Fit for Purpose Approach To Evaluate Detection of Amino Acid Substitutions in Shotgun Proteomics.

Amino acid substitutions (AASs) alter proteins from their genome-expected sequences. Accumulation of substitutions in proteins underlies numerous diseases and antibiotic mechanisms. Accurate global detection of AASs and their frequencies is crucial for understanding these mechanisms. Shotgun proteomics provides an untargeted method for measuring AASs but introduces biases when extrapolating from the genome to identify AASs. To characterize these biases, we created a "ground-truth" approach using the similarities betweenEscherichia coli and Salmonella typhimurium to model the complexity of AAS detection. Shotgun proteomics on mixed lysates generated libraries representing ∼100,000 peptide-spectra and 4161 peptide sequences with a single AAS and defined stoichiometry. Identifying S. typhimurium peptide-spectra with only the E. coli genome resulted in 64.1% correctly identified library peptides. Specific AASs exhibit variable identification efficiencies. There was no inherent bias from the stoichiometry of the substitutions. Short peptides and AASs localized near peptide termini had poor identification efficiency. We identify a new class of "scissor substitutions" that gain or lose protease cleavage sites. Scissor substitutions also had poor identification efficiency. This ground-truth AAS library reveals various sources of bias, which will guide the application of shotgun proteomics to validate AAS hypotheses.

Automatic caries detection in bitewing radiographs-Part II: experimental comparison.

OBJECTIVE: The objective of this study was to compare the detection of caries in bitewing radiographs by multiple dentists with an automatic method and to evaluate the detection performance in the absence of a reliable ground truth. MATERIALS AND METHODS: Four experts and three novices marked caries using bounding boxes in 100 bitewing radiographs. The same dataset was processed by an automatic object detection deep learning method. All annotators were compared in terms of the number of errors and intersection over union (IoU) using pairwise comparisons, with respect to the consensus standard, and with respect to the annotator of the training dataset of the automatic method. RESULTS: The number of lesions marked by experts in 100 images varied between 241 and 425. Pairwise comparisons showed that the automatic method outperformed all dentists except the original annotator in the mean number of errors, while being among the best in terms of IoU. With respect to a consensus standard, the performance of the automatic method was best in terms of the number of errors and slightly below average in terms of IoU. Compared with the original annotator, the automatic method had the highest IoU and only one expert made fewer errors. CONCLUSIONS: The automatic method consistently outperformed novices and performed as well as highly experienced dentists. CLINICAL SIGNIFICANCE: The consensus in caries detection between experts is low. An automatic method based on deep learning can improve both the accuracy and repeatability of caries detection, providing a useful second opinion even for very experienced dentists.

Picky with peakpicking: assessing chromatographic peak quality with simple metrics in metabolomics.

BACKGROUND: Chromatographic peakpicking continues to represent a significant bottleneck in automated LC-MS workflows. Uncontrolled false discovery rates and the lack of manually-calibrated quality metrics require researchers to visually evaluate individual peaks, requiring large amounts of time and breaking replicability. This problem is exacerbated in noisy environmental datasets and for novel separation methods such as hydrophilic interaction columns in metabolomics, creating a demand for a simple, intuitive, and robust metric of peak quality. RESULTS: Here, we manually labeled four HILIC oceanographic particulate metabolite datasets to assess the performance of individual peak quality metrics. We used these datasets to construct a predictive model calibrated to the likelihood that visual inspection by an MS expert would include a given mass feature in the downstream analysis. We implemented two novel peak quality metrics, a custom signal-to-noise metric and a test of similarity to a bell curve, both calculated from the raw data in the extracted ion chromatogram, and found that these outperformed existing measurements of peak quality. A simple logistic regression model built on two metrics reduced the fraction of false positives in the analysis from 70-80% down to 1-5% and showed minimal overfitting when applied to novel datasets. We then explored the implications of this quality thresholding on the conclusions obtained by the downstream analysis and found that while only 10% of the variance in the dataset could be explained by depth in the default output from the peakpicker, approximately 40% of the variance was explained when restricted to high-quality peaks alone. CONCLUSIONS: We conclude that the poor performance of peakpicking algorithms significantly reduces the power of both univariate and multivariate statistical analyses to detect environmental differences. We demonstrate that simple models built on intuitive metrics and derived from the raw data are more robust and can outperform more complex models when applied to new data. Finally, we show that in properly curated datasets, depth is a major driver of variability in the marine microbial metabolome and identify several interesting metabolite trends for future investigation.

Seeing beyond the spikes: reconstructing the complete spatiotemporal membrane potential distribution from paired intra- and extracellular recordings.

Although electrophysiologists have been recording intracellular neural activity routinely ever since the ground-breaking work of Hodgkin and Huxley, and extracellular multichannel electrodes have also been used frequently and extensively, a practical experimental method to track changes in membrane potential along a complete single neuron is still lacking. Instead of obtaining multiple intracellular measurements on the same neuron, we propose an alternative method by combining single-channel somatic patch-clamp and multichannel extracellular potential recordings. In this work, we show that it is possible to reconstruct the complete spatiotemporal distribution of the membrane potential of a single neuron with the spatial resolution of an extracellular probe during action potential generation. Moreover, the reconstruction of the membrane potential allows us to distinguish between the two major but previously hidden components of the current source density (CSD) distribution: the resistive and the capacitive currents. This distinction provides a clue to the clear interpretation of the CSD analysis, because the resistive component corresponds to transmembrane ionic currents (all the synaptic, voltage-sensitive and passive currents), whereas capacitive currents are considered to be the main contributors of counter-currents. We validate our model-based reconstruction approach on simulations and demonstrate its application to experimental data obtained in vitro via paired extracellular and intracellular recordings from a single pyramidal cell of the rat hippocampus. In perspective, the estimation of the spatial distribution of resistive membrane currents makes it possible to distiguish between active and passive sinks and sources of the CSD map and the localization of the synaptic input currents, which make the neuron fire. KEY POINTS: A new computational method is introduced to calculate the unbiased current source density distribution on a single neuron with known morphology. The relationship between extracellular and intracellular electric potential is determined via mathematical formalism, and a new reconstruction method is applied to reveal the full spatiotemporal distribution of the membrane potential and the resistive and capacitive current components. The new reconstruction method was validated on simulations. Simultaneous and colocalized whole-cell patch-clamp and multichannel silicon probe recordings were performed from the same pyramidal neuron in the rat hippocampal CA1 region, in vitro. The method was applied in experimental measurements and returned precise and distinctive characteristics of various intracellular phenomena, such as action potential generation, signal back-propagation and the initial dendritic depolarization preceding the somatic action potential.

Shapes of Hyperspectral Imaged Microplastics.

Shape matters for microplastics, but its definition, particularly for hyperspectral imaged microplastics, remains ambiguous and inexplicit, leading to incomparability across data. Hyperspectral imaging is a common approach for quantification, yet no unambiguous microplastic shape classification exists. We conducted an expert-based survey and proposed a set of clear and concise shapes (fiber, rod, ellipse, oval, sphere, quadrilateral, triangle, free-form, and unidentifiable). The categories were validated on images of 11,042 microplastics from four environmental compartments (seven matrices: indoor air; wastewater influent, effluent, and sludge; marine water; stormwater; and stormwater pond sediments), by inviting five experts to score each shape. We found that the proposed shapes were well defined, representative, and distinguishable to the human eye, especially for fiber and sphere. Ellipse, oval, and rod were though less distinguishable but dominated in all water and solid matrices. Indoor air held more unidentifiable, an abstract shape that appeared mostly for particles below 30 µm. This study highlights the need for assessing the recognizability of chosen shape categories prior to reporting data. Shapes with a clear and stringent definition would increase comparability and reproducibility across data and promote harmonization in microplastic research.

Sensing Psychological Well-being Using Social Media Language: Prediction Model Development Study.

BACKGROUND: Positive mental health is arguably increasingly important and can be revealed, to some extent, in terms of psychological well-being (PWB). However, PWB is difficult to assess in real time on a large scale. The popularity and proliferation of social media make it possible to sense and monitor online users' PWB in a nonintrusive way, and the objective of this study is to test the effectiveness of using social media language expression as a predictor of PWB. OBJECTIVE: This study aims to investigate the predictive power of social media corresponding to ground truth well-being data in a psychological way. METHODS: We recruited 1427 participants. Their well-being was evaluated using 6 dimensions of PWB. Their posts on social media were collected, and 6 psychological lexicons were used to extract linguistic features. A multiobjective prediction model was then built with the extracted linguistic features as input and PWB as the output. Further, the validity of the prediction model was confirmed by evaluating the model's discriminant validity, convergent validity, and criterion validity. The reliability of the model was also confirmed by evaluating the split-half reliability. RESULTS: The correlation coefficients between the predicted PWB scores of social media users and the actual scores obtained using the linguistic prediction model of this study were between 0.49 and 0.54 (P<.001), which means that the model had good criterion validity. In terms of the model's structural validity, it exhibited excellent convergent validity but less than satisfactory discriminant validity. The results also suggested that our model had good split-half reliability levels for every dimension (ranging from 0.65 to 0.85; P<.001). CONCLUSIONS: By confirming the availability and stability of the linguistic prediction model, this study verified the predictability of social media corresponding to ground truth well-being data from the perspective of PWB. Our study has positive implications for the use of social media to predict mental health in nonprofessional settings such as self-testing or a large-scale user study.

NAPS Fusion: A framework to overcome experimental data limitations to predict human performance and cognitive task outcomes.

In the area of human performance and cognitive research, machine learning (ML) problems become increasingly complex due to limitations in the experimental design, resulting in the development of poor predictive models. More specifically, experimental study designs produce very few data instances, have large class imbalances and conflicting ground truth labels, and generate wide data sets due to the diverse amount of sensors. From an ML perspective these problems are further exacerbated in anomaly detection cases where class imbalances occur and there are almost always more features than samples. Typically, dimensionality reduction methods (e.g., PCA, autoencoders) are utilized to handle these issues from wide data sets. However, these dimensionality reduction methods do not always map to a lower dimensional space appropriately, and they capture noise or irrelevant information. In addition, when new sensor modalities are incorporated, the entire ML paradigm has to be remodeled because of new dependencies introduced by the new information. Remodeling these ML paradigms is time-consuming and costly due to lack of modularity in the paradigm design, which is not ideal. Furthermore, human performance research experiments, at times, creates ambiguous class labels because the ground truth data cannot be agreed upon by subject-matter experts annotations, making ML paradigm nearly impossible to model. This work pulls insights from Dempster-Shafer theory (DST), stacking of ML models, and bagging to address uncertainty and ignorance for multi-classification ML problems caused by ambiguous ground truth, low samples, subject-to-subject variability, class imbalances, and wide data sets. Based on these insights, we propose a probabilistic model fusion approach, Naive Adaptive Probabilistic Sensor (NAPS), which combines ML paradigms built around bagging algorithms to overcome these experimental data concerns while maintaining a modular design for future sensor (new feature integration) and conflicting ground truth data. We demonstrate significant overall performance improvements using NAPS (an accuracy of 95.29%) in detecting human task errors (a four class problem) caused by impaired cognitive states and a negligible drop in performance with the case of ambiguous ground truth labels (an accuracy of 93.93%), when compared to other methodologies (an accuracy of 64.91%). This work potentially sets the foundation for other human-centric modeling systems that rely on human state prediction modeling.

Ground truth labels challenge the validity of sepsis consensus definitions in critical illness.

BACKGROUND: Sepsis is the leading cause of death in the intensive care unit (ICU). Expediting its diagnosis, largely determined by clinical assessment, improves survival. Predictive and explanatory modelling of sepsis in the critically ill commonly bases both outcome definition and predictions on clinical criteria for consensus definitions of sepsis, leading to circularity. As a remedy, we collected ground truth labels for sepsis. METHODS: In the Ground Truth for Sepsis Questionnaire (GTSQ), senior attending physicians in the ICU documented daily their opinion on each patient's condition regarding sepsis as a five-category working diagnosis and nine related items. Working diagnosis groups were described and compared and their SOFA-scores analyzed with a generalized linear mixed model. Agreement and discriminatory performance measures for clinical criteria of sepsis and GTSQ labels as reference class were derived. RESULTS: We analyzed 7291 questionnaires and 761 complete encounters from the first survey year. Editing rates for all items were > 90%, and responses were consistent with current understanding of critical illness pathophysiology, including sepsis pathogenesis. Interrater agreement for presence and absence of sepsis was almost perfect but only slight for suspected infection. ICU mortality was 19.5% in encounters with SIRS as the "worst" working diagnosis compared to 5.9% with sepsis and 5.9% with severe sepsis without differences in admission and maximum SOFA. Compared to sepsis, proportions of GTSQs with SIRS plus acute organ dysfunction were equal and macrocirculatory abnormalities higher (p < 0.0001). SIRS proportionally ranked above sepsis in daily assessment of illness severity (p < 0.0001). Separate analyses of neurosurgical referrals revealed similar differences. Discriminatory performance of Sepsis-1/2 and Sepsis-3 compared to GTSQ labels was similar with sensitivities around 70% and specificities 92%. Essentially no difference between the prevalence of SIRS and SOFA ≥ 2 yielded sensitivities and specificities for detecting sepsis onset close to 55% and 83%, respectively. CONCLUSIONS: GTSQ labels are a valid measure of sepsis in the ICU. They reveal suspicion of infection as an unclear clinical concept and refute an illness severity hierarchy in the SIRS-sepsis-severe sepsis spectrum. Ground truth challenges the accuracy of Sepsis-1/2 and Sepsis-3 in detecting sepsis onset. It is an indispensable intermediate step towards advancing diagnosis and therapy in the ICU and, potentially, other health care settings.

Evaluating Driver Features for Cognitive Distraction Detection and Validation in Manual and Level 2 Automated Driving.

OBJECTIVE: This study aimed to investigate the impacts of feature selection on driver cognitive distraction (CD) detection and validation in real-world nonautomated and Level 2 automated driving scenarios. BACKGROUND: Real-time driver state monitoring is critical to promote road user safety. METHOD: Twenty-four participants were recruited to drive a Tesla Model S in manual and Autopilot modes on the highway while engaging in the N-back task. In each driving mode, CD was classified by the random forest algorithm built on three "hand-crafted" glance features (i.e., percent road center [PRC], the standard deviation of gaze pitch, and yaw angles), or through a large number of features that were transformed from the output of a driver monitoring system (DMS) and other sensing systems. RESULTS: In manual driving, the small set of glance features was as effective as the large set of machine-generated features in terms of classification accuracy. Whereas in Level 2 automated driving, both glance and vehicle features were less sensitive to CD. The glance features also revealed that the misclassified driver state was the result of the dynamic fluctuations and individual differences of cognitive loads under CD. CONCLUSION: Glance metrics are critical for the detection and validation of CD in on-road driving. APPLICATIONS: The paper suggests the practical value of human factors domain knowledge in feature selection and ground truth validation for the development of driver monitoring technologies.

A low-cost robotic oculomotor simulator for assessing eye tracking accuracy in health and disease.

Eye tracking accuracy is affected in individuals with vision and oculomotor deficits, impeding our ability to answer important scientific and clinical questions about these disorders. It is difficult to disambiguate decreases in eye movement accuracy and changes in accuracy of the eye tracking itself. We propose the EyeRobot-a low-cost, robotic oculomotor simulator capable of emulating healthy and compromised eye movements to provide ground truth assessment of eye tracker performance, and how different aspects of oculomotor deficits might affect tracking accuracy and performance. The device can operate with eccentric optical axes or large deviations between the eyes, as well as simulate oculomotor pathologies, such as large fixational instabilities. We find that our design can provide accurate eye movements for both central and eccentric viewing conditions, which can be tracked by using a head-mounted eye tracker, Pupil Core. As proof of concept, we examine the effects of eccentric fixation on calibration accuracy and find that Pupil Core's existing eye tracking algorithm is robust to large fixation offsets. In addition, we demonstrate that the EyeRobot can simulate realistic eye movements like saccades and smooth pursuit that can be tracked using video-based eye tracking. These tests suggest that the EyeRobot, an easy to build and flexible tool, can aid with eye tracking validation and future algorithm development in healthy and compromised vision.

An evaluation of different approaches which use Google Street View imagery to ground truth land degradation assessments.

Member states of the United Nations Convention to Combat Desertification are required to report on the proportion of land that is degraded in their countries, a requirement that is also tied into the UN Sustainable Development Goals (SDGs). National land degradation assessments are often conducted with the use of remote sensing data which are not always ground truthed. Google Street View (GSV) provides high resolution, panoramic imagery across large parts of the world that has the potential to be used to ground truth land degradation assessments. We apply three different methodologies (visual interpretation of GSV images, GSV image classification and vegetation index extraction) to derive vegetation cover estimates from Google Street View imagery for the Hardeveld bioregion of the Succulent Karoo biome in South Africa. Visual estimates of cover best predict known habitat condition values (adjusted R2 = 0.86), whilst estimates derived from an unsupervised classification of GSV images also predict habitat condition relatively well (adjusted R2 = 0.52). These results show the potential for using GSV imagery, and other large collections of ground-level landscape photographs, as a rough ground-truthing tool, especially in instances where more traditional ground-truthing approaches are not possible.

Evaluating Localization Accuracy of Automated Driving Systems.

Automated driving systems are in need of accurate localization, i.e., achieving accuracies below 0.1 m at confidence levels above 95%. Although during the last decade numerous localization techniques have been proposed, a common methodology to validate their accuracies in relation to a ground-truth dataset is missing so far. This work aims at closing this gap by evaluating four different methods for validating localization accuracies of a vehicle's position trajectory to different ground truths: (1) a static driving-path, (2) the lane-centerline of a high-definition (HD) map with validated accuracy, (3) localized vehicle body overlaps of the lane-boundaries of a HD map, and (4) longitudinal accuracy at stop points. The methods are evaluated using two localization test datasets, one acquired by an automated vehicle following a static driving path, being additionally equipped with roof-mounted localization systems, and a second dataset acquired from manually-driven connected vehicles. Results show the broad applicability of the approach for evaluating localization accuracy and reveal the pros and cons of the different methods and ground truths. Results also show the feasibility of achieving localization accuracies below 0.1 m at confidence levels up to 99.9% for high-quality localization systems, while at the same time demonstrate that such accuracies are still challenging to achieve.

SEED-G: Simulated EEG Data Generator for Testing Connectivity Algorithms.

EEG signals are widely used to estimate brain circuits associated with specific tasks and cognitive processes. The testing of connectivity estimators is still an open issue because of the lack of a ground-truth in real data. Existing solutions such as the generation of simulated data based on a manually imposed connectivity pattern or mass oscillators can model only a few real cases with limited number of signals and spectral properties that do not reflect those of real brain activity. Furthermore, the generation of time series reproducing non-ideal and non-stationary ground-truth models is still missing. In this work, we present the SEED-G toolbox for the generation of pseudo-EEG data with imposed connectivity patterns, overcoming the existing limitations and enabling control of several parameters for data simulation according to the user's needs. We first described the toolbox including guidelines for its correct use and then we tested its performances showing how, in a wide range of conditions, datasets composed by up to 60 time series were successfully generated in less than 5 s and with spectral features similar to real data. Then, SEED-G is employed for studying the effect of inter-trial variability Partial Directed Coherence (PDC) estimates, confirming its robustness.

Autonomous Fingerprinting and Large Experimental Data Set for Visible Light Positioning.

This paper presents an autonomous method of collecting data for Visible Light Positioning (VLP) and a comprehensive investigation of VLP using a large set of experimental data. Received Signal Strength (RSS) data are efficiently collected using a novel method that utilizes consumer grade Virtual Reality (VR) tracking for accurate ground truth recording. An investigation into the accuracy of the ground truth system showed median and 90th percentile errors of 4.24 and 7.35 mm, respectively. Co-locating a VR tracker with a photodiode-equipped VLP receiver on a mobile robotic platform allows fingerprinting on a scale and accuracy that has not been possible with traditional manual collection methods. RSS data at 7344 locations within a 6.3 × 6.9 m test space fitted with 11 VLP luminaires is collected and has been made available for researchers. The quality and the volume of the data allow for a robust study of Machine Learning (ML)- and channel model-based positioning utilizing visible light. Among the ML-based techniques, ridge regression is found to be the most accurate, outperforming Weighted k Nearest Neighbor, Multilayer Perceptron, and random forest, among others. Model-based positioning is more accurate than ML techniques when a small data set is available for calibration and training. However, if a large data set is available for training, ML-based positioning outperforms its model-based counterparts in terms of localization accuracy.

Convolutional neural network for automated mass segmentation in mammography.

BACKGROUND: Automatic segmentation and localization of lesions in mammogram (MG) images are challenging even with employing advanced methods such as deep learning (DL) methods. We developed a new model based on the architecture of the semantic segmentation U-Net model to precisely segment mass lesions in MG images. The proposed end-to-end convolutional neural network (CNN) based model extracts contextual information by combining low-level and high-level features. We trained the proposed model using huge publicly available databases, (CBIS-DDSM, BCDR-01, and INbreast), and a private database from the University of Connecticut Health Center (UCHC). RESULTS: We compared the performance of the proposed model with those of the state-of-the-art DL models including the fully convolutional network (FCN), SegNet, Dilated-Net, original U-Net, and Faster R-CNN models and the conventional region growing (RG) method. The proposed Vanilla U-Net model outperforms the Faster R-CNN model significantly in terms of the runtime and the Intersection over Union metric (IOU). Training with digitized film-based and fully digitized MG images, the proposed Vanilla U-Net model achieves a mean test accuracy of 92.6%. The proposed model achieves a mean Dice coefficient index (DI) of 0.951 and a mean IOU of 0.909 that show how close the output segments are to the corresponding lesions in the ground truth maps. Data augmentation has been very effective in our experiments resulting in an increase in the mean DI and the mean IOU from 0.922 to 0.951 and 0.856 to 0.909, respectively. CONCLUSIONS: The proposed Vanilla U-Net based model can be used for precise segmentation of masses in MG images. This is because the segmentation process incorporates more multi-scale spatial context, and captures more local and global context to predict a precise pixel-wise segmentation map of an input full MG image. These detected maps can help radiologists in differentiating benign and malignant lesions depend on the lesion shapes. We show that using transfer learning, introducing augmentation, and modifying the architecture of the original model results in better performance in terms of the mean accuracy, the mean DI, and the mean IOU in detecting mass lesion compared to the other DL and the conventional models.

Ordinal labels in machine learning: a user-centered approach to improve data validity in medical settings.

BACKGROUND: Despite the vagueness and uncertainty that is intrinsic in any medical act, interpretation and decision (including acts of data reporting and representation of relevant medical conditions), still little research has focused on how to explicitly take this uncertainty into account. In this paper, we focus on the representation of a general and wide-spread medical terminology, which is grounded on a traditional and well-established convention, to represent severity of health conditions (for instance, pain, visible signs), ranging from Absent to Extreme. Specifically, we will study how both potential patients and doctors perceive the different levels of the terminology in both quantitative and qualitative terms, and if the embedded user knowledge could improve the representation of ordinal values in the construction of machine learning models. METHODS: To this aim, we conducted a questionnaire-based research study involving a relatively large sample of 1,152 potential patients and 31 clinicians to represent numerically the perceived meaning of standard and widely-applied labels to describe health conditions. Using these collected values, we then present and discuss different possible fuzzy-set based representations that address the vagueness of medical interpretation by taking into account the perceptions of domain experts. We also apply the findings of this user study to evaluate the impact of different encodings on the predictive performance of common machine learning models in regard to a real-world medical prognostic task. RESULTS: We found significant differences in the perception of pain levels between the two user groups. We also show that the proposed encodings can improve the performances of specific classes of models, and discuss when this is the case. CONCLUSIONS: In perspective, our hope is that the proposed techniques for ordinal scale representation and ordinal encoding may be useful to the research community, and also that our methodology will be applied to other widely used ordinal scales for improving validity of datasets and bettering the results of machine learning tasks.

Automatic Calibration of the Step Length Model of a Pocket INS by Means of a Foot Inertial Sensor.

All non-foot-mounted inertial localization systems have a common challenge: the need for calibrating the parameters of the step length model. The calibration of the parameters of a step length model is key for an accurate estimation of the pedestrian's step length, and therefore, for the accuracy of the position estimation. In a previous work, we provided a proof of concept on how to calibrate step length models with a foot inertial navigation system (INS), i.e., an INS based on an inertial measurement unit (IMU) mounted on the upper front part of the foot. The reason is that the foot INS does not require calibration thanks to the implementation of the strapdown algorithm. The goal of this article is to automatically calibrate the parameters of a step length model of the pocket INS by means of the foot INS. The step length model of the pocket INS has two parameters: the slope and offset of a first-order linear regression that relates the amplitude of the thigh pitch with the user's step length. Firstly, we show that it is necessary to estimate the two parameters of the step length model. Secondly, we propose a method to automatically estimate these parameters by means of a foot INS. Finally, we propose a practical implementation of the proposed method in the pocket INS. We evaluate the pocket INS with the proposed calibration method and we compare the results to the state of the art implementations of the pocket INS. The results show that the proposed automatic calibration method outperforms the previous work, which proves the need for calibrating all the parameters of the step length model of the pocket INS. In this work, we conclude that it is possible to use a foot INS to automatically calibrate all parameters of the step length model of the pocket INS. Since the calibration of the step length model is always needed, our proposed automatic calibration method is a key enabler for using the pocket INS.

Novel Multi-IMU Tight Coupling Pedestrian Localization Exploiting Biomechanical Motion Constraints.

In this article, we present a novel tight coupling inertial localization system which simultaneously processes the measurements of two inertial measurement units (IMUs) mounted on the leg, namely the upper thigh and the front part of the foot. Moreover, the proposed system exploits motion constraints of each leg link; that is, the thigh and the foot. To derive these constraints, we carry out a motion tracking experiment to collect both ground truth data and inertial measurements from IMUs mounted on the leg. The performance of the tight coupling system is assessed with a data set of approximately 10 h. The evaluation shows that the average 2D-position error of the proposed tight coupling system is at least 50% better than the average 2D-position error of two state-of-the-art systems, whereas the average height error of the tight coupling system is at least 75% better than the average height error of the two state-of-the-art systems. In this work, we improve the accuracy of the position estimation by introducing biomechanical constraints in an inertial localization system. This article allows to observe, for the first time, heading errors of an inertial localization system by using only inertial measurements and without the need for using maps or repeating totally or partially the walked trajectory.

Dimensionality Reduction of Hyperspectral Images Based on Improved Spatial-Spectral Weight Manifold Embedding.

Due to the spectral complexity and high dimensionality of hyperspectral images (HSIs), the processing of HSIs is susceptible to the curse of dimensionality. In addition, the classification results of ground truth are not ideal. To overcome the problem of the curse of dimensionality and improve classification accuracy, an improved spatial-spectral weight manifold embedding (ISS-WME) algorithm, which is based on hyperspectral data with their own manifold structure and local neighbors, is proposed in this study. The manifold structure was constructed using the structural weight matrix and the distance weight matrix. The structural weight matrix was composed of within-class and between-class coefficient representation matrices. These matrices were obtained by using the collaborative representation method. Furthermore, the distance weight matrix integrated the spatial and spectral information of HSIs. The ISS-WME algorithm describes the whole structure of the data by the weight matrix constructed by combining the within-class and between-class matrices and the spatial-spectral information of HSIs, and the nearest neighbor samples of the data are retained without changing when embedding to the low-dimensional space. To verify the classification effect of the ISS-WME algorithm, three classical data sets, namely Indian Pines, Pavia University, and Salinas scene, were subjected to experiments for this paper. Six methods of dimensionality reduction (DR) were used for comparison experiments using different classifiers such as k-nearest neighbor (KNN) and support vector machine (SVM). The experimental results show that the ISS-WME algorithm can represent the HSI structure better than other methods, and effectively improves the classification accuracy of HSIs.

A Peptide-Level Fully Annotated Data Set for Quantitative Evaluation of Precursor-Aware Mass Spectrometry Data Processing Algorithms.

Modern label-free quantitative mass spectrometry workflows are complex experimental chains for devising the composition of biological samples. With benchtop and in silico experimental steps that each have a significant effect on the accuracy, coverage, and statistical significance of the study result, it is crucial to understand the efficacy and biases of each protocol decision. Although many studies have been conducted on wet lab experimental protocols, postacquisition data processing methods have not been adequately evaluated in large part due to a lack of available ground truth data. In this study, we provide a novel ground truth data set for mass spectrometry data analysis at the precursor (MS1) signal level comprised of isolated peptide signals from UPS2, a popular complex standard for proteomics analysis, requiring more than 1000 h of manual curation. The data set consists of more than 62 million points with 1,294,008 grouped into 57,518 extracted ion chromatograms and those grouped into 14,111 isotopic envelopes. This data set can be used to evaluate many aspects of mass spectrometry data processing, including precursor mapping and signal extraction algorithms.