Robust Deep Neural Network for Learning in Noisy Multi-Label Food Images.

Deep networks can facilitate the monitoring of a balanced diet to help prevent various health problems related to eating disorders. Large, diverse, and clean data are essential for learning these types of algorithms. Although data can be collected automatically, the data cleaning process is time-consuming. This study aims to provide the model with the ability to learn even when the data are not completely clean. For this purpose, we extend the Attentive Feature MixUp method to enable its learning on noisy multi-label food data. The extension was based on the hypothesis that during the MixUp phase, when a pair of images are mixed, the resulting soft labels should be different for each ingredient, being larger for ingredients that are mixed with the background because they are better distinguished than when they are mixed with other ingredients. Furthermore, to address data perturbation, the incorporation of the Laplace approximation as a post-hoc method was analyzed. The evaluation of the proposed method was performed on two food datasets, where a notable performance improvement was obtained in terms of Jaccard index and F1 score, which validated the hypothesis raised. With the proposed MixUp, our method reduces the memorization of noisy multi-labels, thereby improving its performance.

Antibiomania: A Case Report of a Manic Episode Potentially Induced by the Interaction of Clarithromycin and Amoxicillin during H. Pylori Eradication Therapy.

Antibiomanic episodes, or as the DSM-5 refers to them, drug-induced manic episodes, pose a clinical challenge that is still poorly understood. There is insufficient information on the most common clinical presentation, patient profile, or underlying aetiopathogenic mechanisms. We present the clinical case of a 67-year-old woman who, after starting treatment (clarithromycin and amoxicillin) for the eradication of Helicobacter pylori, bacteria presented with a brief manic episode, which resolved after withdrawal of both drugs and with antipsychotic treatment. The possible interaction of both drugs, as GABA antagonists, in the generation of such episodes is discussed, and the clinical importance of such episodes in psychiatric emergency departments and liaison and interconsultation psychiatry, is highlighted.

Bayesian DivideMix++ for Enhanced Learning with Noisy Labels.

Leveraging inexpensive and human intervention-based annotating methodologies, such as crowdsourcing and web crawling, often leads to datasets with noisy labels. Noisy labels can have a detrimental impact on the performance and generalization of deep neural networks. Robust models that are able to handle and mitigate the effect of these noisy labels are thus essential. In this work, we explore the open challenges of neural network memorization and uncertainty in creating robust learning algorithms with noisy labels. To overcome them, we propose a novel framework called "Bayesian DivideMix++" with two critical components: (i) DivideMix++, to enhance the robustness against memorization and (ii) Monte-Carlo MixMatch, which focuses on improving the effectiveness towards label uncertainty. DivideMix++ improves the pipeline by integrating the warm-up and augmentation pipeline with self-supervised pre-training and dedicated different data augmentations for loss analysis and backpropagation. Monte-Carlo MixMatch leverages uncertainty measurements to mitigate the influence of uncertain samples by reducing their weight in the data augmentation MixMatch step. We validate our proposed pipeline using four datasets encompassing various synthetic and real-world noise settings. We demonstrate the effectiveness and merits of our proposed pipeline using extensive experiments. Bayesian DivideMix++ outperforms the state-of-the-art models by considerable differences in all experiments. Our findings underscore the potential of leveraging these modifications to enhance the performance and generalization of deep neural networks in practical scenarios.

Antibiomania: A Case Report of a Manic Episode Potentially Induced by the Interaction of Clarithromycin and Amoxicillin during H. Pylori Eradication Therapy

Antibiomanic episodes, or as the DSM-5 refers to them, drug-induced manic episodes, pose a clinical challenge that is still poorly understood. There is insufficient information on the most common clinical presentation, patient profile, or underlying aetiopathogenic mechanisms. We present the clinical case of a 67-year-old woman who, after starting treatment (clarithromycin and amoxicillin) for the eradication of Helicobacter pylori, bacteria presented with a brief manic episode, which resolved after withdrawal of both drugs and with antipsychotic treatment. The possible interaction of both drugs, as GABA antagonists, in the generation of such episodes is discussed, and the clinical importance of such episodes in psychiatric emergency departments and liaison and interconsultation psychiatry, is highlighted. (AU)

Progressive loss of cortical gray matter in first episode psychosis patients with auditory hallucinations.

BACKGROUND: Previous longitudinal magnetic resonance imaging studies have shown progressive gray matter (GM) reduction during the earliest phases of schizophrenia. It is unknown whether these progressive processes are homogeneous in all groups of patients. One way to obtain more valid findings is to focus on the symptoms. Auditory hallucinations (AHs) are frequent and reliable symptoms of psychosis. The present study aims to analyze whether longitudinal changes in structural abnormalities in cortical regions are related to the presence of AHs and the intensity of psychotic symptoms in a large sample. METHODS: A Magnetic Resonance (MR) voxel-based morphometry analysis was applied to a group of 128 first episodes psychosis (FEP) patients (63 patients with AHs and 65 patients without AHs) and 78 matched healthy controls at baseline and at a 2-year follow-up. RESULTS: At baseline, FEP patients exhibited significant GM volume reductions in the temporal, frontal and precentral regions. At follow-up, FEP patients exhibited GM volume changes in the temporal, Rolandic, frontal, precentral and insula regions. At baseline, no significant differences were found between FEP patients with and without AHs. At follow-up, while FEP patients with AHs showed less GM volume in temporal and frontal lobes, non-AH FEP patients showed reductions in the frontal, precentral and fusiform areas. PANSS scores showed statistically significant correlations with GM volume reductions at baseline and follow-up. CONCLUSIONS: Brain cortical loss in the early phases of psychosis is not associated with potentially transitory AHs; however, brain structural changes may emerge as AHs appear in chronic patients.

Volume alterations of the hippocampus and amygdala in patients with schizophrenia and persistent auditory hallucinations.

INTRODUCTION: Auditory hallucinations (AH) are one of the most prevalent symptoms of schizophrenia. They might cause several brain alterations, especially changes in the volumes of hippocampus and amygdala, regions related to the relay and processing of auditory cues and emotional memories. MATERIAL AND METHODS: We have recruited 41 patients with schizophrenia and persistent AH, 35 patients without AH, and 55 healthy controls. Using their MRIs, we have performed semiautomatic segmentations of the hippocampus and amygdala using Freesurfer. We have also performed bilateral correlations between the total PSYRATS score and the volumes of affected subregions and nuclei. RESULTS: In the hippocampus, we found bilateral increases in the volume of its hippocampal fissure and decreases in the right fimbria in patients with and without AH. The volume of the right hippocampal tail and left head of the granule cell layer from the dentate gyrus were decreased in patients with AH. In the amygdala, we found its left total volume was shrunk, and there was a decrease of its left accessory basal nucleus in patients with AH. CONCLUSIONS: We have detected volume alterations of different limbic structures likely due to the presence of AH. The volumes of the right hippocampal tail and left head of the granule cell layer from the dentate gyrus, and total volume of the amygdala and its accessory basal nucleus, were only affected in patients with AH. Bilateral volume alterations in the hippocampal fissure and right fimbria seem inherent of schizophrenia and due to traits not contemplated in our research.

Common genetic variants contribute to heritability of age at onset of schizophrenia.

Schizophrenia (SCZ) is a complex disorder that typically arises in late adolescence or early adulthood. Age at onset (AAO) of SCZ is associated with long-term outcomes of the disease. We explored the genetic architecture of AAO with a genome-wide association study (GWAS), heritability, polygenic risk score (PRS), and copy number variant (CNV) analyses in 4 740 subjects of European ancestry. Although no genome-wide significant locus was identified, SNP-based heritability of AAO was estimated to be between 17 and 21%, indicating a moderate contribution of common variants. We also performed cross-trait PRS analyses with a set of mental disorders and identified a negative association between AAO and common variants for SCZ, childhood maltreatment and attention-deficit/hyperactivity disorder. We also investigated the role of copy number variants (CNVs) in AAO and found an association with the length and number of deletions (P-value = 0.03), whereas the presence of CNVs previously reported in SCZ was not associated with earlier onset. To our knowledge, this is the largest GWAS of AAO of SCZ to date in individuals from European ancestry, and the first study to determine the involvement of common variants in the heritability of AAO. Finally, we evidenced the role played by higher SCZ load in determining AAO but discarded the role of pathogenic CNVs. Altogether, these results shed light on the genetic architecture of AAO, which needs to be confirmed with larger studies.

Shape complexity in cluster analysis.

In cluster analysis, a common first step is to scale the data aiming to better partition them into clusters. Even though many different techniques have throughout many years been introduced to this end, it is probably fair to say that the workhorse in this preprocessing phase has been to divide the data by the standard deviation along each dimension. Like division by the standard deviation, the great majority of scaling techniques can be said to have roots in some sort of statistical take on the data. Here we explore the use of multidimensional shapes of data, aiming to obtain scaling factors for use prior to clustering by some method, like k-means, that makes explicit use of distances between samples. We borrow from the field of cosmology and related areas the recently introduced notion of shape complexity, which in the variant we use is a relatively simple, data-dependent nonlinear function that we show can be used to help with the determination of appropriate scaling factors. Focusing on what might be called "midrange" distances, we formulate a constrained nonlinear programming problem and use it to produce candidate scaling-factor sets that can be sifted on the basis of further considerations of the data, say via expert knowledge. We give results on some iconic data sets, highlighting the strengths and potential weaknesses of the new approach. These results are generally positive across all the data sets used.

Neutrophil extracellular trap-associated carbamylation and histones trigger osteoclast formation in rheumatoid arthritis.

OBJECTIVE: Neutrophil infiltration into the synovial joint is a hallmark of rheumatoid arthritis (RA), a disease characterised by progressive bone erosion. However, the mechanisms by which neutrophils participate in bone destruction remain unclear. Carbamylation is a posttranslational modification linked to increased bone erosion in RA and we previously showed that carbamylation is present in RA neutrophil extracellular traps (NETs). However, it remains unclear whether NETs and their carbamylated protein cargo directly promote bone destruction and alter osteoclast biology. METHODS: NETs and carbamylated NETs (cNETs) were assessed for their capacity to induce osteoclast formation in CD14+ monocytes. Chemical inhibitors and neutralising antibodies were used to elucidate the pathway by which NETs induce osteoclastogenesis. HLA-DRB1*04:01 mice received intra-articular injection of cNETs for 4 weeks. Joints were isolated and assessed for osteoclast formation. Plasma and synovial fluid samples from patients with RA (n=32) were assessed for the presence of carbamylated histone, and correlations to disease specific outcomes were performed. RESULTS: We found that NETs, when cNETs, instruct monocytes to undergo rapid osteoclast formation. NET-mediated osteoclastogenesis appears to depend on Toll-like receptor 4 signalling and NET-associated proteins including histones and neutrophil elastase. In vivo, we identified that the number of osteoclasts increased following immunisation with cNETs in HLA-DRB1*04:01 transgenic mice. Furthermore, carbamylated histones are increased in plasma and synovial fluid from patients with RA and correlate with active bone resorption and inflammatory markers. CONCLUSIONS: Our results suggest that NETs have a direct role in RA-associated bone erosion by promoting osteoclast formation.

Influence of anaerobic biotransformation process of agro-industrial waste with Lactobacillus acidophilus on the rheological parameters: case of study of pig manure.

This study evaluated the rheological behavior of the pig waste biotransformation process to produce lactic acid (LA) and biomass with Lactobacillus acidophilus in a stirred reactor. In addition, cell growth, carbohydrate consumption, and LA production were measured at three different agitation speeds, 100, 150, and 200 rpm at 37 °C, with a reaction time of 52 h. During the development of the process, the kinetic and rheological parameters were obtained using the logistic, Gompertz, generalized Gompertz, Ostwald de Waele, and Herschel-Bulkley mathematical models, respectively. The substrate used was pig manure, to which molasses was added at 12% v/v to increase the concentration of carbohydrates. The results suggest that mass exchange is favorable at low agitation speeds. Nevertheless, the presence of molasses rich in carbohydrates as a carbon source modifies the characteristics of the fluid, dilatant (n > 1) at the beginning of the process to end up as pseudoplastic (n < 1) due to the addition of exopolysaccharides and the modification of the physical structure of the substrate. This effect was confirmed by the Herschel-Bulkley model, which presented a better fit to the data obtained, in addition to finding a direct relationship between viscosity and pH that can be used as variables for the control of bioconversion processes of pig manure into biomass rich in Lactobacillus acidophilus.

How to Improve Training in Descriptive Psychopathology for Psychiatry Residents: A Delphi Study.

PURPOSE: Descriptive psychopathology (DP) is the language of psychiatry and is dedicated to the description of mental symptoms. Despite its core importance, a systematic review identified a series of shortcomings in its training. This Delphi study seeks to provide guidance for better didactic and clinical training in DP. METHOD: The authors used the Delphi method in order to gather, pool, and optimize the knowledgeable opinion of a highly qualified panel of international experts on how to improve DP training. A preliminary phase with open-ended questions was the basis to elaborate a Delphi questionnaire consisting of 14 questions on didactic and clinical training, which was then used in two successive Delphi rounds. Twenty-nine international experts in DP participated throughout the study. RESULTS: A series of hierarchical lists on how to improve DP training were elaborated. Regarding didactic training, the experts valued the contents and educational methods, as well as recommended authors and texts. Regarding clinical training, the experts valued educational methods, desirable characteristics in clinical supervisors, how to improve the supervisors' expertise, useful aspects and facilitating questions for direct supervision, and the suitability of including other evaluators besides the supervisor. In the final survey, 94% of the experts considered that the Delphi method had been effective to obtain and improve their opinions. CONCLUSIONS: Insufficient direct supervision in live interviews was considered the most important problem. A series of general measures were proposed to improve DP training: (i) adapting DP training throughout residency, with introductory and advanced levels; (ii) making DP training compulsory in psychiatry curricula; (iii) assessing residents' DP knowledge and clinical use; and (iv) training the trainers/supervisors in both content (DP) and form (how to train/supervise residents). Within didactic training, epistemology of DP and contemporary and classic authors/texts were the highest rated contents, while supervised discussions based on cases, videos, or readings were the highest rated methods. Within clinical training, 8 aspects of DP that could guide the supervisor were highly rated: mental state examination, dialogue, empathy and understanding, attitude/willingness, knowledge-practice bridge, mental symptom formation, implications for decision-making, and the written report of the encounter.

Three highly variable genome regions of the four dengue virus serotypes can accurately recapitulate the CDS phylogeny.

The circulation of the four-dengue virus (DENV) serotypes has significantly increased in recent years, accompanied by an increase in viral genetic diversity. In order to conduct disease surveillance and understand DENV evolution and its effects on virus transmission and disease, efficient and accurate methods for phylogenetic classification are required. Phylogenetic analysis of different viral genes sequences is the most used method, the envelope gene (E) being the most frequently selected target. We explored the genetic variability of the four DENV serotypes throughout their complete coding sequence (CDS) of sequences available in GenBank and used genomic regions of different variability rate to recapitulate the phylogeny obtained with the DENV CDS. Our results indicate that the use of high or low variable regions accurately recapitulate the phylogeny obtained with CDS of sequences from different DENV genotypes. However, when analyzing the phylogeny of a single genotype, highly variable regions performed better in recapitulating the distance branch length, topology, and support of the CDS phylogeny. The use of three concatenated highly variable regions was not statistically different in distance branch length and support to that obtained in CDS phylogeny.•This study demonstrated the ability of highly variable regions of the DENV genome to recapitulate the phylogeny obtained with the full coding sequence (CDS).•The use of genomic regions of high or low variability did not affect the performance in recapitulating the phylogeny obtained with CDS from different genotypes. However, when phylogeny was analyzed for sequences from a single genotype, highly variable regions performed better in recapitulating the distance branch length, topology, and support of the CDS phylogeny.•The use of concatenated highly variable genome regions represent a useful option for recapitulating genome-wide phylogenies in analyses of sequences belonging to the same DENV genotype.

The association between cannabis use and facial emotion recognition in schizophrenia, siblings, and healthy controls: Results from the EUGEI study.

Schizophrenia is frequently accompanied with social cognitive disturbances. Cannabis represents one established environmental factor associated with the onset and progression of schizophrenia. The present cross-sectional study aimed to investigate the association of facial emotion recognition (FER) performance with cannabis use in 2039 patients with schizophrenia, 2141 siblings, and 2049 healthy controls (HC). FER performance was measured using the Degraded Facial Affect Recognition Task (DFAR). Better FER performance as indicated by higher DFAR-total scores was associated with lifetime regular cannabis use in schizophrenia (B = 1.36, 95% CI 0.02 to 2.69), siblings (B = 2.17, 95% CI 0.79 to 3.56), and HC (B = 3.10, 95% CI 1.14 to 5.06). No associations were found between DFAR-total and current cannabis use. Patients with schizophrenia who started to use cannabis after the age of 16 showed better FER performance than patients who started earlier (B = 2.50, 95% CI 0.15 to 4.84) and non-users (B = 3.72, 95 CI 1.96 to 5.49). Better FER performance was found also in siblings who started to use cannabis after 16 compared to non-users (B = 2.37, 95% CI 0.58 to 4.16), while HC using cannabis performed better than non-users at DFAR-total regardless of the age at onset. Our findings suggest that lifetime regular cannabis use may be associated with better FER regardless of the psychosis risk, but that FER might be moderated by age at first use in people with higher genetic risk. Longitudinal studies may clarify whether there is a cause-and-effect relationship between cannabis use and FER performance in psychotic and non-psychotic samples.

Genetic diversity and spatiotemporal dynamics of DENV-1 and DENV-2 infections during the 2012-2013 outbreak in Mexico.

Dengue fever is caused by four related dengue virus serotypes, DENV-1 to DENV-4, where each serotype comprises distinct genotypes and lineages. The last major outbreak in Mexico occurred during 2012 and 2013, when 112,698 confirmed cases were reported (DENV-1 and DENV-2 were predominant). Following partial E, NS2A and NS5 gene sequencing, based on the virus genome variability, we analyzed 396 DENV-1 and 248 DENV-2 gene sequences from serum samples from dengue acute clinical cases from 13 Mexican states, Mutations were identified, and their genetic variability estimated, along with their evolutionary relationship with DENV sequences sampled globally. DENV-1 genotype V and DENV-2 Asian-American genotype V were the only genotypes circulating during the outbreak. Mutations in NS2A and NS5 proteins were widely disseminated and suggested local emergence of new lineages. Phylogeographic analysis suggested viral spread occurred from coastal regions, and tourist destinations, such as Yucatan and Quintana Roo, which played important roles in disseminating these lineages.

Removal of carbon and nitrogen in wastewater from a poultry processing plant in a photobioreactor cultivated with the microalga Chlorella vulgaris.

This study evaluates the removal of COD and nitrogen from poultry wastewater in photobioreactors. Cell growth, the effect of light intensity (3200, 9800, and 12000 lux) and air flow (1.6, 3.2, and 4.8 L min-1) as a source of CO2 in bold basal medium and wastewater with different concentrations of COD were evaluated. The growth kinetics were modeled by using the Gompertz model and logistic model for both culture media. COD removals of up to 95% were achieved, and poultry wastewater was found to be a viable growing medium for Chlorella vulgaris. Finally, the wastewater met Mexican standards, and biomass was obtained with products valued as lipids (3.2 g lipid/100 g biomass) and proteins (342.94 mg L-1). The culture was found to have a dilatory behavior, and the rheological models of Ostwald de Waele, Ostwald de Waele linealized and Herschel Bulkley were utilized, showing a laminar behavior.

Determinants of sexual function in male patients with systemic lupus erythematosus.

OBJECTIVES: Our study aims to describe the association between SLE and sexual function, analysing demographic variables, comorbidities and other disease-related factors. As an exploratory objective, the impact of asking about sexual function during outpatient consultation was evaluated. METHODS: From 2018 to 2019, we invited sexually active men diagnosed with SLE to complete questionnaires that evaluated their sexual function and quality of life. Additionally, patients were asked if they believed they had sexual dysfunction, whether they would be interested in receiving specialized sexual care, and if they considered SLE to be detrimental to their sexual function. Epidemiological and disease-related data were retrieved from the patients' clinical records. RESULTS: We included 124 men with SLE. Twenty-two (18%) patients answered positively when asked if they believed they had sexual dysfunction. These patients had lower overall erectile function scores and lower physical function scores than those who did not consider they had sexual dysfunction. In the multivariable analysis, factors that were associated with better sexual function were high physical function (B = 0.126, p = .031), lower BMI (B = 0.53, p = .010) and the patient's perception of normal sexual function (B = 13.0, p < .001). Comorbidities associated with worse sexual function were type 2 diabetes (B = -8.1, p = .017) and a history of thrombosis (B = -5.12, p = .019). CONCLUSION: Sexual function of male patients with SLE is impaired, independently of disease activity, chronic disease damage or pharmacological treatment. A simple question about perception of sexual function in the outpatient clinic can be used to help determine which patients could benefit from a multidisciplinary intervention to improve sexual health.

Alterations in the volume of thalamic nuclei in patients with schizophrenia and persistent auditory hallucinations.

The thalamus is a subcortical structure formed by different nuclei that relay information to the neocortex. Several reports have already described alterations of this structure in patients of schizophrenia that experience auditory hallucinations. However, to date no study has addressed whether the volumes of specific thalamic nuclei are altered in chronic patients experiencing persistent auditory hallucinations. We have processed structural MRI images using Freesurfer, and have segmented them into 25 nuclei using the probabilistic atlas developed by Iglesias and collaborators (Iglesias et al., 2018). To homogenize the sample, we have matched patients of schizophrenia, with and without persistent auditory hallucinations, with control subjects, considering sex, age and their estimated intracranial volume. This rendered a group number of 41 patients experiencing persistent auditory hallucinations, 35 patients without auditory hallucinations, and 55 healthy controls. In addition, we have also correlated the volume of the altered thalamic nuclei with the total score of the PSYRATS, a clinical scale used to evaluate the positive symptoms of this disorder. We have found alterations in the volume of 8 thalamic nuclei in both cohorts of patients with schizophrenia: The medial and lateral geniculate nuclei, the anterior, inferior, and lateral pulvinar nuclei, the lateral complex and the lateral and medial mediodorsal nuclei. We have also found some significant correlations between the volume of these nuclei in patients experiencing auditory hallucinations, and the total score of the PSYRATS scale. Altogether our results indicate that volumetric alterations of thalamic nuclei involved in audition may be related to persistent auditory hallucinations in chronic schizophrenia patients, whereas alterations in nuclei related to association cortices are evident in all patients. Future studies should explore whether the structural alterations are cause or consequence of these positive symptoms and whether they are already present in first episodes of psychosis.

Uncertainty-aware selecting for an ensemble of deep food recognition models.

Deep learning is a machine learning technique that has revolutionized the research community due to its impressive results on various real-life problems. Recently, ensembles of Convolutional Neural Networks (CNN) have proven to achieve high robustness and accuracy in numerous computer vision challenges. As expected, the more models we add to the ensemble, the better performance we can obtain, but, in contrast, more computer resources are needed. Hence, the importance of deciding how many models to use and which models to select from a pool of trained models is huge. From the latter, a common strategy in deep learning is to select the models randomly or according to the results on the validation set. However, in this way models are chosen based on individual performance ignoring how they are expected to work together. Alternatively, to ensure a better complement between models, an exhaustive search can be used by evaluating the performance of several ensemble models based on different numbers and combinations of trained models. Nevertheless, this may result in being high computationally expensive. Considering that epistemic uncertainty analysis has recently been successfully employed to understand model learning, we aim to analyze whether an uncertainty-aware epistemic method can help us decide which groups of CNN models may work best. The method was validated on several food datasets and with different CNN architectures. In most cases, our proposal outperforms the results by a statistically significant range with respect to the baseline techniques and is much less computationally expensive compared to the brute-force search.