Micapipe: A pipeline for multimodal neuroimaging and connectome analysis

Multimodal magnetic resonance imaging (MRI) has accelerated human neuroscience by fostering the analysis of brain microstructure, geometry, function, and connectivity across multiple scales and in living brains. The richness and complexity of multimodal neuroimaging, however, demands processing methods to integrate information across modalities and to consolidate findings across different spatial scales. Here, we present micapipe, an open processing pipeline for multimodal MRI datasets. Based on BIDS-conform input data, micapipe can generate i) structural connectomes derived from diffusion tractography, ii) functional connectomes derived from resting-state signal correlations, iii) geodesic distance matrices that quantify cortico-cortical proximity, and iv) microstructural profile covariance matrices that assess inter-regional similarity in cortical myelin proxies. The above matrices can be automatically generated across established 18 cortical parcellations (100-1000 parcels), in addition to subcortical and cerebellar parcellations, allowing researchers to replicate findings easily across different spatial scales. Results are represented on three different surface spaces (native, conte69, fsaverage5), and outputs are BIDS-conform. Processed outputs can be quality controlled at the individual and group level. micapipe was tested on several datasets and is available at https://github.com/MICA-MNI/micapipe, documented at https://micapipe.readthedocs.io/, and containerized as a BIDS App http://bids-apps.neuroimaging.io/apps/. We hope that micapipe will foster robust and integrative studies of human brain microstructure, morphology, function, cand connectivity.

Performance of QR Code Detectors near Nyquist Limits

For the interacting with real world, augmented reality devices need lightweight yet reliable methods for recognition and identification of physical objects. In that regard, promising possibilities are offered by supporting computer vision with 2D barcode tags. These tags, as high contrast and visually well-defined objects, can be used for finding fiducial points in the space or to identify physical items. Currently, QR code readers have certain demands towards the size and visibility of the codes. However, the increase of resolution of built-in cameras makes it possible to identify smaller QR codes in the scene. On the other hand, growing resolutions cause the increase to the computational effort of tag location. Therefore, resolution reduction in decoders is a common trade-off between processing time and recognition capabilities. In this article, we propose the simulation method of QR codes scanning near limits that stem from Shannon's theorem. We analyze the efficiency of three publicly available decoders versus different size-to-sampling ratios (scales) and MTF characteristics of the image capture subsystem. The MTF we used is based on the characteristics of real devices, and it was modeled using Gaussian low-pass filtering. We tested two tasks-decoding and locating-and-decoding. The findings of the work are several-fold. Among others, we identified that, for practical decoding, the QR-code module should be no smaller than 3-3.5 pixels, regardless of MTF characteristics. We confirmed the superiority of Zbar in practical tasks and the worst recognition capabilities of OpenCV. On the other hand, we identified that, for borderline cases, or even below Nyquist limit where the other decoders fail, OpenCV is still capable of decoding some information.

Mindful Workarounds in Bar Code Medication Administration

Bar-Coded Medication Administration systems (BCMA) are often used with workarounds. These workarounds are usually judged against standard operating procedures or the use of the technology as 'designers' intended'. However, some workarounds may be reasonable and justified to prevent safety errors. In this conceptual paper, we clarify BCMA safety mechanisms and provide a framework to identify workarounds with BCMA that nullify the error prevention mechanisms inherent in the technology design and process. We also highlight the importance of understanding the purpose behind a nurse's workaround in BCMA, focusing on the notion of mindful (thoughtful) workarounds that have the potential to improve patient safety.

Pea-KD: Parameter-efficient and accurate Knowledge Distillation on BERT

Knowledge Distillation (KD) is one of the widely known methods for model compression. In essence, KD trains a smaller student model based on a larger teacher model and tries to retain the teacher model’s level of performance as much as possible. However, existing KD methods suffer from the following limitations. First, since the student model is smaller in absolute size, it inherently lacks model capacity. Second, the absence of an initial guide for the student model makes it difficult for the student to imitate the teacher model to its fullest. Conventional KD methods yield low performance due to these limitations. In this paper, we propose Pea-KD (Parameter-efficient and accurate Knowledge Distillation), a novel approach to KD. Pea-KD consists of two main parts: Shuffled Parameter Sharing (SPS) and Pretraining with Teacher’s Predictions (PTP). Using this combination, we are capable of alleviating the KD’s limitations. SPS is a new parameter sharing method that increases the student model capacity. PTP is a KD-specialized initialization method, which can act as a good initial guide for the student. When combined, this method yields a significant increase in student model’s performance. Experiments conducted on BERT with different datasets and tasks show that the proposed approach improves the student model’s performance by 4.4% on average in four GLUE tasks, outperforming existing KD baselines by significant margins.

A data flow process for confidential data and its application in a health research project

BACKGROUND

The use of linked healthcare data in research has the potential to make major contributions to knowledge generation and service improvement. However, using healthcare data for secondary purposes raises legal and ethical concerns relating to confidentiality, privacy and data protection rights. Using a linkage and anonymisation approach that processes data lawfully and in line with ethical best practice to create an anonymous (non-personal) dataset can address these concerns, yet there is no set approach for defining all of the steps involved in such data flow end-to-end. We aimed to define such an approach with clear steps for dataset creation, and to describe its utilisation in a case study linking healthcare data.

METHODS

We developed a data flow protocol that generates pseudonymous datasets that can be reversibly linked, or irreversibly linked to form an anonymous research dataset. It was designed and implemented by the Comprehensive Patient Records (CPR) study in Leeds, UK.

RESULTS

We defined a clear approach that received ethico-legal approval for use in creating an anonymous research dataset. Our approach used individual-level linkage through a mechanism that is not computer-intensive and was rendered irreversible to both data providers and processors. We successfully applied it in the CPR study to hospital and general practice and community electronic health record data from two providers, along with patient reported outcomes, for 365,193 patients. The resultant anonymous research dataset is available via DATA-CAN, the Health Data Research Hub for Cancer in the UK.

CONCLUSIONS

Through ethical, legal and academic review, we believe that we contribute a defined approach that represents a framework that exceeds current minimum standards for effective pseudonymisation and anonymisation. This paper describes our methods and provides supporting information to facilitate the use of this approach in research.

A New Workflow for Drug Metabolite Profiling by Utilizing Advanced Tribrid Mass Spectrometry and Data-Processing Techniques

Drug metabolite profiling utilizes liquid chromatography with tandem mass spectrometry (LC/MS/MS) to acquire ample information for metabolite identification and structural elucidation. However, there are still challenges in detecting and characterizing all potential metabolites that can be masked by a high biological background, especially the unknown and uncommon ones. In this work, a novel metabolite profiling workflow was established on a platform using a state-of-the-art tribrid high-resolution mass spectrometry (HRMS) system. Primarily, an instrumental method was developed based on the novel design of the tribrid system that facilitates in-depth MSⁿ scans with two fragmentation devices. Additionally, different advanced data acquisition techniques were assessed and compared, and automatic background exclusion and deep-scan approaches were adopted to promote assay efficiency and metabolite coverage. Finally, different data-analysis techniques were explored to fully extract metabolite data from the information-rich MS/MS data sets. Overall, a workflow combining tribrid mass spectrometry and advanced acquisition methodology has been developed for metabolite characterization in drug discovery and development. It maximizes the tribrid HRMS platform's utility and enhances the coverage, efficiency, quality, and speed of metabolite profiling assays.

An automatic screening method for strabismus detection based on image processing

Purpose

This study aims to provide an automatic strabismus screening method for people who live in remote areas with poor medical accessibility.

Materials and methods

The proposed method first utilizes a pretrained convolutional neural network-based face-detection model and a detector for 68 facial landmarks to extract the eye region for a frontal facial image. Second, Otsu’s binarization and the HSV color model are applied to the image to eliminate the influence of eyelashes and canthi. Then, the method samples all of the pixel points on the limbus and applies the least square method to obtain the coordinate of the pupil center. Lastly, we calculated the distances from the pupil center to the medial and lateral canthus to measure the deviation of the positional similarity of two eyes for strabismus screening.

Result

We used a total of 60 frontal facial images (30 strabismus images, 30 normal images) to validate the proposed method. The average value of the iris positional similarity of normal images was smaller than one of the strabismus images via the method (p-value<0.001). The sample mean and sample standard deviation of the positional similarity of the normal and strabismus images were 1.073 ± 0.014 and 0.039, as well as 1.924 ± 0.169 and 0.472, respectively.

Conclusion

The experimental results of 60 images show that the proposed method is a promising automatic strabismus screening method for people living in remote areas with poor medical accessibility.

High-speed serial deep learning through temporal optical neurons

Deep learning is able to functionally mimic the human brain and thus, it has attracted considerable recent interest. Optics-assisted deep learning is a promising approach to improve forward-propagation speed and reduce the power consumption of electronic-assisted techniques. However, present methods are based on a parallel processing approach that is inherently ineffective in dealing with the serial data signals at the core of information and communication technologies. Here, we propose and demonstrate a sequential optical deep learning concept that is specifically designed to directly process high-speed serial data. By utilizing ultra-short optical pulses as the information carriers, the neurons are distributed at different time slots in a serial pattern, and interconnected to each other through group delay dispersion. A 4-layer serial optical neural network (SONN) was constructed and trained for classification of both analog and digital signals with simulated accuracy rates of over 79.2% with proper individuality variance rates. Furthermore, we performed a proof-of-concept experiment of a pseudo-3-layer SONN to successfully recognize the ASCII codes of English letters at a data rate of 12 gigabits per second. This concept represents a novel one-dimensional realization of artificial neural networks, enabling a direct application of optical deep learning methods to the analysis and processing of serial data signals, while offering a new overall perspective for temporal signal processing.

Integrating barcoded neuroanatomy with spatial transcriptional profiling enables identification of gene correlates of projections

Functional circuits consist of neurons with diverse axonal projections and gene expression. Understanding the molecular signature of projections requires high-throughput interrogation of both gene expression and projections to multiple targets in the same cells at cellular resolution, which is difficult to achieve using current technology. Here, we introduce BARseq2, a technique that simultaneously maps projections and detects multiplexed gene expression by in situ sequencing. We determined the expression of cadherins and cell-type markers in 29,933 cells and the projections of 3,164 cells in both the mouse motor cortex and auditory cortex. Associating gene expression and projections in 1,349 neurons revealed shared cadherin signatures of homologous projections across the two cortical areas. These cadherins were enriched across multiple branches of the transcriptomic taxonomy. By correlating multigene expression and projections to many targets in single neurons with high throughput, BARseq2 provides a potential path to uncovering the molecular logic underlying neuronal circuits.

Consumer-Based Activity Trackers as a Tool for Physical Activity Monitoring in Epidemiological Studies During the COVID-19 Pandemic: Development and Usability Study

BACKGROUND

Consumer-based physical activity trackers have increased in popularity. The widespread use of these devices and the long-term nature of the recorded data provides a valuable source of physical activity data for epidemiological research. The challenges include the large heterogeneity between activity tracker models in terms of available data types, the accuracy of recorded data, and how this data can be shared between different providers and third-party systems.

OBJECTIVE

The aim of this study is to develop a system to record data on physical activity from different providers of consumer-based activity trackers and to examine its usability as a tool for physical activity monitoring in epidemiological research. The longitudinal nature of the data and the concurrent pandemic outbreak allowed us to show how the system can be used for surveillance of physical activity levels before, during, and after a COVID-19 lockdown.

METHODS

We developed a system (mSpider) for automatic recording of data on physical activity from participants wearing activity trackers from Apple, Fitbit, Garmin, Oura, Polar, Samsung, and Withings, as well as trackers storing data in Google Fit and Apple Health. To test the system throughout development, we recruited 35 volunteers to wear a provided activity tracker from early 2019 and onward. In addition, we recruited 113 participants with privately owned activity trackers worn before, during, and after the COVID-19 lockdown in Norway. We examined monthly changes in the number of steps, minutes of moderate-to-vigorous physical activity, and activity energy expenditure between 2019 and 2020 using bar plots and two-sided paired sample t tests and Wilcoxon signed-rank tests.

RESULTS

Compared to March 2019, there was a significant reduction in mean step count and mean activity energy expenditure during the March 2020 lockdown period. The reduction in steps and activity energy expenditure was temporary, and the following monthly comparisons showed no significant change between 2019 and 2020. A small significant increase in moderate-to-vigorous physical activity was observed for several monthly comparisons after the lockdown period and when comparing March-December 2019 with March-December 2020.

CONCLUSIONS

mSpider is a working prototype currently able to record physical activity data from providers of consumer-based activity trackers. The system was successfully used to examine changes in physical activity levels during the COVID-19 period.

Automatic image annotation for fluorescent cell nuclei segmentation

Dataset annotation is a time and labor-intensive task and an integral requirement for training and testing deep learning models. The segmentation of images in life science microscopy requires annotated image datasets for object detection tasks such as instance segmentation. Although the amount of annotated image data has been steadily reduced due to methods such as data augmentation, the process of manual or semi-automated data annotation is the most labor and cost intensive task in the process of cell nuclei segmentation with deep neural networks. In this work we propose a system to fully automate the annotation process of a custom fluorescent cell nuclei image dataset. By that we are able to reduce nuclei labelling time by up to 99.5%. The output of our system provides high quality training data for machine learning applications to identify the position of cell nuclei in microscopy images. Our experiments have shown that the automatically annotated dataset provides coequal segmentation performance compared to manual data annotation. In addition, we show that our system enables a single workflow from raw data input to desired nuclei segmentation and tracking results without relying on pre-trained models or third-party training datasets for neural networks.

Automated system for diagnosing endometrial cancer by adopting deep-learning technology in hysteroscopy

Endometrial cancer is a ubiquitous gynecological disease with increasing global incidence. Therefore, despite the lack of an established screening technique to date, early diagnosis of endometrial cancer assumes critical importance. This paper presents an artificial-intelligence-based system to detect the regions affected by endometrial cancer automatically from hysteroscopic images. In this study, 177 patients (60 with normal endometrium, 21 with uterine myoma, 60 with endometrial polyp, 15 with atypical endometrial hyperplasia, and 21 with endometrial cancer) with a history of hysteroscopy were recruited. Machine-learning techniques based on three popular deep neural network models were employed, and a continuity-analysis method was developed to enhance the accuracy of cancer diagnosis. Finally, we investigated if the accuracy could be improved by combining all the trained models. The results reveal that the diagnosis accuracy was approximately 80% (78.91–80.93%) when using the standard method, and it increased to 89% (83.94–89.13%) and exceeded 90% (i.e., 90.29%) when employing the proposed continuity analysis and combining the three neural networks, respectively. The corresponding sensitivity and specificity equaled 91.66% and 89.36%, respectively. These findings demonstrate the proposed method to be sufficient to facilitate timely diagnosis of endometrial cancer in the near future.

An Observational Study in Manipur State, India on Preventive Behavior Influenced by Social Media During the COVID-19 Pandemic Mediated by Cyberchondria and Information Overload

OBJECTIVES

The coronavirus disease 2019 (COVID-19) pandemic is a public health emergency posing unprecedented challenges for health authorities. Social media may serve as an effective platform to disseminate health-related information. This study aimed to assess the extent of social media use, its impact on preventive behavior, and negative health effects such as cyberchondria and information overload.

METHODS

A cross-sectional observational study was conducted between June 10, 2020 and August 9, 2020 among people visiting the outpatient department of the authors' institution, and participants were also recruited during field visits for an awareness drive. Questions were developed on preventive behavior, and the Short Cyberchondria Scale and instruments dealing with information overload and perceived vulnerability were used.

RESULTS

The study recruited 767 participants with a mean age of about 45 years. Most of the participants (>90%) engaged in preventive behaviors, which were influenced by the extent of information received through social media platforms (β=3.297; p<0.001) and awareness of infection when a family member tested positive (β=29.082; p<0.001) or a neighbor tested positive (β=27.964; p<0.001). The majority (63.0%) of individuals often searched for COVID-19 related news on social media platforms. The mean±standard deviation scores for cyberchondria and information overload were 9.09±4.05 and 8.69±2.56, respectively. Significant and moderately strong correlations were found between cyberchondria, information overload, and perceived vulnerability to COVID-19.

CONCLUSIONS

This study provides evidence that the use of social media as an information- seeking platform altered preventive behavior. However, excessive and misleading information resulted in cyberchondria and information overload.

Implementation of convolutional neural network approach for COVID-19 disease detection

In this paper, two novel, powerful, and robust convolutional neural network (CNN) architectures are designed and proposed for two different classification tasks using publicly available data sets. The first architecture is able to decide whether a given chest X-ray image of a patient contains COVID-19 or not with 98.92% average accuracy. The second CNN architecture is able to divide a given chest X-ray image of a patient into three classes (COVID-19 versus normal versus pneumonia) with 98.27% average accuracy. The hyperparameters of both CNN models are automatically determined using Grid Search. Experimental results on large clinical data sets show the effectiveness of the proposed architectures and demonstrate that the proposed algorithms can overcome the disadvantages mentioned above. Moreover, the proposed CNN models are fully automatic in terms of not requiring the extraction of diseased tissue, which is a great improvement of available automatic methods in the literature. To the best of the author's knowledge, this study is the first study to detect COVID-19 disease from given chest X-ray images, using CNN, whose hyperparameters are automatically determined by the Grid Search. Another important contribution of this study is that it is the first CNN-based COVID-19 chest X-ray image classification study that uses the largest possible clinical data set. A total of 1,524 COVID-19, 1,527 pneumonia, and 1524 normal X-ray images are collected. It is aimed to collect the largest number of COVID-19 X-ray images that exist in the literature until the writing of this research paper.

Progressive tracking: a novel procedure to facilitate manual digitization of videos

Digitization of video recordings often requires the laborious procedure of manually clicking points of interest on individual video frames. Here, we present progressive tracking, a procedure that facilitates manual digitization of markerless videos. In contrast to existing software, it allows the user to follow points of interest with a cursor in the progressing video, without the need to click. To compare the performance of progressive tracking with the conventional frame-wise tracking, we quantified speed and accuracy of both methods, testing two different input devices (mouse and stylus pen). We show that progressive tracking can be twice as fast as frame-wise tracking while maintaining accuracy, given that playback speed is controlled. Using a stylus pen can increase frame-wise tracking speed. The complementary application of the progressive and frame-wise mode is exemplified on a realistic video recording. This study reveals that progressive tracking can vastly facilitate video analysis in experimental research.

ComMSnDB-An Automatable Strategy to Identify Compounds from MS Data Sets (Identification of Gypenosides as an Example)

Gynostemma pentaphyllum (Thunb.) Makino is a popular functional food and is also used as an important medicinal plant in China. Gypenoside, the main active constituent in G. pentaphyllum (Thunb.) Makino, belongs to dammarane-type triterpenoid saponins. Due to its high molecular weight and high polarity, it is difficult to obtain complete compound information for gypenoside extracts via mass spectrometry experiments. In this study, an automated targeted data postprocessing strategy called Compound MSⁿ Database (ComMSⁿDB) was designed and established to elucidate compounds in gypenoside extracts based on ultrahigh-performance liquid chromatography coupled with electrospray ionization quadrupole time-of-flight tandem mass spectrometry (UHPLC-ESI-Q-TOF-MS/MS). As a result, 18 types of and 199 main saponin constituents, including 47 potential novel compounds, were tentatively identified from different habitats. At the same time, 15 gypenoside standard compounds were used to verify the feasibility of the ComMSⁿDB strategy. These results demonstrated that ComMSⁿDB offers practical value for quick, automated, and effective compound identification.

Optimized backoff scheme for prioritized data in wireless sensor networks: A class of service approach

Data prioritization of heterogeneous data in wireless sensor networks gives meaning to mission-critical data that are time-sensitive as this may be a matter of life and death. However, the standard IEEE 802.15.4 does not consider the prioritization of data. Prioritization schemes proffered in the literature have not adequately addressed this issue as proposed schemes either uses a single or complex backoff algorithm to estimate backoff time-slots for prioritized data. Subsequently, the carrier sense multiple access with collision avoidance scheme exhibits an exponentially increasing range of backoff times. These approaches are not only inefficient but result in high latency and increased power consumption. In this article, the concept of class of service (CS) was adopted to prioritize heterogeneous data (real-time and non-real-time), resulting in an optimized prioritized backoff MAC scheme called Class of Service Traffic Priority-based Medium Access Control (CSTP-MAC). This scheme classifies data into high priority data (HPD) and low priority data (LPD) by computing backoff times with expressions peculiar to the data priority class. The improved scheme grants nodes the opportunity to access the shared medium in a timely and power-efficient manner. Benchmarked against contemporary schemes, CSTP-MAC attained a 99% packet delivery ratio with improved power saving capability, which translates to a longer operational lifetime.

Minimum error correction-based haplotype assembly: Considerations for long read data

The single nucleotide polymorphism (SNP) is the most widely studied type of genetic variation. A haplotype is defined as the sequence of alleles at SNP sites on each haploid chromosome. Haplotype information is essential in unravelling the genome-phenotype association. Haplotype assembly is a well-known approach for reconstructing haplotypes, exploiting reads generated by DNA sequencing devices. The Minimum Error Correction (MEC) metric is often used for reconstruction of haplotypes from reads. However, problems with the MEC metric have been reported. Here, we investigate the MEC approach to demonstrate that it may result in incorrectly reconstructed haplotypes for devices that produce error-prone long reads. Specifically, we evaluate this approach for devices developed by Illumina, Pacific BioSciences and Oxford Nanopore Technologies. We show that imprecise haplotypes may be reconstructed with a lower MEC than that of the exact haplotype. The performance of MEC is explored for different coverage levels and error rates of data. Our simulation results reveal that in order to avoid incorrect MEC-based haplotypes, a coverage of 25 is needed for reads generated by Pacific BioSciences RS systems.

A new method for computing the projection median, its influence curve and techniques for the production of projected quantile plots

This article introduces a new formulation of, and method of computation for, the projection median. Additionally, we explore its behaviour on a specific bivariate set up, providing the first theoretical result on form of the influence curve for the projection median, accompanied by numerical simulations. Via new simulations we comprehensively compare our performance with an established method for computing the projection median, as well as other existing multivariate medians. We focus on answering questions about accuracy and computational speed, whilst taking into account the underlying dimensionality. Such considerations are vitally important in situations where the data set is large, or where the operations have to be repeated many times and some well-known techniques are extremely computationally expensive. We briefly describe our associated R package that includes our new methods and novel functionality to produce animated multidimensional projection quantile plots, and also exhibit its use on some high-dimensional data examples.

The Genomics Research and Innovation Network: creating an interoperable, federated, genomics learning system

PURPOSE

Clinicians and researchers must contextualize a patient's genetic variants against population-based references with detailed phenotyping. We sought to establish globally scalable technology, policy, and procedures for sharing biosamples and associated genomic and phenotypic data on broadly consented cohorts, across sites of care.

METHODS

Three of the nation's leading children's hospitals launched the Genomic Research and Innovation Network (GRIN), with federated information technology infrastructure, harmonized biobanking protocols, and material transfer agreements. Pilot studies in epilepsy and short stature were completed to design and test the collaboration model.

RESULTS

Harmonized, broadly consented institutional review board (IRB) protocols were approved and used for biobank enrollment, creating ever-expanding, compatible biobanks. An open source federated query infrastructure was established over genotype-phenotype databases at the three hospitals. Investigators securely access the GRIN platform for prep to research queries, receiving aggregate counts of patients with particular phenotypes or genotypes in each biobank. With proper approvals, de-identified data is exported to a shared analytic workspace. Investigators at all sites enthusiastically collaborated on the pilot studies, resulting in multiple publications. Investigators have also begun to successfully utilize the infrastructure for grant applications.

CONCLUSIONS

The GRIN collaboration establishes the technology, policy, and procedures for a scalable genomic research network.

Exploring the promise of broadband fisheries echosounders for species discrimination with quantitative assessment of data processing effects

It remains an open question how well the increased bandwidth afforded by broadband echosounders can improve species discrimination in fisheries acoustics. Here, an objective statistical approach was used to determine if there is information available in dual channel broadband data (45-170 kHz) to allow discrimination between in situ echoes obtained from monospecific aggregations of three species (hake, Merluccius productus: anchovy, Engraulis mordax; and krill, Euphausiia pacifica) using a remotely operated vehicle. These data were used to explore the effects of processing choices on the ability to statistically classify the broadband spectra to species. This ability was affected by processing choices including the Fourier transform analysis window size, available bandwidth, and the method and scale of data averaging. The approach to normalizing the spectra and the position of individual targets in the beam, however, had little effect. Broadband volume backscatter and single target spectra were both used to successfully classify acoustic data from these species with ∼6% greater success using volume backscatter data. Broadband data were effectively classified to species while simulated multi-frequency narrowband data were categorized at rates near chance, supporting the presumption that greater bandwidth increases the information available for the characterization and classification of biological targets.

Comparison of quality control methods for automated diffusion tensor imaging analysis pipelines

The processing of brain diffusion tensor imaging (DTI) data for large cohort studies requires fully automatic pipelines to perform quality control (QC) and artifact/outlier removal procedures on the raw DTI data prior to calculation of diffusion parameters. In this study, three automatic DTI processing pipelines, each complying with the general ENIGMA framework, were designed by uniquely combining multiple image processing software tools. Different QC procedures based on the RESTORE algorithm, the DTIPrep protocol, and a combination of both methods were compared using simulated ground truth and artifact containing DTI datasets modeling eddy current induced distortions, various levels of motion artifacts, and thermal noise. Variability was also examined in 20 DTI datasets acquired in subjects with vascular cognitive impairment (VCI) from the multi-site Ontario Neurodegenerative Disease Research Initiative (ONDRI). The mean fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD) were calculated in global brain grey matter (GM) and white matter (WM) regions. For the simulated DTI datasets, the measure used to evaluate the performance of the pipelines was the normalized difference between the mean DTI metrics measured in GM and WM regions and the corresponding ground truth DTI value. The performance of the proposed pipelines was very similar, particularly in FA measurements. However, the pipeline based on the RESTORE algorithm was the most accurate when analyzing the artifact containing DTI datasets. The pipeline that combined the DTIPrep protocol and the RESTORE algorithm produced the lowest standard deviation in FA measurements in normal appearing WM across subjects. We concluded that this pipeline was the most robust and is preferred for automated analysis of multisite brain DTI data.

Arteria: An automation system for a sequencing core facility

BACKGROUND

In recent years, nucleotide sequencing has become increasingly instrumental in both research and clinical settings. This has led to an explosive growth in sequencing data produced worldwide. As the amount of data increases, so does the need for automated solutions for data processing and analysis. The concept of workflows has gained favour in the bioinformatics community, but there is little in the scientific literature describing end-to-end automation systems. Arteria is an automation system that aims at providing a solution to the data-related operational challenges that face sequencing core facilities.

FINDINGS

Arteria is built on existing open source technologies, with a modular design allowing for a community-driven effort to create plug-and-play micro-services. In this article we describe the system, elaborate on the underlying conceptual framework, and present an example implementation. Arteria can be reduced to 3 conceptual levels: orchestration (using an event-based model of automation), process (the steps involved in processing sequencing data, modelled as workflows), and execution (using a series of RESTful micro-services). This creates a system that is both flexible and scalable. Arteria-based systems have been successfully deployed at 3 sequencing core facilities. The Arteria Project code, written largely in Python, is available as open source software, and more information can be found at https://arteria-project.github.io/ .

CONCLUSIONS

We describe the Arteria system and the underlying conceptual framework, demonstrating how this model can be used to automate data handling and analysis in the context of a sequencing core facility.

A systematic evaluation of single cell RNA-seq analysis pipelines

The recent rapid spread of single cell RNA sequencing (scRNA-seq) methods has created a large variety of experimental and computational pipelines for which best practices have not yet been established. Here, we use simulations based on five scRNA-seq library protocols in combination with nine realistic differential expression (DE) setups to systematically evaluate three mapping, four imputation, seven normalisation and four differential expression testing approaches resulting in ~3000 pipelines, allowing us to also assess interactions among pipeline steps. We find that choices of normalisation and library preparation protocols have the biggest impact on scRNA-seq analyses. Specifically, we find that library preparation determines the ability to detect symmetric expression differences, while normalisation dominates pipeline performance in asymmetric DE-setups. Finally, we illustrate the importance of informed choices by showing that a good scRNA-seq pipeline can have the same impact on detecting a biological signal as quadrupling the sample size.

A Roadmap for Automatic Surgical Site Infection Detection and Evaluation Using User-Generated Incision Images

Background: Emerging technologies such as smartphones and wearable sensors have enabled the paradigm shift to new patient-centered healthcare, together with recent mobile health (mHealth) app development. One such promising healthcare app is incision monitoring based on patient-taken incision images. In this review, challenges and potential solution strategies are investigated for surgical site infection (SSI) detection and evaluation using surgical site images taken at home. Methods: Potential image quality issues, feature extraction, and surgical site image analysis challenges are discussed. Recent image analysis and machine learning solutions are reviewed to extract meaningful representations as image markers for incision monitoring. Discussions on opportunities and challenges of applying these methods to derive accurate SSI prediction are provided. Conclusions: Interactive image acquisition as well as customized image analysis and machine learning methods for SSI monitoring will play critical roles in developing sustainable mHealth apps to achieve the expected outcomes of patient-taken incision images for effective out-of-clinic patient-centered healthcare with substantially reduced cost.

Development of a Biobank from a Legacy Collection in Universitas Gadjah Mada, Indonesia: Proposed Approach for Centralized Biobank Development in Low-Resource Institutions

Introduction: The establishment of a biobank requires specific expertise along with relatively expensive infrastructure and appropriate technology. This causes certain challenges in biobank implementation for research in low-middle-income countries. Biobank development with established specimens and data collection (legacy collection) was an approach used in the Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada. This approach aimed to identify the resources available at present, while providing nontechnical information for further development of a centralized biobank. Materials and Methods: Retrospective modeling was done in 2015 by recruiting existing specimen collections and their associated data. The steps were as follows: (1) informing research stakeholders through discussion with experts and stakeholders; (2) identifying specimen collections to be used; (3) determining the system, infrastructure, and consumables needed; (4) determining inclusion criteria; (5) building an in-house database system; (6) organizing data and physical specimen collections; and (7) validating data and physical sample arrangement. All technical procedures were built into standard operating procedures. Results: The model included specimens from one -80°C freezer. The associated data included demographic, clinical diagnosis, and physical sample information. Samples came from six studies, collected between 2001 and 2014. A web-based database was built based on the MySQL programming system. Information on biospecimens from a total of 4196 subjects collected in 11,358 vials was entered into the database, following physical rearrangement of vials in the -80°C freezer with one-dimensional barcodes taped to vials, boxes, and racks. A validation test was done for data concordance between the database and physical arrangement in the -80°C freezer, showing no discrepancies. Conclusion: This report demonstrated current technical and nontechnical insights to further develop a centralized biobank for health research at an academic institution in Indonesia.

Data Processing Pipeline for Lipid Profiling of Carotid Atherosclerotic Plaque with Mass Spectrometry Imaging

Atherosclerosis is a lipid and inflammation-driven disease of the arteries that is characterized by gradual buildup of plaques in the vascular wall. A so-called vulnerable plaque, consisting of a lipid-rich necrotic core contained by a thin fibrous cap, may rupture and trigger thrombus formation, which can lead to ischemia in the heart (heart attack) or in the brain (stroke). In this study, we present a protocol to investigate the lipid composition of advanced human carotid plaques using matrix-assisted laser desorption ionization (MALDI) mass spectrometry imaging (MSI), providing a framework that should enable the discrimination of vulnerable from stable plaques based on lipid composition. We optimized the tissue preparation and imaging methods by systematically analyzing data from three specimens: two human carotid endarterectomy samples (advanced plaque) and one autopsy sample (early stage plaque). We show a robust data reduction method and evaluate the variability of the endarterectomy samples. We found diacylglycerols to be more abundant in a thrombotic area compared to other plaque areas and could distinguish advanced plaque from early stage plaque based on cholesteryl ester composition. We plan to use this systematic approach to analyze a larger dataset of carotid atherosclerotic plaques.

Technological Advances in Clinical Definition and Surveillance Methodology for Surgical Site Infection Incorporating Surgical Site Imaging and Patient-Generated Health Data

Background: Surgical site infection (SSI) continues to be a common and costly complication after surgery. The current commonly used definitions of SSI were devised more than two decades ago and do not take in to account more modern technology that could be used to make diagnosis more consistent and precise. Patient-generated health data (PGHD), including digital imaging, may be able to fulfill this objective. Methods: The published literature was examined to determine the current state of development in terms of using digital imaging as an aide to diagnose SSI. This information was used to devise possible methodology that could be used to integrate digital images to more objectively define SSI, as well as using these data for both surveillance activities and clinical management. Results: Digital imaging is a highly promising means to help define and diagnose SSI, particularly in remote settings. Multiple groups continue to actively study these emerging technologies, however, present methods remain based generally on subjective rather than objective observations. Although current images may be useful on a case-by-case basis, similar to physical examination information, integrating imaging in the definition of SSI to allow more automated diagnosis in the future will require complex image analysis combined with other available quantified data. Conclusions: Digital imaging technology, once adequately evolved, should become a cornerstone of the criteria for both the clinical and surveillance definitions of SSI.

Chemical Barcoding: A Nuclear-Magnetic-Resonance-Based Approach To Ensure the Quality and Safety of Natural Ingredients

One of the greatest challenges facing the functional food and natural health product (NHP) industries is sourcing high-quality, functional, natural ingredients for their finished products. Unfortunately, the lack of ingredient standards, modernized analytical methodologies, and industry oversight creates the potential for low quality and, in some cases, deliberate adulteration of ingredients. By exploring a diverse library of NHPs provided by the independent certification organization ISURA, we demonstrated that nuclear magnetic resonance (NMR) spectroscopy provides an innovative solution to authenticate botanicals and warrant the quality and safety of processed foods and manufactured functional ingredients. Two-dimensional NMR experiments were shown to be a robust and reproducible approach to capture the content of complex chemical mixtures, while a binary normalization step allows for emphasizing the chemical diversity in each sample, and unsupervised statistical methodologies provide key advantages to classify, authenticate, and highlight the potential presence of additives and adulterants.

Biomedical Image Processing with Containers and Deep Learning: An Automated Analysis Pipeline: Data architecture, artificial intelligence, automated processing, containerization, and clusters orchestration ease the transition from data acquisition to insights in medium-to-large datasets

Here, a streamlined, scalable, laboratory approach is discussed that enables medium-to-large dataset analysis. The presented approach combines data management, artificial intelligence, containerization, cluster orchestration, and quality control in a unified analytic pipeline. The unique combination of these individual building blocks creates a new and powerful analysis approach that can readily be applied to medium-to-large datasets by researchers to accelerate the pace of research. The proposed framework is applied to a project that counts the number of plasmonic nanoparticles bound to peripheral blood mononuclear cells in dark-field microscopy images. By using the techniques presented in this article, the images are automatically processed overnight, without user interaction, streamlining the path from experiment to conclusions.

A sparse loudspeaker array for surround sound reproduction using the least absolute shrinkage and selection operator algorithm

This letter explores a least absolute shrinkage and selection operator- (Lasso-) based beamforming algorithm for a sparse cylindrically baffled speaker array, which can be used for low-cost multi-channel surround sound reproduction. The proposed method exploits the inherent sparsity of the Lasso algorithm, and achieves both narrower beamwidth and a smaller side lobe in comparison with existing algorithms in both simulation and experiment. In addition, further study on the dependency of operating speaker sparsity on regularization parameter enables user preference-based adjustment in practice.

The challenges of big data biology

The availability of big data has the potential to transform many areas of the life sciences and usher in new ways of doing research. Here, I argue that big data biology also raises fundamental questions in the philosophy of science: for example, what is a good dataset, and how can reliable knowledge be extracted from big data? Collaborations between biologists, data scientists and philosophers of science will help us to answer these and other questions.

Automated early detection of obstetric complications: theoretic and methodologic considerations

Compared with adults who are admitted to general medical-surgical wards, women who are admitted to labor and delivery services are at much lower risk of experiencing unexpected critical illness. Nonetheless, critical illness and other complications that put either the mother or fetus at risk do occur. One potential approach to prevention is to use automated early warning systems, such as those used for nonpregnant adults. Predictive models that use data extracted in real time from electronic records constitute the cornerstone of such systems. This article addresses several issues that are involved in the development of such predictive models: specification of temporal characteristics, choice of denominator, selection of outcomes for model calibration, potential uses of existing adult severity of illness scores, approaches to data processing, statistical considerations, validation, and options for instantiation. These have not been addressed explicitly in the obstetrics literature, which has focused on the use of manually assigned scores. In addition, this article provides some results from work in progress to develop 2 obstetric predictive models with the use of data from 262,071 women who were admitted to a labor and delivery service at 15 Kaiser Permanente Northern California hospitals between 2010 and 2017.

Identification of urinary tract infections using electronic health record data

BACKGROUND

Population ascertainment of postoperative urinary tract infections (UTIs) is time-consuming and expensive, as it often requires manual chart review. Using the American College of Surgeons National Surgical Quality Improvement Program UTI status of patients who underwent an operation at the University of Colorado Hospital, we sought to develop an algorithm for identifying UTIs using data from the electronic health record.

METHODS

Data were split into training (operations occurring between 2013-2015) and test (operations in 2016) sets. A binomial generalized linear model with an elastic-net penalty was used to fit the model and carry out variables selection. International classification of disease codes, common procedural terminology codes, antibiotics, catheterization, and common procedural terminology-specific UTI event rates were included as predictors. The Youden's J statistic was used to determine the optimal classification threshold.

RESULTS

Of 6,840 patients, 134 (2.0%) had a UTI. The model achieved 92% specificity, 80% sensitivity, 100% negative predictive value, 16% positive predictive value, and an area under the curve of 0.94 using a decision threshold of 0.03.

CONCLUSIONS

A model with 14 predictors from the electronic health record identifies UTIs well, and it could be used to scale up UTI surveillance or to estimate the impact of large-scale interventions on UTI rates.

'Fit-for-purpose?' - challenges and opportunities for applications of blockchain technology in the future of healthcare

Blockchain is a shared distributed digital ledger technology that can better facilitate data management, provenance and security, and has the potential to transform healthcare. Importantly, blockchain represents a data architecture, whose application goes far beyond Bitcoin - the cryptocurrency that relies on blockchain and has popularized the technology. In the health sector, blockchain is being aggressively explored by various stakeholders to optimize business processes, lower costs, improve patient outcomes, enhance compliance, and enable better use of healthcare-related data. However, critical in assessing whether blockchain can fulfill the hype of a technology characterized as 'revolutionary' and 'disruptive', is the need to ensure that blockchain design elements consider actual healthcare needs from the diverse perspectives of consumers, patients, providers, and regulators. In addition, answering the real needs of healthcare stakeholders, blockchain approaches must also be responsive to the unique challenges faced in healthcare compared to other sectors of the economy. In this sense, ensuring that a health blockchain is 'fit-for-purpose' is pivotal. This concept forms the basis for this article, where we share views from a multidisciplinary group of practitioners at the forefront of blockchain conceptualization, development, and deployment.

Translating cancer genomics into precision medicine with artificial intelligence: applications, challenges and future perspectives

In the field of cancer genomics, the broad availability of genetic information offered by next-generation sequencing technologies and rapid growth in biomedical publication has led to the advent of the big-data era. Integration of artificial intelligence (AI) approaches such as machine learning, deep learning, and natural language processing (NLP) to tackle the challenges of scalability and high dimensionality of data and to transform big data into clinically actionable knowledge is expanding and becoming the foundation of precision medicine. In this paper, we review the current status and future directions of AI application in cancer genomics within the context of workflows to integrate genomic analysis for precision cancer care. The existing solutions of AI and their limitations in cancer genetic testing and diagnostics such as variant calling and interpretation are critically analyzed. Publicly available tools or algorithms for key NLP technologies in the literature mining for evidence-based clinical recommendations are reviewed and compared. In addition, the present paper highlights the challenges to AI adoption in digital healthcare with regard to data requirements, algorithmic transparency, reproducibility, and real-world assessment, and discusses the importance of preparing patients and physicians for modern digitized healthcare. We believe that AI will remain the main driver to healthcare transformation toward precision medicine, yet the unprecedented challenges posed should be addressed to ensure safety and beneficial impact to healthcare.

El-MAVEN: A Fast, Robust, and User-Friendly Mass Spectrometry Data Processing Engine for Metabolomics

Analysis of large metabolomic datasets is becoming commonplace with the increased realization of the role that metabolites play in biology and pathophysiology. While there are many open-source analysis tools to extract peaks from liquid chromatography-mass spectrometry (LC-MS), gas chromatography-mass spectrometry (GC-MS), and tandem mass spectrometry (LC-MS/MS) data, these tools are not very interactive and are suboptimal when a large number of samples are to be analyzed. El-MAVEN is an open-source analysis platform that extends MAVEN and provides fast, powerful, and interactive analysis capabilities especially for datasets containing over 100 samples. The El-MAVEN workflow is easy to use with just four steps from loading data to exporting of the results. Advanced analysis and software techniques such as multiprocessing, machine learning, and reduction of memory leaks are implemented so as to provide a seamless and interactive user experience. Results from El-MAVEN can be exported in a range of formats allowing continued analysis on other platforms. Additionally, El-MAVEN is also fully integrated with Polly™, a cloud-based analysis platform that provides a range of tools for flux analysis and integrative-omics analysis. El-MAVEN is a powerful tool that enables fast and efficient analysis of large metabolomic datasets to accelerate the process of gaining insight from raw data.

Live Cell Barcoding for Efficient Analysis of Small Samples by Mass Cytometry

In mass cytometry, sample loss is of considerable concern due to the relative inefficiency of cell event collection compared to similar techniques such as flow cytometry. Cell stimulation and the harsh conditions required in the later stages of certain sample preparations also contribute to cell loss. Low starting cell numbers are especially susceptible to these effects, potentially limiting the ability to use mass cytometry. Here is presented a live cell barcoding scheme and additional efficiency methods to improve recovery and achieve consistent staining for small samples.

A High-Throughput Targeted Metabolomics Workflow for the Detection of 200 Polar Metabolites in Central Carbon Metabolism

Targeted metabolomics aims to analyze a set of pre-selected metabolites from biologically relevant metabolic pathways. The triple quadrupole mass spectrometry (QqQ-MS) based multiple reaction monitoring (MRM) technique is the most widely approach used for targeted metabolomics, and features high selectivity and sensitivity, good reproducibility and wide dynamic range in quantitative analysis. Here, we describe an MRM based targeted metabolomics workflow for the quantitative analysis of 200 polar metabolites in central carbon metabolic pathways, including the data acquisition method and the automated data processing procedures using our in-house R package MRMAnalyzer. The workflow described in this chapter combines a hydrophilic interaction liquid chromatography (HILIC) separation and positive/negative ion polarity switching based MS detection, and is able to acquire data from multiple types of biological samples such as bacteria, cultured mammalian cells, animal tissues and biofluids (e.g., serum and urine). Finally, the MRMAnalyzer software can automatically process the generated large-scale data set with high efficiency. We hope it is a valuable and efficient workflow for researchers to facilitate the respective biological studies using targeted metabolomics.

PENYEK: Automated brown planthopper detection from imperfect sticky pad images using deep convolutional neural network

Rice is a staple food in Asia and it contributes significantly to the Gross Domestic Product (GDP) of Malaysia and other developing countries. Brown Planthopper (BPH) causes high levels of economic loss in Malaysia. Identification of BPH presence and monitoring of its abundance has been conducted manually by experts and is time-consuming, fatiguing and tedious. Automated detection of BPH has been proposed by many studies to overcome human fallibility. However, all studies regarding automated recognition of BPH are investigated based on intact specimen although most of the specimens are imperfect, with missing parts have distorted shapes. The automated recognition of an imperfect insect image is more difficult than recognition of the intact specimen. This study proposes an automated, deep-learning-based detection pipeline, PENYEK, to identify BPH pest in images taken from a readily available sticky pad, constructed by clipping plastic sheets onto steel plates and spraying with glue. This study explores the effectiveness of a convolutional neural network (CNN) architecture, VGG16, in classifying insects as BPH or benign based on grayscale images constructed from Euclidean Distance Maps (EDM). The pipeline identified imperfect images of BPH with an accuracy of 95% using deep-learning’s hyperparameters: softmax, a mini-batch of 30 and an initial learning rate of 0.0001.

Effectiveness of Digital Interventions for Improving Glycemic Control in Persons with Poorly Controlled Type 2 Diabetes: A Systematic Review, Meta-analysis, and Meta-regression Analysis

BACKGROUND

Digital interventions may assist patients with type 2 diabetes in improving glycemic control. We aimed to synthesize effect sizes of digital interventions on glycated hemoglobin (HbA1c) levels and to identify effective features of digital interventions targeting patients with poorly controlled type 2 diabetes.

MATERIALS AND METHODS

MEDLINE, ISI Web of Science, and PsycINFO were searched for randomized controlled trials (RCTs) comparing the effects of digital interventions with usual care. Two reviewers independently assessed studies for eligibility and determined study quality, using the Cochrane Risk of Bias Assessment Tool. The Behavioral Change Technique Taxonomy V1 (BCTTv1) was used to identify BCTs used in interventions. Mean HbA1c differences were pooled using analysis of covariance to adjust for baseline differences and pre-post correlations. To examine effective intervention features and to evaluate differences in effect sizes across groups, meta-regression and subgroup analyses were performed.

RESULTS

Twenty-three arms of 21 RCTs were included in the meta-analysis (n = 3787 patients, 52.6% in intervention arms). The mean HbA1c baseline differences ranged from -0.2% to 0.64%. The pooled mean HbA1c change was statistically significant (-0.39 {95% CI: [-0.51 to -0.26]} with substantial heterogeneity [I² statistic, 80.8%]) and a significant HbA1c reduction was noted for web-based interventions. A baseline HbA1c level above 7.5%, β = -0.44 (95% CI: [-0.81 to -0.06]), the BCTs "problem solving," β = -1.30 (95% CI: [-2.05 to -0.54]), and "self-monitoring outcomes of behavior," β = -1.21 (95% CI: [-1.95 to -0.46]) were significantly associated with reduced HbA1c levels.

CONCLUSIONS

Digital interventions appear effective for reducing HbA1c levels in patients with poorly controlled type 2 diabetes.

Information Processing and Distributed Computation in Plant Organs

The molecular networks plant cells evolved to tune their development in response to the environment are becoming increasingly well understood. Much less is known about how these programs function in the multicellular context of organs and the impact this spatial embedding has on emergent decision-making. Here I address these questions and investigate whether the computational control principles identified in engineered information processing systems also apply to plant development. Examples of distributed computing underlying plant development are presented and support the presence of shared mechanisms of information processing across these domains. The coinvestigation of computation across plant biology and computer science can provide novel insight into the principles of plant development and suggest novel algorithms for use in distributed computing.

An improved memory-based collaborative filtering method based on the TOPSIS technique

This paper describes an approach for improving the accuracy of memory-based collaborative filtering, based on the technique for order of preference by similarity to ideal solution (TOPSIS) method. Recommender systems are used to filter the huge amount of data available online based on user-defined preferences. Collaborative filtering (CF) is a commonly used recommendation approach that generates recommendations based on correlations among user preferences. Although several enhancements have increased the accuracy of memory-based CF through the development of improved similarity measures for finding successful neighbors, there has been less investigation into prediction score methods, in which rating/preference scores are assigned to items that have not yet been selected by a user. A TOPSIS solution for evaluating multiple alternatives based on more than one criterion is proposed as an alternative to prediction score methods for evaluating and ranking items based on the results from similar users. The recommendation accuracy of the proposed TOPSIS technique is evaluated by applying it to various common CF baseline methods, which are then used to analyze the MovieLens 100K and 1M benchmark datasets. The results show that CF based on the TOPSIS method is more accurate than baseline CF methods across a number of common evaluation metrics.

An automated approach to the quantitation of vocalizations and vocal learning in the songbird

Studies of learning mechanisms critically depend on the ability to accurately assess learning outcomes. This assessment can be impeded by the often complex, multidimensional nature of behavior. We present a novel, automated approach to evaluating imitative learning. Conceptually, our approach estimates how much of the content present in a reference behavior is absent from the learned behavior. We validate our approach through examination of songbird vocalizations, complex learned behaviors the study of which has provided many insights into sensory-motor learning in general and vocal learning in particular. Historically, learning has been holistically assessed by human inspection or through comparison of specific song features selected by experimenters (e.g. fundamental frequency, spectral entropy). In contrast, our approach uses statistical models to broadly capture the structure of each song, and then estimates the divergence between the two models. We show that our measure of song learning (the Kullback-Leibler divergence between two distributions corresponding to specific song data, or, Song D_KL) is well correlated with human evaluation of song learning. We then expand the analysis beyond learning and show that Song D_KL also detects the typical song deterioration that occurs following deafening. Finally, we illustrate how this measure can be extended to quantify differences in other complex behaviors such as human speech and handwriting. This approach potentially provides a framework for assessing learning across a broad range of behaviors like song that can be described as a set of discrete and repeated motor actions.

Measuring learning outcomes is a critical objective of research into the mechanisms that support learning. Demonstration that a given manipulation results in better or worse learning outcomes requires an accurate and consistent measurement of learning quality. However, many behaviors (e.g. speaking, walking, and writing) are complex and multidimensional, confounding the assessment of learning. One behavior subject to such confounds, vocal learning in Estrildid finches, has emerged as a vital model for sensory motor learning broadly and human speech learning in particular. Here, we demonstrate a new approach to the assessment of learning for complex high dimensional behaviors. Conceptually, we estimate the amount of content present in a reference behavior that is absent in the resultant learned behavior. We show that this measure provides a holistic and automated assessment of vocal learning in Estrildid finches that is consistent with human assessment. We then illustrate how this measure can be used to quantify changes in other complex behaviors such as human speech. We conclude that this approach could be useful in assessing shared content in other similarly structured behaviors composed of a set of discrete and repeated motor actions.

An efficient outlier removal method for scattered point cloud data

Outlier removal is a fundamental data processing task to ensure the quality of scanned point cloud data (PCD), which is becoming increasing important in industrial applications and reverse engineering. Acquired scanned PCD is usually noisy, sparse and temporarily incoherent. Thus the processing of scanned data is typically an ill-posed problem. In the paper, we present a simple and effective method based on two geometrical characteristics constraints to trim the noisy points. One of the geometrical characteristics is the local density information and another is the deviation from the local fitting plane. The local density based method provides a preprocessing step, which could remove those sparse outlier and isolated outlier. The non-isolated outlier removal in this paper depends on a local projection method, which placing those points onto objects. There is no doubt that the deviation of any point from the local fitting plane should be a criterion to reduce the noisy points. The experimental results demonstrate the ability to remove the noisy point from various man-made objects consisting of complex outlier.

Developing a System for Processing Health Data of Children Using Digitalized Toys: Ethical and Privacy Concerns for the Internet of Things Paradigm

EDUCERE (Ubiquitous Detection Ecosystem to Care and Early Stimulation for Children with Developmental Disorders) is a government funded research and development project. EDUCERE objectives are to investigate, develop, and evaluate innovative solutions for society to detect changes in psychomotor development through the natural interaction of children with toys and everyday objects, and perform stimulation and early attention activities in real environments such as home and school. In the EDUCERE project, an ethical impact assessment is carried out linked to a minors' data protection rights. Using a specific methodology, the project has achieved some promising results. These include use of a prototype of smart toys to detect development difficulties in children. In addition, privacy protection measures which take into account the security concerns of health data, have been proposed and applied. This latter security framework could be useful in other Internet of Things related projects. It consists of legal and technical measures. Special attention has been placed in the transformation of bulk data such as acceleration and jitter of toys into health data when patterns of atypical development are found. The article describes the different security profiles in which users are classified.

Data reconstruction using iteratively reweighted L1-principal component analysis for an electronic nose system

We propose a method to reconstruct damaged data based on statistical learning during data acquisition. In the process of measuring the data using a sensor, the damage of the data caused by the defect of the sensor or the environmental factor greatly degrades the performance of data classification. Instead of the traditional PCA based on L2-norm, the PCA features were extracted based on L1-norm and updated by iteratively reweighted fitting using the generalized objective function to obtain robust features for the outlier data. The damaged data samples were reconstructed using weighted linear combination using these features and the projection vectors of L1-norm based PCA. The experimental results on various types of volatile organic compounds (VOCs) data show that the proposed method can be used to reconstruct the damaged data to the original form of the undamaged data and to prevent degradation of classification performance due to data corruption through data reconstruction.

Big-BOE: Fusing Spanish Official Gazette with Big Data Technology

The proliferation of new data sources, stemmed from the adoption of open-data schemes, in combination with an increasing computing capacity causes the inception of new type of analytics that process Internet of things with low-cost engines to speed up data processing using parallel computing. In this context, the article presents an initiative, called BIG-Boletín Oficial del Estado (BOE), designed to process the Spanish official government gazette (BOE) with state-of-the-art processing engines, to reduce computation time and to offer additional speed up for big data analysts. The goal of including a big data infrastructure is to be able to process different BOE documents in parallel with specific analytics, to search for several issues in different documents. The application infrastructure processing engine is described from an architectural perspective and from performance, showing evidence on how this type of infrastructure improves the performance of different types of simple analytics as several machines cooperate.