The Text to Medical Image (T2MedI) approach using latent diffusion models holds significant promise for addressing the scarcity of medical imaging data and elucidating the appearance distribution of lesions corresponding to specific patient status descriptions. Like natural image synthesis models, our investigations reveal that the T2MedI model may exhibit biases towards certain subgroups, potentially neglecting minority groups present in the training dataset. In this study, we initially developed a T2MedI model adapted from the pre-trained Imagen framework. This model employs a fixed Contrastive Language-Image Pre-training (CLIP) text encoder, with its decoder fine-tuned using medical images from the Radiology Objects in Context (ROCO) dataset. We conduct both qualitative and quantitative analyses to examine its gender bias. To address this issue, we propose a subgroup distribution alignment method during fine-tuning on a target application dataset. Specifically, this process involves an alignment loss, guided by an off-the-shelf sensitivity-subgroup classifier, which aims to synchronize the classification probabilities between the generated images and those expected in the target dataset. Additionally, we preserve image quality through a CLIP-consistency regularization term, based on a knowledge distillation framework. For evaluation purposes, we designated the BraTS18 dataset as the target, and developed a gender classifier based on brain magnetic resonance (MR) imaging slices derived from it. Our methodology significantly mitigates gender representation inconsistencies in the generated MR images, aligning them more closely with the gender distribution in the BraTS18 dataset.

Background: Manually labeling sleep stages is time-consuming and labor-intensive, making automatic sleep staging methods crucial for practical sleep monitoring. While both single- and multi-channel data are commonly used in automatic sleep staging, limited research has adequately investigated the differences in their effectiveness. Methods: In this study, four public data sets-Sleep-SC, APPLES, SHHS1, and MrOS1-are utilized, and an advanced hybrid attention neural network composed of a multi-branch convolutional neural network and the multi-head attention mechanism is employed for automatic sleep staging. Results: The experimental results show that, for sleep staging using 2-5 classes, a combination of single-channel electroencephalography (EEG) and dual-channel electrooculography (EOG) consistently outperforms single-channel EEG with single-channel EOG, which in turn outperforms single-channel EEG or single-channel EOG alone. For instance, for five-class sleep staging using the MrOS1 data set, the combination of single-channel EEG and dual-channel EOG resulted in an accuracy of 87.18%, whereas the combination of single-channel EEG and single-channel EOG yielded an accuracy of 85.77%. In comparison, single-channel EEG alone achieved an accuracy of 85.25% and single-channel EOG alone achieved an accuracy of 83.66%. Conclusions: This study highlights the significance of combining EEG and EOG signals in automatic sleep staging, while also providing valuable insights for the channel design of portable sleep monitoring devices.

The hippocampus (HPC) plays a pivotal role in fear learning and memory. Our two recent studies suggest that rapid eye movement (REM) sleep via the HPC downregulates fear memory consolidation and promotes fear extinction. However, it is not clear whether and how the dorsal and the ventral HPC regulates fear memory differently; and how the HPC in wake regulates fear memory. By chemogenetic stimulating in the HPC directly and its afferent entorhinal cortex that selectively activated the HPC in REM sleep for 3-6 h post-fear-acquisition, we found that HPC activation in REM sleep consolidated fear extinction memory. In particular, dorsal HPC (dHPC) stimulation in REM sleep virtually eliminated fear memory by enhancing fear extinction and reducing fear memory consolidation. By contrast, chemogenetic stimulating HPC afferent the supramammillary nucleus (SUM) induced 3-hr wake with HPC activation impaired fear extinction. Finally, desipramine (DMI) injection that selectively eliminated REM sleep for >6 h impaired fear extinction. Our results demonstrate that the HPC is critical for fear memory regulation; and wake HPC and REM sleep HPC have an opposite role in fear extinction of respective impairment and consolidation.

Pontine sub-laterodorsal tegmental nucleus (SLD) is crucial for REM sleep. However, the necessary role of SLD for REM sleep, cataplexy that resembles REM sleep, and emotion memory by REM sleep has remained unclear. To address these questions, we focally ablated SLD neurons using adenoviral diphtheria-toxin (DTA) approach and found that SLD lesions completely eliminated REM sleep accompanied by wake increase, significantly reduced baseline cataplexy amounts by 40% and reward (sucrose) induced cataplexy amounts by 70% and altered cataplexy EEG Fast Fourier Transform (FFT) from REM sleep-like to wake-like in orexin null (OXKO) mice. We then used OXKO animals with absence of REM sleep and OXKO controls and examined elimination of REM sleep in anxiety and fear extinction. Our resulted showed that REM sleep elimination significantly increased anxiety-like behaviors in open field test (OFT), elevated plus maze test (EPM) and defensive aggression and impaired fear extinction. The data indicate that in OXKO mice the SLD is the sole generator for REM sleep; (2) the SLD selectively mediates REM sleep cataplexy (R-cataplexy) that merges with wake cataplexy (W-cataplexy); (3) REM sleep enhances positive emotion (sucrose induced cataplexy) response, reduces negative emotion state (anxiety), and promotes fear extinction.

Taking the queen fetus as the experimental object, the physical and chemical components of the queen fetus were determined, and the enzymatic hydrolysis process of the protein of the queen fetus Was optimized by using single-factor experiments and response surface analysis. Ultrafiltration, gel filtration and other methods Were used to isolate and purify the royal jelly fetus enzymolysis product, and the in vitro antioxidant method was used to measure the in vitro antioxidant activity of the royal jelly fetus protein hydrolysate, in order to obtain antioxidant peptides with application value. Provide a basis for the development and utilization of queen fetus protein resources. And the lyophilized royal queen fetal anti aging granules were prepared to observe the effect of anti-aging granules on the learning and memory ability of D-galactose-induced aging model mice. A metabolism research method based on 1H-NMR was used. Differential metabolites Were detected in the urine of aging model mice. D-galactose-induced aging model mice had cognitive dysfunction, which showed reduced learning and memory abilities, poor spatial location ability, and decreased memory. Anti aging particles can change the cognitive function of aging model mice and improve the learning and incinory ability of D-galactosc aging model mice. Anti-aging granules can improve the disturbance of energy metabolism, amino acid metabolism and inflanunatory response-related pathways in aging model mice induced by D-galactose.

Fatigue detection is valued for people to keep mental health and prevent safety accidents. However, detecting facial fatigue, especially mild fatigue in the real world via machine vision is still a challenging issue due to lack of non-lab dataset and well-defined algorithms. In order to improve the detection capability on facial fatigue that can be used widely in daily life, this paper provided an audiovisual dataset named DLFD (daily-life fatigue dataset) which reflected people's facial fatigue state in the wild. A framework using 3D-ResNet along with non-local attention mechanism was training for extraction of local and long-range features in spatial and temporal dimensions. Then, a compacted loss function combining mean squared error and cross-entropy was designed to predict both continuous and categorical fatigue degrees. Our proposed framework has reached an average accuracy of 90.8% on validation set and 72.5% on test set for binary classification, standing a good position compared to other state-of-the-art methods. The analysis of feature map visualization revealed that our framework captured facial dynamics and attempted to build a connection with fatigue state. Our experimental results in multiple metrics proved that our framework captured some typical, micro and dynamic facial features along spatiotemporal dimensions, contributing to the mild fatigue detection in the wild.

With the rapid development of the fourth-generation technological revolution, the continuous emergence of new technologies has had a huge impact on the market, industry and individuals. The emerging technologies emerging in the fourth-generation technological revolution are mainly artificial intelligence technology, big data technology and block chain technology. The correlation between industries is getting closer and closer, which has had a significant impact on traditional industries. The application of artificial intelligence technology in financial fields such as insurance, credit investigation, asset allocation, big data risk control, etc., has caused brand-new changes in the financial industry. This article will do a simple analysis of artificial intelligence technology, summarize the current situation and development of artificial intelligence applications in the financial field, and then briefly discuss the complexity of artificial intelligence technology applications and the shortcomings of artificial intelligence technology applications, and finally propose artificial intelligence Suggestions for improvements in the application of smart technology in financial industry measures.

Environmental problems resultant from organic pollutants are a major current challenge for modern societies. White rot fungi (WRF) are well known for their extensive organic compound degradation abilities. The unique oxidative and extracellular ligninolytic systems of WRF that exhibit low substrate specificity, enable them to display a considerable ability to transform or degrade different environmental contaminants. In recent decades, WRF and their ligninolytic enzymes have been widely applied in the removal of polycyclic aromatic hydrocarbons (PAHs), pharmaceutically active compounds (PhACs), endocrine disruptor compounds (EDCs), pesticides, synthetic dyes, and other environmental pollutants, wherein promising results have been achieved. This review focuses on advances in WRF-based bioremediation of organic pollutants over the last 10 years. We comprehensively document the application of WRF and their lignocellulolytic enzymes for removing organic pollutants. Moreover, potential problems and intriguing observations that are worthy of additional research attention are highlighted. Lastly, we discuss trends in WRF-remediation system development and avenues that should be considered to advance research in the field.

The widely-used cross-entropy (CE) loss-based deep networks achieved significant progress w.r.t. the classification accuracy. However, the CE loss can essentially ignore the risk of misclassification which is usually measured by the distance between the prediction and label in a semantic hierarchical tree. In this paper, we propose to incorporate the risk-aware inter-class correlation in a discrete optimal transport (DOT) training framework by configuring its ground distance matrix. The ground distance matrix can be pre-defined following a priori of hierarchical semantic risk. Specifically, we define the tree induced error (TIE) on a hierarchical semantic tree and extend it to its increasing function from the optimization perspective. The semantic similarity in each level of a tree is integrated with the information gain. We achieve promising results on several large scale image classification tasks with a semantic tree structure in a plug and play manner.

The anti-interference, anti-interception, anti-spoofing ability and complex electromagnetic environment of the data link signal of the frequency hopping system pose formidable challenges to the reconnaissance of the frequency hopping signal. Therefore, this paper proposes a joint STFT-HOC detection method for FH data link signals. This method combines time-frequency analysis and high-order cumulant to realize effective detection of frequency hopping data link signals. Theoretical analysis and simulation experiments show that the proposed method can successfully detect the data link signal of the frequency hopping system when the signal-to-noise ratio is greater than -2 dB.

Deep neural networks are usually data-starved in real-world applications, while manually annotation can be costly—for example, the audio emotion recognition from the audio. In contrast, the continued research in image-based facial expression recognition grants us a rich source of public available labeled IFER datasets. Using images to support audio emotion recognition with limited labeled data according to their inherent correlations can be a meaningful and challenging task. This paper proposes a system that facilitates knowledge transfer from the labeled visual to the heterogeneous labeled audio domain by learning a joint distribution of examples in different modalities then the system can map an IFER example to a corresponding audio spectrogram. Next, our work reformulates the audio emotion classification into a K+1 class discriminator of GAN-based semi-supervised learning. Good semi-supervised learning requires that the generator does NOT sample from a distribution well matching the true data distribution. Therefore, we demand the generated examples are from the low-density areas of the marginal distribution in the audio spectrogram modality. Concretely, the proposed model translates image samples to audios class-wisely in the form of spectrograms. To harness the decoded samples in a sparsely distributed area and construct a tighter decision boundary, we give a solution to precisely estimate the density on feature space and incorporate low-density pieces with an annealing scheme. Our method requires the network to discriminate against the low-density data points from high-density data points throughout the classification, and we evidence that this technique effectively improves task performance. • we give a solution to precisely estimate the density on feature space and incorporate low-density pieces with an annealing scheme. • This paper proposes a semi-supervised adversarial network that facilitates knowledge transfer from the labeled visual to the heterogeneous labeled audio domain, enhancing the audio emotion recognition performance.

This paper targets for the ordinal regression/classification, which objective is to learn a rule to predict labels from a discrete but ordered set. For instance, the classification for medical diagnosis usually involves inherently ordered labels corresponding to the level of health risk. Previous multi-task classifiers on ordinal data often use several binary classification branches to compute a series of cumulative probabilities. However, these cumulative probabilities are not guaranteed to be monotonically decreasing. It also introduces a large number of hyper-parameters to be fine-tuned manually. This paper aims to eliminate or at least largely reduce the effects of those problems. We propose a simple yet efficient way to rephrase the output layer of the conventional deep neural network. Besides, in order to alleviate the effects of label noise in ordinal datasets, we propose a unimodal label regularization strategy. It also explicitly encourages the class predictions to distribute on nearby classes of ground truth. We show that our methods lead to the state-of-the-art accuracy on the medical diagnose task (e.g., Diabetic Retinopathy and Ultrasound Breast dataset) as well as the face age prediction (e.g., Adience face and MORPH Album II) with very little additional cost.

Bowel cancer, which is easily affected by diet and drugs, has some restrictive factors such as the fecal occult blood test (FOBT) in the routine detection and the high cost and inconvenience of microscopy. In order to break through these restrictive factors, a possible alternative method of FOBT is sought. In this paper, error back propagation neural network (BPNN) algorithm is used, and expression spectrum is used as an auxiliary method to detect medical images, and a colorectal cancer (CRC) diagnosis model based on neural network is constructed. The results show that the accuracy of the model on the training set and the test set are 0.943 and 0.935, respectively, the AUC reaches more than 0.95. Therefore, the CRC diagnosis model based on neural network provides a possible alternative method of FOBT. Experimental results show that the proposed algorithm have high robustness and accuracy, which meets the current clinical needs.

Deep neural networks are usually data-starved, but manually annotation can be costly in many specific tasks. For instance, the emotion recognition from the audio. However, there is a large amount of public available labeled image-based facial expression recognition datasets. How could these images help for the audio emotion recognition with limited labeled data according to their inherent correlations can be a meaningful and challenging task. In this paper, we propose a semi-supervised adversarial network that allows the knowledge transfer from the labeled videos to the heterogeneous labeled audio domain hence enhancing the audio emotion recognition performance. Specifically, face image samples are translated to the spectrograms class-wisely. To harness the translated samples in a sparsely distributed area and construct a tighter decision boundary, we propose to precisely estimate the density on feature space and incorporate the reliable low-density sample with an annealing scheme. Moreover, the unlabeled audios are collected with the high-density path in a graph representation.As a possible "recognition via generation" framework, we empirically demonstrated its effectiveness on several audio emotional recognition benchmarks. We also demonstrated its generality on recent large-scaled semisupervised domain adaptation tasks.

Person re-identification (re-ID) presents various applications in surveillance system, but most existing models are proposed under supervised framework. These methods require large amounts of annotated pedestrian data, which limits their scalability and flexibility in a new application scenario. Aiming to relax this limitation, this article exploits the attribute-invariant characteristics and domain correlations into cross-domain person re-ID, while most unsupervised methods only consider the identity features and ignore the different importance of each source image to the target domain. Specifically, this article proposes an Attribute Memory Transfer Network (AMTNet) with two major contributions of domain-balanced memory and attribute-invariant memory modules. The first domain balanced-memory integrates a domain correlation learning method to evaluate the importance of each source image to the target domain, which is involved into the transfer learning; The second attribute-invariant memory can transfer the source attribute knowledge into the target domain with preserving the identity information to conduct the re-ID process. Extensive evaluated experiments elaborate the superiority of AMTNet on two large datasets of Market-1501 and DukeMTMC-reID, compared with hand-crafted and deep learning feature-based methods.

In order to improve the transmission data rate and the three-resistance characteristics of the network ammunition communication link, this paper proposes a new communication technology with two dimensions. In this technology, orthogonal frequency division multiplexing (OFDM) is used for one-dimensional data transmission, and the G-function-driven subcarrier frequency change rule in differential frequency hopping (DFH) is used for two-dimensional data transmission, which breaks through the conventional idea of using carrier amplitude frequency phase modulation information, and establishes a new two-dimensional communication model based on DFH-OFDM. The theoretical analysis and simulation results show that the proposed two-dimensional new communication technology, without changing the original communication system and signal characteristics, effectively reduces the probability of signal interception, improves the communication system capacity, and has good three-resistance performance.

There is a large amount of public available labeled image-based facial expression recognition datasets. How could these images help for the audio emotion recognition with limited labeled data according to their inherent correlations can be a meaningful and challenging task. In this paper, we propose a semi-supervised adversarial network that allows the knowledge transfer from the labeled videos to the heterogeneous labeled audio domain hence enhancing the audio emotion recognition performance. Specifically, face image samples are translated to the spectrograms class-wisely. To harness the translated samples in a sparsely distributed area and construct a tighter decision boundary, we propose to precisely estimate the density on feature space and incorporate the reliable low-density sample with an annealing scheme. Moreover, the unlabeled audios are collected with the high-density path in a graph representation. As a possible "recognition via generation" framework, we empirically demonstrated its effectiveness on several audio emotional recognition benchmarks.

The present review presents multiple techniques in which ocular assessments may serve as a noninvasive approach for the early diagnoses of various cognitive and psychiatric disorders, such as Alzheimer's disease (AD), autism spectrum disorder (ASD), schizophrenia (SZ), and major depressive disorder (MDD). Real-time ocular responses are tightly associated with emotional and cognitive processing within the central nervous system. Patterns seen in saccades, pupillary responses, and blinking, as well as retinal microvasculature and morphology visualized via office-based ophthalmic imaging, are potential biomarkers for the screening and evaluation of cognitive and psychiatric disorders. Additionally, rapid advances in artificial intelligence (AI) present a growing opportunity to use machine-learning-based AI, especially deep-learning neural networks, to shed new light on the field of cognitive neuroscience, which may lead to novel evaluations and interventions via ocular approaches for cognitive and psychiatric disorders.

AIK is a novel cationic peptide with potential antitumor activity. In order to construct the AIK expression vector by Gateway technology, and establish an optimal expression and purification method for recombinant AIK, a set of primers containing AttB sites were designed and used to create the AttB-TEV-FLAG-AIR fusion gene by overlapping PCR. The resulting fusion gene was cloned into the donor vector pDONR223 by attB and attP mediated recombination (BP reaction), then, transferred into the destination vector pDESTl 5 by attL and attR mediated recombination (LR reaction). All the cloning was verified by both colony PCR and DNA sequencing. The BL21 F. coli transformed by the GST-AIR expression plasmid was used to express the GST-AIK fusion protein with IPTG induction and the induction conditions were optimized. GST-AIR fusion protein was purified by glutathione magnetic beads, followed by rTEV cleavage to remove GST tag and MTS assay to test the growth inhibition activity of the recombinant AIR on human leukemia HL-60 cells. We found that a high level of soluble expression of GST-AIK protein (more than 30% out of the total bacterial proteins) was achieved upon 0.1 mmol/L ITPG induction for 4 h at 37 °C in the transformed BL21 F. coli with starting OD₆₀₀ at 1.0. Through GST affinity purification and rTEV cleavage, the purity of the resulting recombinant AIK was greater than 95%. And the MTS assays on HL-60 cells confirmed that the recombinant AIK retains an antitumor activity at a level similar to the chemically synthesized AIK. Taken together, we have established a method for expression and purification of recombinant AIK with a potent activity against tumor cells, which will be beneficial for the large-scale production and application of recombinant AIK in the future.

Background Lung cancer is emerging rapidly as the leading death cause in Chinese cancer patients. The causal factors for Chinese lung cancer development remain largely unclear. Here we employed an shRNA library-based loss-of-function screen in a genome-wide and unbiased manner to interrogate potential tumor suppressor candidates in the immortalized human lung epithelial cell line BEAS-2B. Methods/Results Soft agar assays were conducted for screening BEAS-2B cells infected with the retroviral shRNA library with the acquired feature of anchorage-independent growth, large (>0.5mm in diameter) and well—separated colonies were isolated for proliferation. PCRs were performed to amplify the integrated shRNA fragment from individual genomic DNA extracted from each colony, and each PCR product is submitted for DNA sequencing to reveal the integrated shRNA and its target gene. A total of 6 candidate transformation suppressors including INPP4B, Sesn2, TIAR, ACRC, Nup210, LMTK3 were identified. We validated Sesn2 as the candidate of lung cancer tumor suppressor. Knockdown of Sesn2 by an shRNA targeting 3’ UTR of Sesn2 transcript potently stimulated the proliferation and malignant transformation of lung bronchial epithelial cell BEAS-2B via activation of Akt-mTOR-p70S6K signaling, whereas ectopic expression of Sens2 re-suppressed the malignant transformation elicited by the Sesn2 shRNA. Moreover, knockdown of Sesn2 in BEAS-2B cells promoted the BEAS-2B cell-transplanted xenograft tumor growth in nude mice. Lastly, DNA sequencing indicated mutations of Sesn2 gene are rare, the protein levels of Sesn2 of 77 Chinese lung cancer patients varies greatly compared to their adjacent normal tissues, and the low expression level of Sesn2 associates with the poor survival in these examined patients by Kaplan Meier analysis. Conclusions Our shRNA-based screen has demonstrated Sesn2 is a potential tumor suppressor in lung epithelial cells. The expression level of Sesn2 may serve as a prognostic marker for Chinese lung cancer patients in the clinic.