identify possible solutions. Finally, based on this conceptualization, we identify potentials for future research, as some relevant application domains are not addressed by current approaches.

Use of nonintrusive sensor-based information and communication technology for real-world evidence for clinical trials in dementia

Cognitive function is an important end point of treatments in dementia clinical trials. Measuring cognitive function by standardized tests, however, is biased toward highly constrained environments (such as hospitals) in selected samples. Patient-powered real-world evidence using information and communication technology devices, including environmental and wearable sensors, may help to overcome these limitations. This position paper describes current and novel information and communication technology devices and algorithms to monitor behavior and function in people with prodromal and manifest stages of dementia continuously, and discusses clinical, technological, ethical, regulatory, and user-centered requirements for collecting real-world evidence in future randomized controlled trials. Challenges of data safety, quality, and privacy and regulatory requirements need to be addressed by future smart sensor technologies. When these requirements are satisfied, these technologies will provide access to truly user relevant outcomes and broader cohorts of participants than currently sampled in clinical trials.

Situation Model for Situation-Aware Assistance of Dementia Patients in Outdoor Mobility

Background:

Dementia impairs spatial orientation and route planning, thus often affecting the patient’s ability to move outdoors and maintain social activities. Situation-aware deliberative assistive technology devices (ATD) can substitute impaired cognitive function in order to maintain one’s level of social activity. To build such a system, one needs domain knowledge about the patient’s situation and needs. We call this collection of knowledge situation model.

Objective:

To construct a situation model for the outdoor mobility of people with dementia (PwD). The model serves two purposes: 1) as a knowledge base from which to build an ATD describing the mobility of PwD; and 2) as a codebook for the annotation of the recorded behavior.

Methods:

We perform systematic knowledge elicitation to obtain the relevant knowledge. The OBO Edit tool is used for implementing and validating the situation model. The model is evaluated by using it as a codebook for annotating the behavior of PwD during a mobility study and interrater agreement is computed. In addition, clinical experts perform manual evaluation and curation of the model.

Results:

The situation model consists of 101 concepts with 11 relation types between them. The results from the annotation showed substantial overlapping between two annotators (Cohen’s kappa of 0.61).

Conclusion:

The situation model is a first attempt to systematically collect and organize information related to the outdoor mobility of PwD for the purposes of situation-aware assistance. The model is the base for building an ATD able to provide situation-aware assistance and to potentially improve the quality of life of PwD.

Time and Memory Efficient Online Piecewise Linear Approximation of Sensor Signals

Piecewise linear approximation of sensor signals is a well-known technique in the fields of Data Mining and Activity Recognition. In this context, several algorithms have been developed, some of them with the purpose to be performed on resource constrained microcontroller architectures of wireless sensor nodes. While microcontrollers are usually constrained in computational power and memory resources, all state-of-the-art piecewise linear approximation techniques either need to buffer sensor data or have an execution time depending on the segment’s length. In the paper at hand, we propose a novel piecewise linear approximation algorithm, with a constant computational complexity as well as a constant memory complexity. Our proposed algorithm’s worst-case execution time is one to three orders of magnitude smaller and its average execution time is three to seventy times smaller compared to the state-of-the-art Piecewise Linear Approximation (PLA) algorithms in our experiments. In our evaluations, we show that our algorithm is time and memory efficient without sacrificing the approximation quality compared to other state-of-the-art piecewise linear approximation techniques, while providing a maximum error guarantee per segment, a small parameter space of only one parameter, and a maximum latency of one sample period plus its worst-case execution time.

Fast care – real-time sensor data analysis framework for intelligent assistance systems

Main target of fast care is the development of a real-time capable sensor data analysis framework for intelligent assistive systems in the field of Ambient Assisted Living, eHealth, Tele Rehabilitation, and Tele Care. The aim is to provide a medically valid integrated situation model based on a distributed, ad-hoc connected, energy-efficient sensor infrastructure suitable for daily use. The integrated situation model combining physiological, cognitive, and kinematic information about the patient is grounded on the intelligent fusion of heterogeneous sensor data on different levels. The model can serve as a tool for quickly identifying risk and hazards as well as enable medical assistance systems to autonomously intervene in real-time and actively give telemedical feedback.

Multidimensional assessment of challenging behaviors in advanced stages of dementia in nursing homes-The insideDEM framework

INTRODUCTION

Assessment of challenging behaviors in dementia is important for intervention selection. Here, we describe the technical and experimental setup and the feasibility of long-term multidimensional behavior assessment of people with dementia living in nursing homes.

METHODS

We conducted 4 weeks of multimodal sensor assessment together with real-time observation of 17 residents with moderate to very severe dementia in two nursing care units. Nursing staff received extensive training on device handling and measurement procedures. Behavior of a subsample of eight participants was further recorded by videotaping during 4 weeks during day hours. Sensors were mounted on the participants' wrist and ankle and measured motion, rotation, as well as surrounding loudness level, light level, and air pressure.

RESULTS

Participants were in moderate to severe stages of dementia. Almost 100% of participants exhibited relevant levels of challenging behaviors. Automated quality control detected 155 potential issues. But only 11% of the recordings have been influenced by noncompliance of the participants. Qualitative debriefing of staff members suggested that implementation of the technology and observation platform in the routine procedures of the nursing home units was feasible and identified a range of user- and hardware-related implementation and handling challenges.

DISCUSSION

Our results indicate that high-quality behavior data from real-world environments can be made available for the development of intelligent assistive systems and that the problem of noncompliance seems to be manageable. Currently, we train machine-learning algorithms to detect episodes of challenging behaviors in the recorded sensor data.

A Process for Systematic Development of Symbolic Models for Activity Recognition

Several emerging approaches to activity recognition (AR) combine symbolic representation of user actions with probabilistic elements for reasoning under uncertainty. These approaches provide promising results in terms of recognition performance, coping with the uncertainty of observations, and model size explosion when complex problems are modelled. But experience has shown that it is not always intuitive to model even seemingly simple problems. To date, there are no guidelines for developing such models. To address this problem, in this work we present a development process for building symbolic models that is based on experience acquired so far as well as on existing engineering and data analysis workflows. The proposed process is a first attempt at providing structured guidelines and practices for designing, modelling, and evaluating human behaviour in the form of symbolic models for AR. As an illustration of the process, a simple example from the office domain was developed. The process was evaluated in a comparative study of an intuitive process and the proposed process. The results showed a significant improvement over the intuitive process. Furthermore, the study participants reported greater ease of use and perceived effectiveness when following the proposed process. To evaluate the applicability of the process to more complex AR problems, it was applied to a problem from the kitchen domain. The results showed that following the proposed process yielded an average accuracy of 78%. The developed model outperformed state-of-the-art methods applied to the same dataset in previous work, and it performed comparably to a symbolic model developed by a model expert without following the proposed development process.

Multimodal analysis of functional and structural disconnection in Alzheimer's disease using multiple kernel SVM

Alzheimer's disease (AD) patients exhibit alterations in the functional connectivity between spatially segregated brain regions which may be related to both local gray matter (GM) atrophy as well as a decline in the fiber integrity of the underlying white matter tracts. Machine learning algorithms are able to automatically detect the patterns of the disease in image data, and therefore, constitute a suitable basis for automated image diagnostic systems. The question of which magnetic resonance imaging (MRI) modalities are most useful in a clinical context is as yet unresolved. We examined multimodal MRI data acquired from 28 subjects with clinically probable AD and 25 healthy controls. Specifically, we used fiber tract integrity as measured by diffusion tensor imaging (DTI), GM volume derived from structural MRI, and the graph-theoretical measures 'local clustering coefficient' and 'shortest path length' derived from resting-state functional MRI (rs-fMRI) to evaluate the utility of the three imaging methods in automated multimodal image diagnostics, to assess their individual performance, and the level of concordance between them. We ran the support vector machine (SVM) algorithm and validated the results using leave-one-out cross-validation. For the single imaging modalities, we obtained an area under the curve (AUC) of 80% for rs-fMRI, 87% for DTI, and 86% for GM volume. When it came to the multimodal SVM, we obtained an AUC of 82% using all three modalities, and 89% using only DTI measures and GM volume. Combined multimodal imaging data did not significantly improve classification accuracy compared to the best single measures alone.

Predicting Prodromal Alzheimer's Disease in Subjects with Mild Cognitive Impairment Using Machine Learning Classification of Multimodal Multicenter Diffusion-Tensor and Magnetic Resonance Imaging Data

BACKGROUND

Alzheimer's disease (AD) patients show early changes in white matter (WM) structural integrity. We studied the use of diffusion tensor imaging (DTI) in assessing WM alterations in the predementia stage of mild cognitive impairment (MCI).

METHODS

We applied a Support Vector Machine (SVM) classifier to DTI and volumetric magnetic resonance imaging data from 35 amyloid-β42 negative MCI subjects (MCI-Aβ42-), 35 positive MCI subjects (MCI-Aβ42+), and 25 healthy controls (HC) retrieved from the European DTI Study on Dementia. The SVM was applied to DTI-derived fractional anisotropy, mean diffusivity (MD), and mode of anisotropy (MO) maps. For comparison, we studied classification based on gray matter (GM) and WM volume.

RESULTS

We obtained accuracies of up to 68% for MO and 63% for GM volume when it came to distinguishing between MCI-Aβ42- and MCI-Aβ42+. When it came to separating MCI-Aβ42+ from HC we achieved an accuracy of up to 77% for MD and a significantly lower accuracy of 68% for GM volume. The accuracy of multimodal classification was not higher than the accuracy of the best single modality.

CONCLUSIONS

Our results suggest that DTI data provide better prediction accuracy than GM volume in predementia AD.

Computational state space models for activity and intention recognition. A feasibility study

Background

Computational state space models (CSSMs) enable the knowledge-based construction of Bayesian filters for recognizing intentions and reconstructing activities of human protagonists in application domains such as smart environments, assisted living, or security. Computational, i. e., algorithmic, representations allow the construction of increasingly complex human behaviour models. However, the symbolic models used in CSSMs potentially suffer from combinatorial explosion, rendering inference intractable outside of the limited experimental settings investigated in present research. The objective of this study was to obtain data on the feasibility of CSSM-based inference in domains of realistic complexity.

Methods

A typical instrumental activity of daily living was used as a trial scenario. As primary sensor modality, wearable inertial measurement units were employed. The results achievable by CSSM methods were evaluated by comparison with those obtained from established training-based methods (hidden Markov models, HMMs) using Wilcoxon signed rank tests. The influence of modeling factors on CSSM performance was analyzed via repeated measures analysis of variance.

Results

The symbolic domain model was found to have more than states, exceeding the complexity of models considered in previous research by at least three orders of magnitude. Nevertheless, if factors and procedures governing the inference process were suitably chosen, CSSMs outperformed HMMs. Specifically, inference methods used in previous studies (particle filters) were found to perform substantially inferior in comparison to a marginal filtering procedure.

Conclusions

Our results suggest that the combinatorial explosion caused by rich CSSM models does not inevitably lead to intractable inference or inferior performance. This means that the potential benefits of CSSM models (knowledge-based model construction, model reusability, reduced need for training data) are available without performance penalty. However, our results also show that research on CSSMs needs to consider sufficiently complex domains in order to understand the effects of design decisions such as choice of heuristics or inference procedure on performance.

Detecting the effect of Alzheimer's disease on everyday motion behavior

BACKGROUND

Early detection of behavioral changes in Alzheimer's disease (AD) would help the design and implementation of specific interventions.

OBJECTIVE

The target of our investigation was to establish a correlation between diagnosis and unconstrained motion behavior in subjects without major clinical behavior impairments.

METHOD

We studied everyday motion behavior in 23 dyads with one partner suffering from AD dementia and one cognitively healthy partner in the subjects' home, employing ankle-mounted three-axes accelerometric sensors. We determined frequency features obtained from the signal envelopes computed by an envelope detector for the carrier band 0.5 Hz to 5 Hz. Based on these features, we employed quadratic discriminant analysis for building models discriminating between AD patients and healthy controls.

RESULTS

After leave-one-out cross-validation, the classification accuracy of motion features reached 91% and was superior to the classification accuracy based on the Cohen-Mansfield Agitation Inventory (CMAI). Motion features were significantly correlated with MMSE and CMAI scores.

CONCLUSION

Our findings suggest that changes of everyday behavior are detectable in accelerometric behavior protocols even in the absence of major clinical behavioral impairments in AD.