Semi-supervised clustering of quaternion time series: Application to gait analysis in multiple sclerosis using motion sensor data.

Recent approaches in gait analysis involve the use of wearable motion sensors to extract spatio-temporal parameters that characterize multiple aspects of an individual's gait. In particular, the medical community could largely benefit from this type of devices as they could provide the clinicians with a valuable tool for assessing gait impairment. Motion sensor data are however complex and there is an urgent unmet need to develop sound statistical methods for analyzing such data and extracting clinically relevant information. In this article, we measure gait by following the hip rotation over time and the resulting statistical unit is a time series of unit quaternions. We explore the possibility to form groups of patients with similar walking impairment by taking into account their walking data and their global decease severity with semi-supervised clustering. We generalize a compromise-based method named hclustcompro to unit quaternion time series by combining it with the proper dissimilarity quaternion dynamic time warping. We apply this method on patients diagnosed with multiple sclerosis to form groups of patients with similar walking deficiencies while accounting for the clinical assessment of their overall disability. We also compare the compromise-based clustering approach with the method mergeTrees that falls into a sub-class of ensemble clustering named collaborative clustering. The results provide a first proof of both the interest of using wearable motion sensors for assessing gait impairment and the use of prior knowledge to guide the clustering process. It also demonstrates that compromise-based clustering is a more appropriate approach in this context.

A Novel Walking Activity Recognition Model for Rotation Time Series Collected by a Wearable Sensor in a Free-Living Environment.

Solutions to assess walking deficiencies are widespread and largely used in healthcare. Wearable sensors are particularly appealing, as they offer the possibility to monitor gait in everyday life, outside a facility in which the context of evaluation biases the measure. While some wearable sensors are powerful enough to integrate complex walking activity recognition models, non-invasive lightweight sensors do not always have the computing or memory capacity to run them. In this paper, we propose a walking activity recognition model that offers a viable solution to this problem for any wearable sensors that measure rotational motion of body parts. Specifically, the model was trained and tuned using data collected by a motion sensor in the form of a unit quaternion time series recording the hip rotation over time. This time series was then transformed into a real-valued time series of geodesic distances between consecutive quaternions. Moving average and moving standard deviation versions of this time series were fed to standard machine learning classification algorithms. To compare the different models, we used metrics to assess classification performance (precision and accuracy) while maintaining the detection prevalence at the level of the prevalence of walking activities in the data, as well as metrics to assess change point detection capability and computation time. Our results suggest that the walking activity recognition model with a decision tree classifier yields the best compromise in terms of precision and computation time. The sensor that was used had purposely low computing and memory capacity so that reported performances can be thought of as the lower bounds of what can be achieved. Walking activity recognition is performed online, i.e., on-the-fly, which further extends the range of applicability of our model to sensors with very low memory capacity.

Location of intracranial aneurysms is the main factor associated with rupture in the ICAN population.

BACKGROUND AND PURPOSE: The ever-growing availability of imaging led to increasing incidentally discovered unruptured intracranial aneurysms (UIAs). We leveraged machine-learning techniques and advanced statistical methods to provide new insights into rupture intracranial aneurysm (RIA) risks. METHODS: We analysed the characteristics of 2505 patients with intracranial aneurysms (IA) discovered between 2016 and 2019. Baseline characteristics, familial history of IA, tobacco and alcohol consumption, pharmacological treatments before the IA diagnosis, cardiovascular risk factors and comorbidities, headaches, allergy and atopy, IA location, absolute IA size and adjusted size ratio (aSR) were analysed with a multivariable logistic regression (MLR) model. A random forest (RF) method globally assessed the risk factors and evaluated the predictive capacity of a multivariate model. RESULTS: Among 994 patients with RIA (39.7%) and 1511 patients with UIA (60.3 %), the MLR showed that IA location appeared to be the most significant factor associated with RIA (OR, 95% CI: internal carotid artery, reference; middle cerebral artery, 2.72, 2.02-3.58; anterior cerebral artery, 4.99, 3.61-6.92; posterior circulation arteries, 6.05, 4.41-8.33). Size and aSR were not significant factors associated with RIA in the MLR model and antiplatelet-treatment intake patients were less likely to have RIA (OR: 0.74; 95% CI: 0.55-0.98). IA location, age, following by aSR were the best predictors of RIA using the RF model. CONCLUSIONS: The location of IA is the most consistent parameter associated with RIA. The use of 'artificial intelligence' RF helps to re-evaluate the contribution and selection of each risk factor in the multivariate model.

Testing the burden of rare variation in arrhythmia-susceptibility genes provides new insights into molecular diagnosis for Brugada syndrome.

The Brugada syndrome (BrS) is a rare heritable cardiac arrhythmia disorder associated with ventricular fibrillation and sudden cardiac death. Mutations in the SCN5A gene have been causally related to BrS in 20-30% of cases. Twenty other genes have been described as involved in BrS, but their overall contribution to disease prevalence is still unclear. This study aims to estimate the burden of rare coding variation in arrhythmia-susceptibility genes among a large group of patients with BrS. We have developed a custom kit to capture and sequence the coding regions of 45 previously reported arrhythmia-susceptibility genes and applied this kit to 167 index cases presenting with a Brugada pattern on the electrocardiogram as well as 167 individuals aged over 65-year old and showing no history of cardiac arrhythmia. By applying burden tests, a significant enrichment in rare coding variation (with a minor allele frequency below 0.1%) was observed only for SCN5A, with rare coding variants carried by 20.4% of cases with BrS versus 2.4% of control individuals (P = 1.4 × 10(-7)). No significant enrichment was observed for any other arrhythmia-susceptibility gene, including SCN10A and CACNA1C. These results indicate that, except for SCN5A, rare coding variation in previously reported arrhythmia-susceptibility genes do not contribute significantly to the occurrence of BrS in a population with European ancestry. Extreme caution should thus be taken when interpreting genetic variation in molecular diagnostic setting, since rare coding variants were observed in a similar extent among cases versus controls, for most previously reported BrS-susceptibility genes.

On Comparison of Clustering Methods for Pharmacoepidemiological Data.

The high consumption of psychotropic drugs is a public health problem. Rigorous statistical methods are needed to identify consumption characteristics in post-marketing phase. Agglomerative hierarchical clustering (AHC) and latent class analysis (LCA) can both provide clusters of subjects with similar characteristics. The objective of this study was to compare these two methods in pharmacoepidemiology, on several criteria: number of clusters, concordance, interpretation, and stability over time. From a dataset on bromazepam consumption, the two methods present a good concordance. AHC is a very stable method and it provides homogeneous classes. LCA is an inferential approach and seems to allow identifying more accurately extreme deviant behavior.

Smoothing method for unit quaternion time series in a classification problem: an application to motion data.

Smoothing orientation data is a fundamental task in different fields of research. Different methods of smoothing time series in quaternion algebras have been described in the literature, but their application is still an open point. This paper develops a smoothing approach for smoothing quaternion time series to obtain good performance in classification problems. Starting from an existing method which involves an angular velocity transformation of unit quaternion time series, a new method which employ the logarithm function to transform the quaternion time series to a real three-dimensional time series is proposed. Empirical evidences achieved on real data set and artificially noisy data sets confirm the effectiveness of the proposed method compared with the classical approach based on angular velocity transformation. The R functions developed for this paper will be provided in a Github repository.

Lead contamination in French children's homes and environment.

Lead in homes is a well-known source of childhood lead exposure, which is still of concern due to the health effects of low lead doses. This study aims to describe lead contamination in the homes of children aged 6 months to 6 years in France (without overseas). Between October 2008 and August 2009, 484 housing units were investigated. Lead in tap water and total and leachable lead levels from floor dust, outdoor soils and paint chips were measured. X-ray fluorescence measurements were carried out on non-metallic and metallic substrates. Nationwide results are provided. The indoor floor dust lead (PbD) geometric mean (GM) was 8.8 µg/m² (0.8 µg/ft²) and 6.8 µg/m² (0.6 µg/ft²) for total and leachable lead respectively; 0.21% of homes had an indoor PbD loading above 430.5 µg/m² (40 µg/ft²). The outdoor play area concentration GM was 33.5 mg/kg and 21.7 mg/kg in total and leachable lead respectively; 1.4% of concentrations were higher than or equal to 400 mg/kg. Outdoor floor PbD GM was 44.4 µg/m² (4.1 µg/ft²) that was approximately 3.2 times higher than the GM of indoor PbD. Lead-based paint (LBP) was present in 25% of dwellings, LBP on only non-metallic substrates was present in 19% of homes and on metallic substrates in 10% of dwellings. The GM of lead concentrations in tap water was below 1 µg/L; 58% of concentrations were lower than 1 µg/L and 2.9% were higher than or equal to 10 µg/L. The age cut-off for homes with lead would be 1974 for paint and 1993 for indoor floor dust. This study provides, for the first time, a look at the state of lead contamination to which children are exposed in French housing. Moreover, it provides policy makers an estimate of the number of French dwellings sheltering children where abatement should be conducted.

A non-targeted LC-HRMS approach for detecting exposure to illegal veterinary treatments: The case of cephalosporins in commercial laying Hens.

Cephalosporins are of particular importance in human medicine and should be reserved for second-line curative treatment in the veterinary field to avoid any emerging antimicrobial resistance. Due to misuse of ceftiofur in the poultry sector in France, it is now recommended to completely stop using cephalosporins in this sector. Methods currently used for the control of veterinary practices are mostly based on liquid chromatography coupled to mass spectrometry in a targeted mode, including parent compounds and any major metabolites. The aim of the present study was to evaluate the relevance of untargeted metabolomic approaches to highlight a possible exposure of laying hens to cephalosporins using a predictive model including selected treatment biomarkers. An experimentation carried out on living animals involved the administration of cefquinome and ceftiofur. Three biological matrices-droppings, eggs and liver-were investigated. Metabolites were extracted and analysed by liquid chromatography coupled to high resolution mass spectrometry in a full scan mode. Metabolites impacted by the treatment were selected by using univariate and multivariate statistical analyses. Predictive models built from the potential biomarkers selected in the "droppings" matrix were validated and able to classify "treated" and "control" hens. PLS-DA and logistic regression models were compared and both models gave satisfactory results in terms of prediction. Results were of less interest for other matrices in which only biomarkers of exposure to cefquinome were detected.

The impact of a fine-scale population stratification on rare variant association test results.

Population stratification is a well-known confounding factor in both common and rare variant association analyses. Rare variants tend to be more geographically clustered than common variants, because of their more recent origin. However, it is not yet clear if population stratification at a very fine scale (neighboring administrative regions within a country) would lead to statistical bias in rare variant analyses. As the inclusion of convenience controls from external studies is indeed a common procedure, in order to increase the power to detect genetic associations, this problem is important. We studied through simulation the impact of a fine scale population structure on different rare variant association strategies, assessing type I error and power. We showed that principal component analysis (PCA) based methods of adjustment for population stratification adequately corrected type I error inflation at the largest geographical scales, but not at finest scales. We also showed in our simulations that adding controls obviously increased power, but at a considerably lower level when controls were drawn from another population.

DoEstRare: A statistical test to identify local enrichments in rare genomic variants associated with disease.

Next-generation sequencing technologies made it possible to assay the effect of rare variants on complex diseases. As an extension of the "common disease-common variant" paradigm, rare variant studies are necessary to get a more complete insight into the genetic architecture of human traits. Association studies of these rare variations show new challenges in terms of statistical analysis. Due to their low frequency, rare variants must be tested by groups. This approach is then hindered by the fact that an unknown proportion of the variants could be neutral. The risk level of a rare variation may be determined by its impact but also by its position in the protein sequence. More generally, the molecular mechanisms underlying the disease architecture may involve specific protein domains or inter-genic regulatory regions. While a large variety of methods are optimizing functionality weights for each single marker, few evaluate variant position differences between cases and controls. Here, we propose a test called DoEstRare, which aims to simultaneously detect clusters of disease risk variants and global allele frequency differences in genomic regions. This test estimates, for cases and controls, variant position densities in the genetic region by a kernel method, weighted by a function of allele frequencies. We compared DoEstRare with previously published strategies through simulation studies as well as re-analysis of real datasets. Based on simulation under various scenarios, DoEstRare was the sole to consistently show highest performance, in terms of type I error and power both when variants were clustered or not. DoEstRare was also applied to Brugada syndrome and early-onset Alzheimer's disease data and provided complementary results to other existing tests. DoEstRare, by integrating variant position information, gives new opportunities to explain disease susceptibility. DoEstRare is implemented in a user-friendly R package.

Source contributions of lead in residential floor dust and within-home variability of dust lead loading.

Evidence of the impact of exposure to low levels of lead on children's health is increasing. Residential floor dust is the assumed origin of lead exposure by young children. In this study, we estimate the contribution of different lead sources to household interior floor dust contamination. We also estimate the within-home variability of interior floor dust lead loadings. A multilevel model was developed based on data collected in a French survey in 2008-2009 (484 housing units, 1834 rooms). Missing data were handled by multiple imputation using chained equations. The intra-home correlation between interior floor Log dust lead loadings was approximately 0.6. Dust lead from the landing of an apartment, mostly originating outside the building, was the major contributor to interior floor dust lead. Secondary contributors included the lead-based paint on exterior railings, track-in of the exterior soil of the children's play area into the dwelling, smoking inside the home, demolition of nearby old buildings and sites of pollution in the vicinity. Interior lead-based paint contaminated interior floor dust only in old and non-renovated dwellings. To reduce interior floor dust lead levels in the general population of dwellings, common areas should be maintained, and track-in from the outside should be limited as much as possible.