Recognition of activities of daily living in healthy subjects using two ad-hoc classifiers

Activities of daily living (ADL) are important for quality of life. They are indicators of cognitive health status and their assessment is a measure of independence in everyday living. ADL are difficult to reliably assess using questionnaires due to self-reporting biases. Various sensor-based (wearable, in-home, intrusive) systems have been proposed to successfully recognize and quantify ADL without relying on self-reporting. New classifiers required to classify sensor data are on the rise. We propose two ad-hoc classifiers that are based only on non-intrusive sensor data. METHODS: A wireless sensor system with ten sensor boxes was installed in the home of ten healthy subjects to collect ambient data over a duration of 20 consecutive days. A handheld protocol device and a paper logbook were also provided to the subjects. Eight ADL were selected for recognition. We developed two ad-hoc ADL classifiers, namely the rule based forward chaining inference engine (RBI) classifier and the circadian activity rhythm (CAR) classifier. The RBI classifier finds facts in data and matches them against the rules. The CAR classifier works within a framework to automatically rate routine activities to detect regular repeating patterns of behavior. For comparison, two state-of-the-art [Naïves Bayes (NB), Random Forest (RF)] classifiers have also been used. All classifiers were validated with the collected data sets for classification and recognition of the eight specific ADL. RESULTS: Out of a total of 1,373 ADL, the RBI classifier correctly determined 1,264, while missing 109 and the CAR determined 1,305 while missing 68 ADL. The RBI and CAR classifier recognized activities with an average sensitivity of 91.27 and 94.36%, respectively, outperforming both RF and NB. CONCLUSIONS: The performance of the classifiers varied significantly and shows that the classifier plays an important role in ADL recognition. Both RBI and CAR classifier performed better than existing state-of-the-art (NB, RF) on all ADL. Of the two ad-hoc classifiers, the CAR classifier was more accurate and is likely to be better suited than the RBI for distinguishing and recognizing complex ADL.

Evaluation of Three State-of-the-Art Classifiers for Recognition of Activities of Daily Living from Smart Home Ambient Data

Smart homes for the aging population have recently started attracting the attention of the research community. The "health state" of smart homes is comprised of many different levels; starting with the physical health of citizens, it also includes longer-term health norms and outcomes, as well as the arena of positive behavior changes. One of the problems of interest is to monitor the activities of daily living (ADL) of the elderly, aiming at their protection and well-being. For this purpose, we installed passive infrared (PIR) sensors to detect motion in a specific area inside a smart apartment and used them to collect a set of ADL. In a novel approach, we describe a technology that allows the ground truth collected in one smart home to train activity recognition systems for other smart homes. We asked the users to label all instances of all ADL only once and subsequently applied data mining techniques to cluster in-home sensor firings. Each cluster would therefore represent the instances of the same activity. Once the clusters were associated to their corresponding activities, our system was able to recognize future activities. To improve the activity recognition accuracy, our system preprocessed raw sensor data by identifying overlapping activities. To evaluate the recognition performance from a 200-day dataset, we implemented three different active learning classification algorithms and compared their performance: naive Bayesian (NB), support vector machine (SVM) and random forest (RF). Based on our results, the RF classifier recognized activities with an average specificity of 96.53%, a sensitivity of 68.49%, a precision of 74.41% and an F-measure of 71.33%, outperforming both the NB and SVM classifiers. Further clustering markedly improved the results of the RF classifier. An activity recognition system based on PIR sensors in conjunction with a clustering classification approach was able to detect ADL from datasets collected from different homes. Thus, our PIR-based smart home technology could improve care and provide valuable information to better understand the functioning of our societies, as well as to inform both individual and collective action in a smart city scenario.

The role of metabotropic glutamate receptor 5 in the pathogenesis of mood disorders and addiction: combining preclinical evidence with human Positron Emission Tomography (PET) studies.

In the present review, we deliver an overview of the involvement of metabotropic glutamate receptor 5 (mGluR5) activity and density in pathological anxiety, mood disorders and addiction. Specifically, we will describe mGluR5 studies in humans that employed Positron Emission Tomography (PET) and combined the findings with preclinical animal research. This combined view of different methodological approaches-from basic neurobiological approaches to human studies-might give a more comprehensive and clinically relevant view of mGluR5 function in mental health than the view on preclinical data alone. We will also review the current research data on mGluR5 along the Research Domain Criteria (RDoC). Firstly, we found evidence of abnormal glutamate activity related to the positive and negative valence systems, which would suggest that antagonistic mGluR5 intervention has prominent anti-addictive, anti-depressive and anxiolytic effects. Secondly, there is evidence that mGluR5 plays an important role in systems for social functioning and the response to social stress. Finally, mGluR5's important role in sleep homeostasis suggests that this glutamate receptor may play an important role in RDoC's arousal and modulatory systems domain. Glutamate was previously mostly investigated in non-human studies, however initial human clinical PET research now also supports the hypothesis that, by mediating brain excitability, neuroplasticity and social cognition, abnormal metabotropic glutamate activity might predispose individuals to a broad range of psychiatric problems.

Combining qualitative and quantitative methods to analyze serious games outcomes: A pilot study for a new cognitive screening tool

Computer games for a serious purpose - so called serious games can provide additional information for the screening and diagnosis of cognitive impairment. Moreover, they have the advantage of being an ecological tool by involving daily living tasks. However, there is a need for better comprehensive designs regarding the acceptance of this technology, as the target population is older adults that are not used to interact with novel technologies. Moreover given the complexity of the diagnosis and the need for precise assessment, an evaluation of the best approach to analyze the performance data is required. The present study examines the usability of a new screening tool and proposes several new outlines for data analysis.

Improving 4D plan quality for PBS-based liver tumour treatments by combining online image guided beam gating with rescanning.

Pencil beam scanned (PBS) proton therapy has many advantages over conventional radiotherapy, but its effectiveness for treating mobile tumours remains questionable. Gating dose delivery to the breathing pattern is a well-developed method in conventional radiotherapy for mitigating tumour-motion, but its clinical efficiency for PBS proton therapy is not yet well documented. In this study, the dosimetric benefits and the treatment efficiency of beam gating for PBS proton therapy has been comprehensively evaluated. A series of dedicated 4D dose calculations (4DDC) have been performed on 9 different 4DCT(MRI) liver data sets, which give realistic 4DCT extracting motion information from 4DMRI. The value of 4DCT(MRI) is its capability of providing not only patient geometries and deformable breathing characteristics, but also includes variations in the breathing patterns between breathing cycles. In order to monitor target motion and derive a gating signal, we simulate time-resolved beams' eye view (BEV) x-ray images as an online motion surrogate. 4DDCs have been performed using three amplitude-based gating window sizes (10/5/3 mm) with motion surrogates derived from either pre-implanted fiducial markers or the diaphragm. In addition, gating has also been simulated in combination with up to 19 times rescanning using either volumetric or layered approaches. The quality of the resulting 4DDC plans has been quantified in terms of the plan homogeneity index (HI), total treatment time and duty cycle. Results show that neither beam gating nor rescanning alone can fully retrieve the plan homogeneity of the static reference plan. Especially for variable breathing patterns, reductions of the effective duty cycle to as low as 10% have been observed with the smallest gating rescanning window (3 mm), implying that gating on its own for such cases would result in much longer treatment times. In addition, when rescanning is applied on its own, large differences between volumetric and layered rescanning have been observed as a function of increasing number of re-scans. However, once gating and rescanning is combined, HI to within 2% of the static plan could be achieved in the clinical target volume, with only moderately prolonged treatment times, irrespective of the rescanning strategy used. Moreover, these results are independent of the motion surrogate used. In conclusion, our results suggest image guided beam gating, combined with rescanning, is a feasible, effective and efficient motion mitigation approach for PBS-based liver tumour treatments.

A vaccine formulation combining rhoptry proteins NcROP40 and NcROP2 improves pup survival in a pregnant mouse model of neosporosis.

Currently there are no effective vaccines for the control of bovine neosporosis. During the last years several subunit vaccines based on immunodominant antigens and other proteins involved in adhesion, invasion and intracellular proliferation of Neospora caninum have been evaluated as targets for vaccine development in experimental mouse infection models. Among them, the rhoptry antigen NcROP2 and the immunodominant NcGRA7 protein have been assessed with varying results. Recent studies have shown that another rhoptry component, NcROP40, and NcNTPase, a putative dense granule antigen, exhibit higher expression levels in tachyzoites of virulent N. caninum isolates, suggesting that these could be potential vaccine candidates to limit the effects of infection. In the present work, the safety and efficacy of these recombinant antigens formulated in Quil-A adjuvant as monovalent vaccines or pair-wise combinations (rNcROP40+rNcROP2 and rNcGRA7+rNcNTPase) were evaluated in a pregnant mouse model of neosporosis. All the vaccine formulations elicited a specific immune response against their respective native proteins after immunization. Mice vaccinated with rNcROP40 and rNcROP2 alone or in combination produced the highest levels of IFN-γ and exhibited low parasite burdens and low IgG antibody levels after the challenge. In addition, most of the vaccine formulations were able to increase the median survival time in the offspring. However, pup survival only ensued in the groups vaccinated with rNcROP40+rNcROP2 (16.2%) and rNcROP2 (6.3%). Interestingly, vertical transmission was not observed in those survivor pups immunized with rNcROP40+rNcROP2, as shown by PCR analyses. These results show a partial protection against N. caninum infection after vaccination with rNcROP40+rNcROP2, suggesting a synergistic effect of the two recombinant rhoptry antigens.