Deep Bayesian active learning-to-rank with relative annotation for estimation of ulcerative colitis severity.

Automatic image-based severity estimation is an important task in computer-aided diagnosis. Severity estimation by deep learning requires a large amount of training data to achieve a high performance. In general, severity estimation uses training data annotated with discrete (i.e., quantized) severity labels. Annotating discrete labels is often difficult in images with ambiguous severity, and the annotation cost is high. In contrast, relative annotation, in which the severity between a pair of images is compared, can avoid quantizing severity and thus makes it easier. We can estimate relative disease severity using a learning-to-rank framework with relative annotations, but relative annotation has the problem of the enormous number of pairs that can be annotated. Therefore, the selection of appropriate pairs is essential for relative annotation. In this paper, we propose a deep Bayesian active learning-to-rank that automatically selects appropriate pairs for relative annotation. Our method preferentially annotates unlabeled pairs with high learning efficiency from the model uncertainty of the samples. We prove the theoretical basis for adapting Bayesian neural networks to pairwise learning-to-rank and demonstrate the efficiency of our method through experiments on endoscopic images of ulcerative colitis on both private and public datasets. We also show that our method achieves a high performance under conditions of significant class imbalance because it automatically selects samples from the minority classes.

A Hybrid of Generative and Discriminative Models Based on the Gaussian-Coupled Softmax Layer.

Generative models offer advantageous characteristics for classification tasks, such as the availability of unsupervised data and calibrated confidence. In contrast, discriminative models have advantages in terms of their potential to outperform their generative counterparts and the simplicity of their model structures and learning algorithms. In this article, we propose a method to train a hybrid of discriminative and generative models in a single neural network (NN), which exhibits the characteristics of both models. The key idea is the Gaussian-coupled softmax layer, which is a fully connected layer with a softmax activation function coupled with Gaussian distributions. This layer can be embedded into an NN-based classifier and allows the classifier to estimate both the class posterior distribution and the input data distribution. We demonstrate that the proposed hybrid model can be applied to semi-supervised learning and confidence calibration.

Housing Systems Affect Eggshell Lightness and Free Amino Acid Contents of Egg Albumen in Tosa-Jidori Chickens: A Preliminary Research.

Many countries have gradually shifted to animal welfare-friendly housing systems for egg layers. However, there is still no consensus among researchers on whether the housing system affects egg quality traits. Therefore, this study aimed to determine the effects of housing systems on egg traits and free amino acid contents of albumen and yolk using two types of housing systems, the conventional cage (cage) system and a floor rearing (litter) system. Tosa-jidori (n = 20) hens were divided into two groups. Experiments during the 7 weeks were performed twice by switching the housing systems (first and second stages). One-way analysis of variance was used to evaluate the effects of housing systems on body weight gain, egg traits, albumen and yolk amino acid contents, and fecal corticosterone. We observed significant housing effects in body weight gain, eggshell weight, yolk weight, eggshell thickness, eggshell lightness, and several albumen amino acids (A_Gln, A_His, A_Met, A_Cys, A_Lys, A_Asp, A_Glu, A_Ser, A_Thr, A_Ala, A_Pro, and A_Phe). Notably, a robust effect was seen in eggshell lightness, even after switching housing systems. These results suggest that eggshell lightness and several egg traits, including albumen amino acid contents, can be changed by using the different housing systems.

Expression of chemerin in intestinal mucosa of calves with comparable expression level with other antimicrobial proteins.

Neonatal calves are highly susceptible to infectious disorders including diarrhea. Therefore, epithelial innate immunity, including antimicrobial peptides/proteins (AMPs), is important during the early stage of their lives. Chemerin, a multifunctional protein that was originally identified as a chemokine, possesses a potent antimicrobial activity. The present study investigated the expression levels of chemerin in the gastrointestinal (GI) tract of growing calves. Chemerin and its coding gene, retinoic acid receptor responder protein 2 (RARRES2), were highly expressed in duodenum, jejunum, and ileum compared with other parts of the GI tract. Immunohistochemistry demonstrated that chemerin-producing cells were localized in the crypt of the intestinal mucosa. Finally, the expression level of RARRES2 was higher compared with those of other major AMPs in duodenum, although it was lower compared with that of enteric ß-defensin but mostly higher than those of other AMPs in jejunum and ileum at various ages in calves. The expression levels of RARRES2 were not influenced by the age of calves in duodenum and jejunum, whereas a higher expression level of RARRES2 in ileum was observed in younger calves. This study revealed that chemerin is produced in the small intestine of calves and has the potential to contribute to the gut epithelial barrier system.

Neurotensin and xenin stimulates pancreatic exocrine secretion through the peripheral cholinergic nerves in conscious sheep.

The present study aimed to clarify the effects of neurotensin and xenin on pancreatic exocrine secretion in conscious sheep and their mechanism of actions. The animals were equipped with two silastic cannulae in the common bile duct to separately collect pancreatic fluid and bile, and a silastic cannula in the proximal duodenum to continuously return the mixed fluids. NT and xenin were intravenously injected at range of 0.01-3.0 nmol/kg during the phase I of duodenal migrating motor complex. A single intravenous NT injection significantly and dose-dependently increased pancreatic fluid, protein, and bicarbonate outputs. The effect of NT at 1 nmol/kg was completely inhibited by a background intravenous infusion of atropine methyl nitrate at a dose of 10 nmol/kg/min, however, the effect was not altered by a prior injection of the neurotensin receptor subtype (NTR)-1 antagonist SR 48692 at 60 nmol/kg. Moreover, a single intravenous xenin-25 injection significantly and dose-dependently increased pancreatic fluid and protein output, whereas the effect of xenin-25 did not clearly show dose-dependence. The prior SR 48692 injection at 30 nmol/kg did not significantly alter the effects of xenin-25 at 0.3 nmol/kg, while the atropine infusion significantly inhibited the increase in fluid secretion. Under the atropine infusion, xenin-25 at 0.3 nmol/kg did not increase protein and bicarbonate outputs, whereas the inhibitory effect of the atropine was not significant compared to that of the single injection of xenin-25. A single intravenous injection of NTR-2 agonist levocabastine at 0.1-3 nmol/kg did not alter pancreatic exocrine secretion. These results suggest that both NT and xenin-25 effectively stimulates pancreatic exocrine secretion through the peripheral cholinergic system in sheep and that NTR-2 is not involved in the regulation of pancreatic exocrine secretion, however, we did not precisely determine the role of NTR-1 in the actions of both the peptides on pancreatic exocrine secretion.

Soft and self constrained clustering for group-based labeling.

When using deep neural networks in medical image classification tasks, it is mandatory to prepare a large-scale labeled image set, and this often requires significant effort by medical experts. One strategy to reduce the labeling cost is group-based labeling, where image samples are clustered and then a label is attached to each cluster. The efficiency of this strategy depends on the purity of the clusters. Constrained clustering is an effective way to improve the purity of the clusters if we can give appropriate must-links and cannot-links as constraints. However, for medical image clustering, the conventional constrained clustering methods encounter two issues. The first issue is that constraints are not always appropriate due to the gap between semantic and visual similarities. The second issue is that attaching constraints requires extra effort from medical experts. To deal with the first issue, we propose a novel soft-constrained clustering method, which has the ability to ignore inappropriate constraints. To deal with the second issue, we propose a self-constrained clustering method that utilizes prior knowledge about the target images to set the constraints automatically. Experiments with the endoscopic image datasets demonstrated that the proposed methods give clustering results with higher purity.

Effects of xenin-25 on insulin and glucagon secretions in healthy conscious sheep.

The aim of present study was to determine effect of an intravenous injection of xenin-25 on insulin and glucagon secretion in healthy conscious sheep. After feeding once at 17:00, the experiment was started from 9:00 on the next day. Xenin-25 was intravenously (i.v.) injected at a dose of 100 to 1000 pmol/kg with and without the simultaneous injection of glucose at a dose of 200 µmol/kg, and blood was withdrawn before and after the injections. A single xenin-25 injection at 100 and 300 pmol/kg significantly increased the plasma insulin concentration, whereas the 1000 pmol/kg dose did not elicit significantly enhanced insulin response. Plasma glucose and glucagon concentrations did not significantly change after a single xenin-25 injection. Xenin-25 injection significantly and dose-dependently augmented the glucose-induced insulin secretion. However, the changes in the plasma glucose and glucagon level after the glucose injection were not altered by xenin injection. A prior intravenous injection of the neurotensin receptor subtype-1 (NTR-1) antagonist SR 48692 at 100 nmol/kg did not modify the glucose-induced change in plasma insulin caused by xenin-25 at 300 pmol/kg, and intravenous injection of the NTR-2 agonist levocabastine at 1000 pmol/kg did not augment the insulin response to the glucose injection. On the other hand, no xenin-25 immunopositive cells were detected in the ovine pancreas. The mRNAs of the three NTR subtypes were highly expressed in the ovine pancreas in comparison with the expression in the abomasum. These results suggest that xenin-25 released from the upper gastrointestinal tract plays a role of an insulinotropic factor in sheep, possibly through NTRs in the pancreatic islets, but not via NTR-2.

Effect of growth and parturition on hair cortisol in Holstein cattle.

The effect of growth and parturition on hair cortisol concentrations of cattle was investigated. Plasma, saliva, and hair (black and white from the shoulders and hip) samples were collected from calves at 6 and 24 weeks old and from dairy cattle at the dry (1 and 2 months prepartum) and lactation (10, 50, 150, and 250 days postpartum) periods. Plasma and saliva cortisol concentrations were lower in 24-week-old calves than those of 6-week-old calves, and hair cortisol concentrations decreased regardless of color and position. In 6-week-old calves, hair cortisol concentrations differed between sampling positions, but this difference was not observed in 24-week-old calves. Plasma and saliva cortisol concentrations increased before parturition until 10 days postpartum then decreased until 50 days postpartum. The same trend was observed in the cortisol concentrations of white hair. Contrarily, cortisol concentrations in black hair remained unchanged and was lower than that in white hair. Hair cortisol concentration can vary greatly depending on the location on the body, hair color, cattle age, or parturition. When this method is used, all of the above factors must be considered.

Forestomach developmental failure in an 11-month-old Japanese Black steer with severely retarded growth and chronic ruminal tympany.

This study reports findings from the pathological examination of the forestomach of an 11-month-old Japanese Black steer with severely retarded growth (41% of expected weight) and chronic ruminal tympany. The ruminal papillae were weakly formed (0.3-0.5 cm long) and unevenly distributed. The cellulae and cristae reticuli were underdeveloped; the cristae were 0.4-0.7 cm in height and milky white. The keratinized layer in the stratified squamous epithelium was thickened. Ruminal pH was 5.25, and ruminal volatile fatty acid concentration was 11.7 mM. The steer's severely retarded growth was considered to be caused by malnutrition due to developmental and functional failure of the forestomach.

Longitudinal assessment of U-shaped and inverted U-shaped developmental changes in the spontaneous movements of infants via markerless video analysis.

Various attempts have been made to elucidate the development patterns in the spontaneous movements of infants through longitudinal evaluations. Movement complexity has been found to demonstrate u-shaped changes in the measurements focusing on limb movements. However, researchers have not yet clarified how other characteristics, besides movement complexity, change over time. This paper presents a longitudinal evaluation of spontaneous movements in infants using evaluation indices calculated through markerless video analysis. Nine infants with corrected ages from [Formula: see text] to 15 weeks participated in the experiments. We confirmed the change in indices over time using statistical methods. Index changes can be classified as positively correlated, u-shaped, inverted u-shaped, and uncorrelated. We also confirmed that the u-shaped and inverted u-shaped indices are negatively correlated. Furthermore, the principal component analysis revealed that the first principal component had the inverted u-shaped changes with the corrected age. These results suggest that it is important to synchronize the inverted u-shaped variations in the movement and velocity with the u-shaped changes in the movement complexity for infant development.

Video-based evaluation of infant crawling toward quantitative assessment of motor development.

This study proposes a quantitative evaluation support system for infant motor development and uses the system to analyze hands-and-knees creeping and belly crawling. This system measures movements using two video cameras and extracts movement features via background and inter-frame subtractions of original images. Eight evaluation indices for each crawling cycle are calculated, enabling markerless movement analysis of infants. Cross-sectional analysis of 16 10-month-olds confirmed significant differences between hands-and-knees creeping and belly crawling in five of the eight indices, demonstrating the system capability to quantitatively differentiate between creeping and crawling. Longitudinal analysis of one infant (aged 7-10 months) also suggested that the proposed quantitative indices can follow changes in crawling characteristics and evaluate infants' motor development process. The results from the experiments suggest that the proposed system may enable diagnosis support in clinical practice.

Leptin and ghrelin expressions in the gastrointestinal tracts of calves and cows.

This study aims to investigate and compare the expressions of leptin and ghrelin in the gastrointestinal tracts of calves and cows. The mRNA expression of leptin in the rumen, abomasum, and jejunum of calves was significantly higher than that in cows. In both calves and cows, abomasum ghrelin mRNA expression was significantly higher than that in other gastrointestinal tracts. In calves, leptin protein expression in the abomasum was the highest. In addition, leptin protein expression in the abomasum and jejunum of calves was significantly higher than that in cows. Results indicated that leptin in the abomasum and jejunum plays an important role during the suckling period in a ruminant.

Markerless Measurement and Evaluation of General Movements in Infants.

General movements (GMs), a type of spontaneous movement, have been used for the early diagnosis of infant disorders. In clinical practice, GMs are visually assessed by qualified licensees; however, this presents a difficulty in terms of quantitative evaluation. Various measurement systems for the quantitative evaluation of GMs track target markers attached to infants; however, these markers may disturb infants' spontaneous movements. This paper proposes a markerless movement measurement and evaluation system for GMs in infants. The proposed system calculates 25 indices related to GMs, including the magnitude and rhythm of movements, by video analysis, that is, by calculating background subtractions and frame differences. Movement classification is performed based on the clinical definition of GMs by using an artificial neural network with a stochastic structure. This supports the assessment of GMs and early diagnoses of disabilities in infants. In a series of experiments, the proposed system is applied to movement evaluation and classification in full-term infants and low-birth-weight infants. The experimental results confirm that the average agreement between four GMs classified by the proposed system and those identified by a licensee reaches up to 83.1 ± 1.84%. In addition, the classification accuracy of normal and abnormal movements reaches 90.2 ± 0.94%.

Total synthesis of verucopeptin, an inhibitor of hypoxia-inducible factor 1 (HIF-1).

Verucopeptin is an inhibitor of hypoxia-inducible factor 1 (HIF-1), which is a promising target for cancer chemotherapy. Here, we report the first total synthesis of verucopeptin via condensation of the depsipeptide core and the polyketide side chain unit including three branched methyl groups after the synthesis of each segment.

A Scale Mixture-Based Stochastic Model of Surface EMG Signals With Variable Variances.

OBJECTIVE: Surface electromyogram (EMG) signals have typically been assumed to follow a Gaussian distribution. However, the presence of non-Gaussian signals associated with muscle activity has been reported in recent studies, and there is no general model of the distribution of EMG signals that can explain both non-Gaussian and Gaussian distributions within a unified scheme. METHODS: In this paper, we describe the formulation of a non-Gaussian EMG model based on a scale mixture distribution. In the model, an EMG signal at a certain time follows a Gaussian distribution, and its variance is handled as a random variable that follows an inverse gamma distribution. Accordingly, the probability distribution of EMG signals is assumed to be a mixture of Gaussians with the same mean but different variances. The EMG variance distribution is estimated via marginal likelihood maximization. RESULTS: Experiments involving nine participants revealed that the proposed model provides a better fit to recorded EMG signals than conventional EMG models. It was also shown that variance distribution parameters may reflect underlying motor unit activity. CONCLUSION: This study proposed a scale mixture distribution-based stochastic EMG model capable of representing changes in non-Gaussianity associated with muscle activity. A series of experiments demonstrated the validity of the model and highlighted the relationship between the variance distribution and muscle force. SIGNIFICANCE: The proposed model helps to clarify conventional wisdom regarding the probability distribution of surface EMG signals within a unified scheme.

Endoscopic Image Clustering with Temporal Ordering Information Based on Dynamic Programming.

In this paper, we propose a clustering method with temporal ordering information for endoscopic image sequences. It is difficult to collect a sufficient amount of endoscopic image datasets to train machine learning techniques by manual labeling. The clustering of endoscopic images leads to group-based labeling, which is useful for reducing the cost of dataset construction. Therefore, in this paper, we propose a clustering method where the property of endoscopic image sequences is fully utilized. For the proposed method, a deep neural network was used to extract features from endoscopic images, and clustering with temporal ordering information was solved by dynamic programming. In the experiments, we clustered the esophagogastroduodenoscopy images. From the results, we confirmed that the performance was improved by using the sequential property.

An EMG Pattern Classification Method Based on a Mixture of Variance Distribution Models.

This paper proposes an electromyogram (EMG) pattern classification method based on a mixture of variance distribution models. A variance distribution model is a stochastic model of raw surface EMG signals in which the EMG variance is taken as a random variable, allowing the representation of uncertainty in the variance. In this paper, we extend the variance distribution model to the multidimensional case and enhance its flexibility for multichannel and processed EMG signals. The enhanced model enables the accurate classification of EMG patterns while considering the uncertainty in the EMG variance. The robustness and applicability of the proposed method are demonstrated through a simulation experiment using artificially generated data and EMG classification experiments using two real datasets.

Biosignal Data Augmentation Based on Generative Adversarial Networks.

In this paper, we propose a synthetic generationmethod for time-series data based on generative adversarial networks (GANs) and apply it to data augmentation for biosinal classification. GANs are a recently proposed framework for learning a generative model, where two neural networks, one generating synthetic data and the other discriminating synthetic and real data, are trained while competing with each other. In the proposed method, each neural network in GANs is developed based on a recurrent neural network using long short-term memories, thereby allowing the adaptation of the GANs framework to time-series data generation. In the experiments, we confirmed the capability of the proposed method for generating synthetic biosignals using the electrocardiogram and electroencephalogram datasets. We also showed the effectiveness of the proposed method for data augmentation in the biosignal classification problem.

Anti-Depressant Fluoxetine Reveals its Therapeutic Effect Via Astrocytes.

Although psychotropic drugs act on neurons and glial cells, how glia respond, and whether glial responses are involved in therapeutic effects are poorly understood. Here, we show that fluoxetine (FLX), an anti-depressant, mediates its anti-depressive effect by increasing the gliotransmission of ATP. FLX increased ATP exocytosis via vesicular nucleotide transporter (VNUT). FLX-induced anti-depressive behavior was decreased in astrocyte-selective VNUT-knockout mice or when VNUT was deleted in mice, but it was increased when astrocyte-selective VNUT was overexpressed in mice. This suggests that VNUT-dependent astrocytic ATP exocytosis has a critical role in the therapeutic effect of FLX. Released ATP and its metabolite adenosine act on P2Y11 and adenosine A2b receptors expressed by astrocytes, causing an increase in brain-derived neurotrophic factor in astrocytes. These findings suggest that in addition to neurons, FLX acts on astrocytes and mediates its therapeutic effects by increasing ATP gliotransmission.

Quantitative Blood Flow Assessment by Multiparameter Analysis of Indocyanine Green Video Angiography.

BACKGROUND: Measurements of quantitative blood flow are crucial during brain vascular surgery. Indocyanine green video angiography (ICG-VAG) is an accepted method of blood flow visualization; however, quantitative techniques have not yet been established. Thus, the aim of this study was to further develop ICG analysis for visualizing intraoperative flow changes. METHODS: We conducted basic experiments and clinical investigations to establish a relationship between ICG-VAG and measured blood flow. We evaluated several parameters and identified optimal indicators that precisely reflect blood (or fluid) flow. Both in vitro and in vivo studies were performed to calculate the interval between baseline and the intensity peak (Grad) and to measure actual flow rate. RESULTS: Grad and actual flow rate showed good exponential correlation, with R2 values of 0.90 in vitro and 0.82 in vivo. In a representative patient (case 3), we performed intraoperative flow analysis using FlowInsight, which identified a marked elevation in Grad on the brain surface. Because this observation is predictive of brain hyperperfusion, we used these data to carefully manage blood pressure postoperatively. CONCLUSIONS: Grad is the optimum parameter for estimating flow conditions. Although ICG-VAG provides only visual profiles of blood circulation in the brain, this procedure has the potential to be widely used in clinical situations. ICG-based flow measurement can be used to identify normal and abnormal blood flow conditions, such as graft flow and vascular pathology. The novelty of this technique is that the fluorescence intensity of Grad enables surgeons to quantitatively measure real blood flow.