Large Public Data Research Articles

Machine learning for predicting compressive strength of sustainable cement paste incorporating copper mine tailings as supplementary cementitious materials

Copper mining produces significant amounts of copper mine tailings (CMT), necessitating appropriate waste handling and disposal practices. By substituting a portion of cement with CMT as supplementary cementitious materials (SCMs), we aim to address two environmental issues simultaneously: reducing copper mine waste in landfills and decreasing embodied carbon by using less cement. The exploration of CMT recycling as a cement replacement requires evaluation of its impact on material performance, such as compressive strength. In this paper, we address this by machine learning that features data fusion of large public data with our own small data on compressive strength of CMT-incorporated cement. We developed and critically evaluated three machine learning models: a simple linear model, Gaussian process, and random forest that predict the compressive strength of CMT-incorporated cement pastes with different mix designs (e.g., varying amounts of CMT and water-binder ratios) and curing ages. Hyperparameters in the random forest model were tuned using Bayesian optimization. Following a comprehensive evaluation of the models, we find that the random forest model can accurately estimate the compressive strength of cement paste across the mix designs. Furthermore, results from SHapley Additive exPlanation (SHAP), Individual Conditional Expectation (ICE), and Partial Dependence Plots (PDP) revealed that cement, ground-granulated blast furnace slag, superplasticizers, and curing ages positively influence compressive strength. This study contributes to acceleration of sustainable material technology to obtain the best mix design and desired compressive strength.

Predicting Chemical Immunotoxicity through Data-Driven QSAR Modeling of Aryl Hydrocarbon Receptor Agonism and Related Toxicity Mechanisms.

Computational modeling has emerged as a time-saving and cost-effective alternative to traditional animal testing for assessing chemicals for their potential hazards. However, few computational modeling studies for immunotoxicity were reported, with few models available for predicting toxicants due to the lack of training data and the complex mechanisms of immunotoxicity. In this study, we employed a data-driven quantitative structure-activity relationship (QSAR) modeling workflow to extensively enlarge the limited training data by revealing multiple targets involved in immunotoxicity. To this end, a probe data set of 6,341 chemicals was obtained from a high-throughput screening (HTS) assay testing for the activation of the aryl hydrocarbon receptor (AhR) signaling pathway, a key event leading to immunotoxicity. Searching this probe data set against PubChem yielded 3,183 assays with testing results for varying proportions of these 6,341 compounds. 100 assays were selected to develop QSAR models based on their correlations to AhR agonism. Twelve individual QSAR models were built for each assay using combinations of four machine-learning algorithms and three molecular fingerprints. 5-fold cross-validation of the resulting models showed good predictivity (average CCR = 0.73). A total of 20 assays were further selected based on QSAR model performance, and their resulting QSAR models showed good predictivity of potential immunotoxicants from external chemicals. This study provides a computational modeling strategy that can utilize large public toxicity data sets for modeling immunotoxicity and other toxicity endpoints, which have limited training data and complicated toxicity mechanisms.

An Analysis of Orientation and Preceptorship Influence on Nurse Job Satisfaction Across Two States.

A large public nursing data set was used to determine whether orientation and/or preceptor programs impact job satisfaction among registered nurses in Maine and Massachusetts. There was no association between orientation and preceptor programs and satisfaction, nor evidence that new nurse status modified the relationship. There is a need for evaluation of orientation and preceptor programs' structure and effectiveness, and innovation is needed in promoting job satisfaction, thereby increasing nurse retention.

Increasing Sufu gene dosage reveals its unorthodox role in promoting polydactyly and medulloblastoma tumorigenesis.

Suppressor of fused (SUFU) is widely regarded as a key negative regulator of the sonic hedgehog (SHH) morphogenic pathway and a known tumor suppressor of medulloblastoma (MB). However, we report here that SUFU expression was markedly increased in 75% of specimens compiled in a tissue array comprising 49 unstratified MBs. The SUFU and GLI1 expression levels in this MB array showed strong positive correlation, which was also identified in a large public data set containing 736 MBs. We further report that increasing Sufu gene dosage in mice caused preaxial polydactyly, which was associated with the expansion of the Gli3 domain in the anterior limb bud and heightened Shh signaling responses during embryonic development. Increasing Sufu gene dosage also led to accelerated cerebellar development and, when combined with ablation of the Shh receptor encoded by Patched1 (Ptch1), promoted MB tumorigenesis. These data reveal multifaceted roles of SUFU in promoting MB tumorigenesis by enhancing SHH signaling. This revelation clarifies potentially counterintuitive clinical observation of high SUFU expression in MBs and may pave way for novel strategies to reduce or reverse MB progression.

MetaNorm: incorporating meta-analytic priors into normalization of NanoString nCounter data.

Non-informative or diffuse prior distributions are widely employed in Bayesian data analysis to maintain objectivity. However, when meaningful prior information exists and can be identified, using an informative prior distribution to accurately reflect current knowledge may lead to superior outcomes and great efficiency. We propose MetaNorm, a Bayesian algorithm for normalizing NanoString nCounter gene expression data. MetaNorm is based on RCRnorm, a powerful method designed under an integrated series of hierarchical models that allow various sources of error to be explained by different types of probes in the nCounter system. However, a lack of accurate prior information, weak computational efficiency, and instability of estimates that sometimes occur weakens the approach despite its impressive performance. MetaNorm employs priors carefully constructed from a rigorous meta-analysis to leverage information from large public data. Combined with additional algorithmic enhancements, MetaNorm improves RCRnorm by yielding more stable estimation of normalized values, better convergence diagnostics and superior computational efficiency. R Code for replicating the meta-analysis and the normalization function can be found at github.com/jbarth216/MetaNorm.

Comparing the performance of a deep learning-based lung gross tumour volume segmentation algorithm before and after transfer learning in a new hospital.

Radiation therapy for lung cancer requires a gross tumour volume (GTV) to be carefully outlined by a skilled radiation oncologist (RO) to accurately pinpoint high radiation dose to a malignant mass while simultaneously minimizing radiation damage to adjacent normal tissues. This is manually intensive and tedious however, it is feasible to train a deep learning (DL) neural network that could assist ROs to delineate the GTV. However, DL trained on large openly accessible data sets might not perform well when applied to a superficially similar task but in a different clinical setting. In this work, we tested the performance of DL automatic lung GTV segmentation model trained on open-access Dutch data when used on Indian patients from a large public tertiary hospital, and hypothesized that generic DL performance could be improved for a specific local clinical context, by means of modest transfer-learning on a small representative local subset. X-ray computed tomography (CT) series in a public data set called "NSCLC-Radiomics" from The Cancer Imaging Archive was first used to train a DL-based lung GTV segmentation model (Model 1). Its performance was assessed using a different open access data set (Interobserver1) of Dutch subjects plus a private Indian data set from a local tertiary hospital (Test Set 2). Another Indian data set (Retrain Set 1) was used to fine-tune the former DL model using a transfer learning method. The Indian data sets were taken from CT of a hybrid scanner based in nuclear medicine, but the GTV was drawn by skilled Indian ROs. The final (after fine-tuning) model (Model 2) was then re-evaluated in "Interobserver1" and "Test Set 2." Dice similarity coefficient (DSC), precision, and recall were used as geometric segmentation performance metrics. Model 1 trained exclusively on Dutch scans showed a significant fall in performance when tested on "Test Set 2." However, the DSC of Model 2 recovered by 14 percentage points when evaluated in the same test set. Precision and recall showed a similar rebound of performance after transfer learning, in spite of using a comparatively small sample size. The performance of both models, before and after the fine-tuning, did not significantly change the segmentation performance in "Interobserver1." A large public open-access data set was used to train a generic DL model for lung GTV segmentation, but this did not perform well initially in the Indian clinical context. Using transfer learning methods, it was feasible to efficiently and easily fine-tune the generic model using only a small number of local examples from the Indian hospital. This led to a recovery of some of the geometric segmentation performance, but the tuning did not appear to affect the performance of the model in another open-access data set. Caution is needed when using models trained on large volumes of international data in a local clinical setting, even when that training data set is of good quality. Minor differences in scan acquisition and clinician delineation preferences may result in an apparent drop in performance. However, DL models have the advantage of being efficiently "adapted" from a generic to a locally specific context, with only a small amount of fine-tuning by means of transfer learning on a small local institutional data set.

Predicting Patients' Satisfaction With Mental Health Drug Treatment Using Their Reviews: Unified Interchangeable Model Fusion Approach.

After the COVID-19 pandemic, the conflict between limited mental health care resources and the rapidly growing number of patients has become more pronounced. It is necessary for psychologists to borrow artificial intelligence (AI)-based methods to analyze patients' satisfaction with drug treatment for those undergoing mental illness treatment. Our goal was to construct highly accurate and transferable models for predicting the satisfaction of patients with mental illness with medication by analyzing their own experiences and comments related to medication intake. We extracted 41,851 reviews in 20 categories of disorders related to mental illnesses from a large public data set of 161,297 reviews in 16,950 illness categories. To discover a more optimal structure of the natural language processing models, we proposed the Unified Interchangeable Model Fusion to decompose the state-of-the-art Bidirectional Encoder Representations from Transformers (BERT), support vector machine, and random forest (RF) models into 2 modules, the encoder and the classifier, and then reconstruct fused "encoder+classifer" models to accurately evaluate patients' satisfaction. The fused models were divided into 2 categories in terms of model structures, traditional machine learning-based models and neural network-based models. A new loss function was proposed for those neural network-based models to overcome overfitting and data imbalance. Finally, we fine-tuned the fused models and evaluated their performance comprehensively in terms of F1-score, accuracy, κ coefficient, and training time using 10-fold cross-validation. Through extensive experiments, the transformer bidirectional encoder+RF model outperformed the state-of-the-art BERT, MentalBERT, and other fused models. It became the optimal model for predicting the patients' satisfaction with drug treatment. It achieved an average graded F1-score of 0.872, an accuracy of 0.873, and a κ coefficient of 0.806. This model is suitable for high-standard users with sufficient computing resources. Alternatively, it turned out that the word-embedding encoder+RF model showed relatively good performance with an average graded F1-score of 0.801, an accuracy of 0.812, and a κ coefficient of 0.695 but with much less training time. It can be deployed in environments with limited computing resources. We analyzed the performance of support vector machine, RF, BERT, MentalBERT, and all fused models and identified the optimal models for different clinical scenarios. The findings can serve as evidence to support that the natural language processing methods can effectively assist psychologists in evaluating the satisfaction of patients with drug treatment programs and provide precise and standardized solutions. The Unified Interchangeable Model Fusion provides a different perspective on building AI models in mental health and has the potential to fuse the strengths of different components of the models into a single model, which may contribute to the development of AI in mental health.

P10.10.B BIRC3 AS A PREDICTOR OF OUTCOME AND A NEW THERAPEUTIC TARGET IN TMZ RESISTANT GLIOBLASTOMA TUMORS

Abstract BACKGROUND Glioblastoma (GB) is the most aggressive malignant tumor of the central nervous system, with a highly unfavorable prognosis with a median survival of 15 months. The standard treatment consists of surgical resection, chemotherapy, and radiotherapy. Despite intense research, overall survival remains stagnant, with very few approved therapeutic agents. Understanding the mutational profile of a tumor is important to inform diagnosis and guide therapeutic options, as well as identify patients that may not respond to treatment. The key advantage of precision medicine in oncology is the ability to match the right drug to the right patient to help improve patient outcomes. Aside from MGMT promoter methylation and TMZ response, molecular biomarkers associated with therapeutic response or prognostic of survival are still lacking in GB. In a recent study, using NADH-fluorescence lifetime imaging (NADH-FLIM), as a new drug screening precision medicine ex-vivo approach, we perturbed IDH1/2 WT GB patients-derived vital tumors by TMZ treatment and obtained a stratification in TMZ sensitive (S) and non-sensitive (NS) tumors, identifying several genes differentially expressed between the two groups. MATERIAL AND METHODS In this study, we enlarged our previous glioblastoma case study by performing NADH-FLIM metabolic approach to test TMZ drug tumor response on several primary and recurrent ex-vivo live GB tumors. Transcriptome analysis was performed to molecularly characterize TMZ S and NS cases. Moreover, functional cellular experiments were carried on to explore biomarkers' efficiency. RESULTS The majority of the genes, found dysregulated in the previous study, showed the same directional regulation of expression and we confirmed, in TMZ-resistant samples, the statistically significant upregulation of BIRC3, a strong inhibitor of apoptosis. The large public TCGA data set provided an opportunity to fully assess the BIRC3 mRNA expression complementarity with the MGMT status as a predictor of outcome in several cohorts analyzed with different RNA analysis approaches. BIRC3 functioned as a prognostic factor of survival also in a French and Italian private glioblastoma cohort analyzed by RNAseq and RT-PCR, respectively. In addition, most importantly, we show that the drug AZD5582, which is a specific antagonist of BIRC3, combined with TMZ treatment, is able to revert TMZ resistance, by restoring apoptosis, in glioblastoma cell lines and in patients derived organoids that are non-sensitive to TMZ-treatment. Higher levels of BIRC3 mRNA expression are associated with AZD5582 efficacy. CONCLUSION Investigating the mRNA expression levels of BIRC3 in GB tumors can be used as a tool to predict survival and response to TMZ. Moreover, the presence of high levels of BIRC3 mRNA is indicative of a new therapeutic target in glioblastoma tumors that are resistant to standard TMZ therapy.

Tackling the lack of diversity in cancer research.

Despite the clear benefit of studying biological samples from diverse genetic backgrounds and geographical locations, our current knowledge of disease is mostly derived from the study of European-descent individuals. In the cancer field, this is reflected in the poor representation of African and Amerindian/Latino samples in most large public data repositories. This lack of diversity is due to several reasons, but here we focus on (1) the lack of support for studies on non-European populations that are performed in low- and middle-income countries (LMICs), and (2) unequal partnerships between scientists in LMICs and those in high-income countries. We argue that expanding access to research funding, increasing the participation of underrepresented scientists in editorial boards and international conferences, facilitating the publication of studies conducted in these countries, and properly acknowledging LMIC researchers' contributions in publications and grant applications will promote equity for scientists working in LMICs. We envisage that this will translate to more impactful research in these countries, which will include more samples from diverse populations. For the cancer field, this will broaden our understanding of pathomechanisms and may help to improve the treatment of patients from all backgrounds.

The use of generative adversarial networks for multi-site one-class follicular lymphoma classification

Recent advances in digital technologies have lowered the costs and improved the quality of digital pathology Whole Slide Images (WSI), opening the door to apply Machine Learning (ML) techniques to assist in cancer diagnosis. ML, including Deep Learning (DL), has produced impressive results in diverse image classification tasks in pathology, such as predicting clinical outcomes in lung cancer and inferring regional gene expression signatures. Despite these promising results, the uptake of ML as a common diagnostic tool in pathology remains limited. A major obstacle is the insufficient labelled data for training neural networks and other classifiers, especially for new sites where models have not been established yet. Recently, image synthesis from small, labelled datasets using Generative Adversarial Networks (GAN) has been used successfully to create high-performing classification models. Considering the domain shift and complexity in annotating data, we investigated an approach based on GAN that minimized the differences in WSI between large public data archive sites and a much smaller data archives at the new sites. The proposed approach allows the tuning of a deep learning classification model for the class of interest to be improved using a small training set available at the new sites. This paper utilizes GAN with the one-class classification concept to model the class of interest data. This approach minimizes the need for large amounts of labelled data from the new site to train the network. The GAN generates synthesized one-class WSI images to jointly train the classifier with WSIs available from the new sites. We tested the proposed approach for follicular lymphoma data of a new site by utilizing the data archives from different sites. The synthetic images for the one-class data generated from the data obtained from different sites with minimum amount of data from the new site have resulted in a significant improvement of 15% for the Area Under the curve (AUC) for the new site that we want to establish a new follicular lymphoma classifier. The test results have shown that the classifier can perform well without the need to obtain more training data from the test site, by utilizing GAN to generate the synthetic data from all existing data in the archives from all the sites.

Abstract 6685: Targeted liquid biopsy mutation profiling and genome-wide cfDNA copy-number landscape in 3000 samples from six major cancer indications

Abstract Introduction. PredicineCARE is a CLIA-certified targeted next-generation sequencing (NGS) assay that provides comprehensive, non-invasive genomic profiling of circulating tumor DNA (ctDNA) in blood and urine. Here we present liquid biopsy mutational profiles and genome-wide cell-free DNA (cfDNA) copy-number burden levels for historical cancer cohorts analyzed with PredicineCARE. We include a total of 3,147 samples from six major cancer indications: prostate (n=1681), lung (n=630), breast (n=300), colorectal (n=217), pancreatic (n=190), and bladder (n=129). Methods. PredicineCARE employs a 152 targeted gene panel covering all major guideline-recommended alterations for patient care and clinical drug development including targeted therapies and immunotherapies. The assay detects single nucleotide variants (SNVs), small insertions and deletions, gene-level copy number loss/gain, and targeted re-arrangements/fusions. Concurrent low-pass whole-genome sequencing (without the need for additional sample volume) is performed to measure blood-based genome-wide copy-number burden (bCNB) and tumor fraction, providing important complementary molecular insights to variant detection. Results. We present data from a representative cohort of blood and urine samples processed at Predicine lab in Hayward California (n=3,147). Sequencing was performed using a minimum of 5 ng total cfDNA input from 1-5 mL of plasma or 20-40mL of urine, at one or more clinical timepoints per patient. Samples were sequenced at ~20,000X target panel coverage and the majority also had ~2.5X low-pass whole genome sequencing. Using the Predicine in-house DeepSea bioinformatics pipeline, we report variants as low as 0.05% allele frequency and estimate genome-wide CNVs and bCNB from low-pass data. We compare the mutational and CNV profiles of important genes in each liquid biopsy cancer cohort to variants reported in solid tumor tissues from several large public data sets including Project GENIE (n=71,817 patients) and 136 additional study programs including TCGA (n=38,619 patients, www.cBioPortal.org). Conclusion. PredicineCARE is a highly sensitive and robust assay for detecting cancer variants in blood and urine. The cfDNA mutation profiles are mostly consistent with public tissue datasets. A small but significant portion of the detected ctDNA variants have been shown to be important for treatment selection or are currently being tested for targeted drug development. Increased bCNB provides a useful measure to monitor treatment response and disease progression. PredicineCARE has been CE-marked and received breakthrough device designation from FDA. A harmonized PredicineCARE assay is available in the US and China to support patient testing in the clinic, global clinical trials and CDx development. Citation Format: Billie Gould, Lu Tan, Chao Dai, Tiantian Zheng, Wei Mo, Xianghong Wang, Michael Sneddon, Shidong Jia, Kemin Zhou, Pan Du. Targeted liquid biopsy mutation profiling and genome-wide cfDNA copy-number landscape in 3000 samples from six major cancer indications [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6685.

JPEG Pleno holography presents the numerical reconstruction software for holograms: an excursion in holographic views.

Digital reconstructions of numerical holograms enable data visualization and serve a multitude of purposes ranging from microscopy to holographic displays. Over the years, many pipelines have been developed for specific hologram types. Within the standardization effort of JPEG Pleno holography, an open-source MATLAB toolbox was developed that reflects the best current consensus. It can process Fresnel, angular spectrum, and Fourier-Fresnel holograms with one or more color channels; it also allows for diffraction-limited numerical reconstructions. The latter provides a way to reconstruct holograms at their intrinsic physical instead of an arbitrarily chosen numerical resolution. The Numerical Reconstruction Software for Holograms v10 supports all large public data sets featured by UBI, BCOM, ETRI, and ETRO, in their native and vertical off-axis binary forms. Through the release of this software, we hope to improve the reproducibility of research, thus enabling consistent comparison of data between research groups and the quality of specific numerical reconstructions.

NuKit: A deep learning platform for fast nucleus segmentation of histopathological images.

Nucleus segmentation represents the initial step for histopathological image analysis pipelines, and it remains a challenge in many quantitative analysis methods in terms of accuracy and speed. Recently, deep learning nucleus segmentation methods have demonstrated to outperform previous intensity- or pattern-based methods. However, the heavy computation of deep learning provides impression of lagging response in real time and hampered the adoptability of these models in routine research. We developed and implemented NuKit a deep learning platform, which accelerates nucleus segmentation and provides prompt results to the users. NuKit platform consists of two deep learning models coupled with an interactive graphical user interface (GUI) to provide fast and automatic nucleus segmentation "on the fly". Both deep learning models provide complementary tasks in nucleus segmentation. The whole image segmentation model performs whole image nucleus whereas the click segmentation model supplements the nucleus segmentation with user-driven input to edits the segmented nuclei. We trained the NuKit whole image segmentation model on a large public training data set and tested its performance in seven independent public image data sets. The whole image segmentation model achieves average [Formula: see text] and [Formula: see text]. The outputs could be exported into different file formats, as well as provides seamless integration with other image analysis tools such as QuPath. NuKit can be executed on Windows, Mac, and Linux using personal computers.

Enhancing plant-derived smart nanoinhibitor in targeting mammalian target of rapamycin (mTOR) in breast cancer using Curcuma longa-derived compound curcumin.

Breast cancer is a diverse female malignancy; its classification is based on clinical evidence and pathologicalelucidation. Large public drug screening data databases combined with transcriptome measures have helped developpredictive computational models. Breast cancer is frequent among women worldwide. Several genes increase breast cancerrisk. The Mammalian Target of Rapamycin (popularly known as mTOR) is a risk factor mutated in numerous breastcarcinoma types. This has caught the scientific community's focus, which is attempting to generate creative, potent, andbio-available ligands for future anti-cancer treatments to establish a practical therapeutic approach. mTOR is a proteinkinase involved in cell proliferation, survival, metabolism, and immune response. Activating mTOR promotes cancer growthand spread. To generate a bioavailable and effective mTOR inhibitor, we used computer-aided drug design to studychromones and flavonoids, two naturally occurring chemicals with many biological activities. We used Curcuma longaderivedtiny nano-molecules, which can be coated using liposomes to target mTOR to prevent breast cancer growth.The significant interactions of Curcumin were anticipated using molecular docking. It had the highest binding affinityat -12.26 kcal/mol. 100 nanoseconds of molecular dynamic modelling confirmed Curcumin and mTOR receptor interaction.Liposomes are a form of medicine carrier. To improve healthcare, more liposome-like nanostructures are being made.Nanostructures' interactions with living creatures are being studied. Half-life, tissue accumulation, and toxicity have beenstudied. Future medication distribution may use nanocarriers having a liposome-like form, enabling targeted nano-delivery.Curcumin's interaction with the active site increased the complex's structural stability during its expansion. Ourresults may help future investigations of Curcumin's efficacy as a possible lead treatment targeting mTOR receptors inbreast cancer. Using Curcumin as a potential anti-cancer drug with lipid-coated nano-particles allows for tailoredadministration.

Study on Improved VGGNet and SK Convolution Identification Model for Defect Classification of Single Molten Salt Battery

Aiming at the lack of large public single molten salt battery data sets, to reduce the labour-consuming, to improve the insufficient learning ability of traditional diagnostic methods in the production of single molten salt battery, an image recognition model for molten salt battery defects based on transfer learning is proposed. First, some pre-processing operations and image enhancement on the single molten salt battery image are performed. Second, the backbone of recognition model is built based on VGG16 network, and the selective kernel (SK) convolution module is adopted after the bottleneck layer, convolution kernel with an appropriate size can be selected adaptively through input feature map; Third, the FC is taken the place of a GAP layer, a dropout layer, and other fine-tuning operations are added, a simplified model called V-VGGNet is got; Finally, the weight parameters obtained from the pre-training on the ImageNet data set are transferred to the single molten salt battery image recognition model V-VGGNet. For different network structures and different training strategies, comparative experiments of performance tests are conducted. The test data manifest that the accuracy rates of V-VGGNet network for three categories of defective images (Missing Negative Electrode, Broken Tab, and Missing Current Collector) and Assembly Normal images can reach 95.14%, 98.79%, 98.21%, and 99.41%, the average accuracy can achieve 97.91%, good performance improvement of the single molten salt battery is improved, it is about 3% higher compared to other well-knows networks, which verified the feasibility of V-VGGNet model and the effectiveness of the improvement.

Artistic motivations are intrinsic, specific, and temporally stable by nature: Evidence from large real-life Brazilian public data between 1987–2004

AbstractArtistic behavior as aesthetically enhancing activities is conceptualized as a functionally autonomous activity within the evolved human behavioral repertoire. Accordingly, it should be intrinsically motivated, and it might also be expected to be temporally stable and domain specific. Preferential freely-pursued activities reflect intrinsic motivation and offer a valuable measure of artistic motivation. We used a large decades-long real-life public Brazilian data set from university applications to test these ideas. We analysed data on extra-class activities from 674.699 late-adolescents applying for university courses between 1987 and 2004, mostly between 17 and 19 years of age; approximately half men and half women. We found that 27% of individuals reported that Artistic/cultural activities were the leisure-time activity they participated in most frequently, and 32% reported they spent the longest period of free-time doing Artistic-activities (theater/cinema, music, dance, art-craft/plastic arts). Interestingly, from this whole sample, only less than 3% actually applied for artistic careers, which suggests that the prevalence of prioritizing artistic activities is higher than commonly assumed and includes not only professional artists, but also many hobbyists, amateurs and dedicated fans. Further, artistic careers applicants prioritize art almost three times more than the total of applicants, suggesting its specificity. After controlling for inconsistency of answer options during the period, prioritizing both Artistic/cultural and Artistic-activities remained temporally stable, as predicted. Despite limitations, overall results supported the hypotheses that artistic behavior is more intrinsically motivated, domain specific, and temporally stable. This plausibly demonstrates that artistic propensity has at least partly an evolved nature.

Is inventory performance helping to improve SME credit ratings? The moderating role of supply chain concentration

ABSTRACT This study examines the relationship between inventory performance and credit ratings and explores how this relationship is affected by supply chain concentration in the context of small- and medium-sized enterprises (SMEs). Financial indicators and bond rating data for SMEs in China’s manufacturing sector are obtained from two large public data sources, and hierarchical ordered logistic regression analysis is used to investigate our research questions. The findings show that inventory performance is positively associated with SME credit ratings and that this relationship is positively moderated by customer concentration but negatively moderated by supplier concentration. This study provides valuable theoretical contributions to the interface of supply chain and finance research and offers important managerial implications that SMEs can follow to increase their credit ratings.

Headache classification and automatic biomarker extraction from structural MRIs using deep learning.

Data-driven machine-learning methods on neuroimaging (e.g. MRI) are of great interest for the investigation and classification of neurological diseases. However, traditional machine learning requires domain knowledge to delineate the brain regions first, followed by feature extraction from the regions. Compared with this semi-automated approach, recently developed deep learning methods have advantages since they do not require such prior knowledge; instead, deep learning methods can automatically find features that differentiate MRIs from different cohorts. In the present study, we developed a deep learning-based classification pipeline distinguishing brain MRIs of individuals with one of three types of headaches [migraine (n = 95), acute post-traumatic headache (n = 48) and persistent post-traumatic headache (n = 49)] from those of healthy controls (n = 532) and identified the brain regions that most contributed to each classification task. Our pipeline included: (i) data preprocessing; (ii) binary classification of healthy controls versus headache type using a 3D ResNet-18; and (iii) biomarker extraction from the trained 3D ResNet-18. During the classification at the second step of our pipeline, we resolved two common issues in deep learning methods, limited training data and imbalanced samples from different categories, by incorporating a large public data set and resampling among the headache cohorts. Our method achieved the following classification accuracies when tested on independent test sets: (i) migraine versus healthy controls-75% accuracy, 66.7% sensitivity and 83.3% specificity; (2) acute post-traumatic headache versus healthy controls-75% accuracy, 66.7% sensitivity and 83.3% specificity; and (3) persistent post-traumatic headache versus healthy controls-91.7% accuracy, 100% sensitivity and 83.3% specificity. The most significant biomarkers identified by the classifier for migraine were caudate, caudal anterior cingulate, superior frontal, thalamus and ventral diencephalon. For acute post-traumatic headache, lateral occipital, cuneus, lingual, pericalcarine and superior parietal regions were identified as most significant biomarkers. Finally, for persistent post-traumatic headache, the most significant biomarkers were cerebellum, middle temporal, inferior temporal, inferior parietal and superior parietal. In conclusion, our study shows that the deep learning methods can automatically detect aberrations in the brain regions associated with different headache types. It does not require any human knowledge as input which significantly reduces human effort. It uncovers the great potential of deep learning methods for classification and automatic extraction of brain imaging-based biomarkers for these headache types.

Electroencephalograph-Based Emotion Recognition Using Brain Connectivity Feature and Domain Adaptive Residual Convolution Model.

In electroencephalograph (EEG) emotion recognition research, obtaining high-level emotional features with more discriminative information has become the key to improving the classification performance. This study proposes a new end-to-end emotion recognition method based on brain connectivity (BC) features and domain adaptive residual convolutional network (short for BC-DA-RCNN), which could effectively extract the spatial connectivity information related to the emotional state of the human brain and introduce domain adaptation to achieve accurate emotion recognition within and across the subject’s EEG signals. The BC information is represented by the global brain network connectivity matrix. The DA-RCNN is used to extract high-level emotional features between different dimensions of EEG signals, reduce the domain offset between different subjects, and strengthen the common features between different subjects. The experimental results on the large public DEAP data set show that the accuracy of the subject-dependent and subject-independent binary emotion classification in valence reaches 95.15 and 88.28%, respectively, which outperforms all the benchmark methods. The proposed method is proven to have lower complexity, better generalization ability, and domain robustness that help to lay a solid foundation for the development of high-performance affective brain-computer interface applications.

Dnabarcoder: An open-source software package for analysing and predicting DNA sequence similarity cutoffs for fungal sequence identification.

The accuracy and precision of fungal molecular identification and classification are challenging, particularly in environmental metabarcoding approaches as these often trade accuracy for efficiency given the large data volumes at hand. In most ecological studies, only a single similarity cutoff value is used for sequence identification. This is not sufficient since the most commonly used DNA markers are known to vary widely in terms of inter‐ and intraspecific variability. We address this problem by presenting a new tool, dnabarcoder, to predict local similarity cutoffs and measure the resolving powers of a biomarker for sequence identification for different clades of fungi. It was shown that the predicted similarity cutoffs varied significantly between the clades of a recently released ITS DNA barcode data set from the CBS culture collection of the Westerdijk Fungal Biodiversity Institute. When classifying a large public fungal ITS data set—the UNITE database—against the barcode data set, the local similarity cutoffs assigned fewer sequences than the traditional cutoffs used in metabarcoding studies. However, the obtained accuracy and precision were significantly improved. Our study showed that it might be better to extract the ITS region from the ITS barcodes to optimize taxonomic assignment accuracy. Furthermore, 15.3, 25.6, and 26.3% of the fungal species of the barcode data set were indistinguishable by full‐length ITS, ITS1, and ITS2, respectively. Except for these indistinguishable species, the resolving powers of full‐length ITS, ITS1, and ITS2 sequences were similar at the species level. Nevertheless, the complete ITS region had a better resolving power at higher taxonomic levels.