Uncovering Flat and Hierarchical Topics by Community Discovery on Word Co-occurrence Network.

Topic modeling aims to discover latent themes in collections of text documents. It has various applications across fields such as sociology, opinion analysis, and media studies. In such areas, it is essential to have easily interpretable, diverse, and coherent topics. An efficient topic modeling technique should accurately identify flat and hierarchical topics, especially useful in disciplines where topics can be logically arranged into a tree format. In this paper, we propose Community Topic, a novel algorithm that exploits word co-occurrence networks to mine communities and produces topics. We also evaluate the proposed approach using several metrics and compare it with usual baselines, confirming its good performances. Community Topic enables quick identification of flat topics and topic hierarchy, facilitating the on-demand exploration of sub- and super-topics. It also obtains good results on datasets in different languages.

Reconstructing brain functional networks through identifiability and deep learning.

We propose a novel approach for the reconstruction of functional networks representing brain dynamics based on the idea that the coparticipation of two brain regions in a common cognitive task should result in a drop in their identifiability, or in the uniqueness of their dynamics. This identifiability is estimated through the score obtained by deep learning models in supervised classification tasks and therefore requires no a priori assumptions about the nature of such coparticipation. The method is tested on EEG recordings obtained from Alzheimer's and Parkinson's disease patients, and matched healthy volunteers, for eyes-open and eyes-closed resting-state conditions, and the resulting functional networks are analysed through standard topological metrics. Both groups of patients are characterised by a reduction in the identifiability of the corresponding EEG signals, and by differences in the patterns that support such identifiability. Resulting functional networks are similar, but not identical to those reconstructed by using a correlation metric. Differences between control subjects and patients can be observed in network metrics like the clustering coefficient and the assortativity in different frequency bands. Differences are also observed between eyes open and closed conditions, especially for Parkinson's disease patients.

Simple-based Dynamic Decentralized Community Detection Algorithm in socially aware networks.

This paper introduces a novel, Simple-based Dynamic Decentralized Community Detection Algorithm (S-DCDA) for Socially Aware Networks. This algorithm aims to address the resource-intensive nature, instabilities and inaccuracies of traditional distributed community detection algorithms. The dynamics of decentralization is evident in the threefold nature of the algorithm: (i) each node of the community is the core of the entire network or community for a certain period of time dependent on their need, (ii) nodes are not centralized around themselves, requiring the consent of the other node to join a community, and (iii) Communities start from a single node to form an initial scale community, the number of nodes and the relationship among them are constantly changing. The algorithm requires low processor performance and memory capacity size of each node, to a certain extent, effectively improve the accuracy and stability of community detection and maintenance. Experimental results demonstrate that in comparison to classical and classical-based improved community detection algorithms, S-DCDA yields superior detection results.

Human protein-protein interaction networks: A topological comparison review.

Protein-Protein Interaction Networks aim to model the interactome, providing a powerful tool for understanding the complex relationships governing cellular processes. These networks have numerous applications, including functional enrichment, discovering cancer driver genes, identifying drug targets, and more. Various databases make protein-protein networks available for many species, including Homo sapiens. This work topologically compares four Homo sapiens networks using a coarse-to-fine approach, comparing global characteristics, sub-network topology, specific nodes centrality, and interaction significance. Results show that the four human protein networks share many common protein-encoding genes and some global measures, but significantly differ in the interactions and neighbourhood. Small sub-networks from cancer pathways performed better than the whole networks, indicating an improved topological consistency in functional pathways. The centrality analysis shows that the same genes play different roles in different networks. We discuss how studies and analyses that rely on protein-protein networks for humans should consider their similarities and distinctions.

Optimizing echo state networks for continuous gesture recognition in mobile devices: A comparative study.

Continuous gesture recognition can be used to enhance human-computer interaction. This can be accomplished by capturing human movement with the use of the Inertial Measurement Units in smartphones and using machine learning algorithms to predict the intended gestures. Echo State Networks (ESNs) consist of a fixed internal reservoir that is able to generate rich and diverse nonlinear dynamics in response to input signals that capture temporal dependencies within the signal. This makes ESNs well-suited for time series prediction tasks, such as continuous gesture recognition. However, their application has not been rigorously explored, with regard to gesture recognition. In this study, we sought to enhance the efficacy of ESN models in continuous gesture recognition by exploring diverse model structures, fine-tuning hyperparameters, and experimenting with various training approaches. We used three different training schemes that used the Leave-one-out Cross-validation (LOOCV) protocol to investigate the performance in real-world scenarios with different levels of data availability: Leaving out data from one user to use for testing (F1-score: 0.89), leaving out a fraction of data from all users to use in testing (F1-score: 0.96), and training and testing using LOOCV on a single user (F1-score: 0.99). The obtained results outperformed the Long Short-Term Memory (LSTM) performance from past research (F1-score: 0.87) while maintaining a low training time of approximately 13 seconds compared to 63 seconds for the LSTM model. Additionally, we further explored the performance of the ESN models through behaviour space analysis using memory capacity, Kernel Rank, and Generalization Rank. Our results demonstrate that ESNs can be optimized to achieve high performance on gesture recognition in mobile devices on multiple levels of data availability. These findings highlight the practical ability of ESNs to enhance human-computer interaction.

The Alcatraz-Strategy: a roadmap to break the connectivity barrier in malignant brain tumours.

In recent years, the discovery of functional and communicative cellular tumour networks has led to a new understanding of malignant primary brain tumours. In this review, the authors shed light on the diverse nature of cell-to-cell connections in brain tumours and propose an innovative treatment approach to address the detrimental connectivity of these networks. The proposed therapeutic outlook revolves around three main strategies: (a) supramarginal resection removing a substantial portion of the communicating tumour cell front far beyond the gadolinium-enhancing tumour mass, (b) morphological isolation at the single cell level disrupting structural cell-to-cell contacts facilitated by elongated cellular membrane protrusions known as tumour microtubes (TMs), and (c) functional isolation at the single cell level blocking TM-mediated intercellular cytosolic exchange and inhibiting neuronal excitatory input into the malignant network. We draw an analogy between the proposed therapeutic outlook and the Alcatraz Federal Penitentiary, where inmates faced an impassable sea barrier and experienced both spatial and functional isolation within individual cells. Based on current translational efforts and ongoing clinical trials, we propose the Alcatraz-Strategy as a promising framework to tackle the harmful effects of cellular brain tumour networks.

Botany and geogenomics: Constraining geological hypotheses in the neotropics with large-scale genetic data derived from plants.

Decades of empirical research have revealed how the geological history of our planet shaped plant evolution by establishing well-known patterns (e.g., how mountain uplift resulted in high rates of diversification and replicate radiations in montane plant taxa). This follows a traditional approach where botanical data are interpreted in light of geological events. In this synthesis, I instead describe how by integrating natural history, phylogenetics, and population genetics, botanical research can be applied alongside geology and paleontology to inform our understanding of past geological and climatic processes. This conceptual shift aligns with the goals of the emerging field of geogenomics. In the neotropics, plant geogenomics is a powerful tool for the reciprocal exploration of two long standing questions in biology and geology: how the dynamic landscape of the region came to be and how it shaped the evolution of the richest flora. Current challenges that are specific to analytical approaches for plant geogenomics are discussed. I describe the scale at which various geological questions can be addressed from biological data and what makes some groups of plants excellent model systems for geogenomics research. Although plant geogenomics is discussed with reference to the neotropics, the recommendations given here for approaches to plant geogenomics can and should be expanded to exploring long-standing questions on how the earth evolved with the use of plant DNA.

Experimental Demonstration of Reservoir Computing with Self-Assembled Percolating Networks of Nanoparticles.

The complex self-assembled network of neurons and synapses that comprizes the biological brain enables natural information processing with remarkable efficiency. Percolating Networks of Nanoparticles (PNNs) are complex self-assembled nanoscale systems that have been shown to possess many promising brain-like attributes and which are therefore appealing systems for neuromorphic computation. Here we perform experiments that show that PNNs can be utilized as physical reservoirs within a nanoelectronic reservoir computing framework and demonstrate successful computation for several benchmark tasks (chaotic time series prediction, non-linear transformation and memory capacity). For each task we compile relevant literature results and show that the performance of the PNNs compares favourably to that previously reported from nanoelectronic reservoirs. We then demonstrate experimentally that PNNs can be used for spoken digit recognition with state-of-the-art accuracy. Finally, we emulate a parallel reservoir architecture, which increases the dimensionality and richness of the reservoir outputs and results in further improvements in performance across all tasks. This article is protected by copyright. All rights reserved.

Clinical trial recommendations using Semantics-Based inductive inference and knowledge graph embeddings.

OBJECTIVE: Designing a new clinical trial entails many decisions, such as defining a cohort and setting the study objectives to name a few, and therefore can benefit from recommendations based on exhaustive mining of past clinical trial records. This study proposes an approach based on knowledge graph embeddings and semantics-driven inductive inference for generating such recommendations. METHOD: The proposed recommendation methodology is based on neural embeddings trained on first-of-its-kind knowledge graph constructed from clinical trials data. The methodology includes design of a knowledge graph for clinical trial data, evaluation of various knowledge graph embedding techniques for it, application of a novel inductive inference method using these embeddings, and generation of recommendations for clinical trial design. The study uses freely available data from clinicaltrials.gov and related sources. RESULTS: The proposed approach for recommendations obtained relevance scores ranging from 70% to 83%. These scores were determined by evaluating the text similarity of recommended elements to actual elements used in clinical trials that are in progress. Furthermore, the most pertinent recommendations were consistently located towards the top of the list, indicating the effectiveness of our method. CONCLUSION: Our study suggests that inductive inference using node semantics is a viable approach for generating recommendations using graphs neural embeddings, and that there is a potential for improvement in training graph embeddings using node semantics.

Single-Cell Nuclear Sequencing Atlas Revealed the Induction of Parkinson's Disease by RELN+ Neuron 3 and the Gene Regulatory Network of MSRA.

AIMS: To determine the cell types that promoted the progression of Parkinson's disease (PD) using the substantia nigra in the brain tissues derived from patients with PD and normal controls. BACKGROUND: PD is an incurable neurodegenerative disease that threatens the physical activity of the aging population, and the complex molecular mechanisms remain be comprehensively elucidated. OBJECTIVE: To describe potential disease-promoting cell types in PD and to provide a theoretical basis. METHODS: Single-cell nuclear sequencing data of nine PD samples and control samples from Gene Expression Omnibus (GEO) were included, and heterogeneous cell subpopulations in the substantia nigra were identified by annotation analysis. Potential pathogenic cell subpopulations of PD were determined based on the expression data of marker genes. Cell differentiation trajectories and communication networks were generated by Pseudotime trajectory analysis and cell communication analysis. Furthermore, single-- cell regulatory network inference and clustering (SCENIC) analysis was conducted to determine the regulatory network of transcription factor-target genes in PD. RESULTS: Among the nine cell subpopulations classified, RELN+neuron 3 showed reduced abundance and dopamine secretion capacity in PD and was therefore considered as a promoter of PD pathogenesis and progression. The regulatory network of MSRA action was involved in the developmental process of cells in the central nervous system, indicating that MSRA and its targets might serve as potential therapeutic targets for PD. RELN+neuron 3 had two directions of differentiation, specifically, branch 1 exhibited a high apoptotic profile and branch 2 exhibited a high cell death profile. In addition, the intensity of EPHA and EPHB signaling was attenuated between RELN+neuron 3 and other cell subpopulations. CONCLUSION: To conclude, this study identified a subpopulation of RELN+neuron 3 cells with markedly reduced abundance in the brain substantia nigra in PD. The MSRA-involved gene regulatory networks was considered as a novel therapeutic network for PD.

Virtual MOLLI Target: Generative Adversarial Networks Toward Improved Motion Correction in MRI Myocardial T1 Mapping.

BACKGROUND: The modified Look-Locker inversion recovery (MOLLI) sequence is commonly used for myocardial T1 mapping. However, it acquires images with different inversion times, which causes difficulty in motion correction for respiratory-induced misregistration to a given target image. HYPOTHESIS: Using a generative adversarial network (GAN) to produce virtual MOLLI images with consistent heart positions can reduce respiratory-induced misregistration of MOLLI datasets. STUDY TYPE: Retrospective. POPULATION: 1071 MOLLI datasets from 392 human participants. FIELD STRENGTH/SEQUENCE: Modified Look-Locker inversion recovery sequence at 3 T. ASSESSMENT: A GAN model with a single inversion time image as input was trained to generate virtual MOLLI target (VMT) images at different inversion times which were subsequently used in an image registration algorithm. Four VMT models were investigated and the best performing model compared with the standard vendor-provided motion correction (MOCO) technique. STATISTICAL TESTS: The effectiveness of the motion correction technique was assessed using the fitting quality index (FQI), mutual information (MI), and Dice coefficients of motion-corrected images, plus subjective quality evaluation of T1 maps by three independent readers using Likert score. Wilcoxon signed-rank test with Bonferroni correction for multiple comparison. Significance levels were defined as P < 0.01 for highly significant differences and P < 0.05 for significant differences. RESULTS: The best performing VMT model with iterative registration demonstrated significantly better performance (FQI 0.88 ± 0.03, MI 1.78 ± 0.20, Dice 0.84 ± 0.23, quality score 2.26 ± 0.95) compared to other approaches, including the vendor-provided MOCO method (FQI 0.86 ± 0.04, MI 1.69 ± 0.25, Dice 0.80 ± 0.27, quality score 2.16 ± 1.01). DATA CONCLUSION: Our GAN model generating VMT images improved motion correction, which may assist reliable T1 mapping in the presence of respiratory motion. Its robust performance, even with considerable respiratory-induced heart displacements, may be beneficial for patients with difficulties in breath-holding. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY: Stage 1.

Investigating the impact of primary care networks on continuity of care in English general practice: Analysis of interviews with patients and clinicians from a mixed methods study.

INTRODUCTION: In England, primary care networks (PCNs) offer opportunities to improve access to and sustainability of general practice through collaboration between groups of practices to provide care with a broader range of practitioner roles. However, there are concerns that these changes may undermine continuity of care. Our study investigates what the organisational shift to PCNs means for continuity of care. METHODS: The paper uses thematic analysis of qualitative data from interviews with general practitioners and other healthcare professionals (HCPs, n = 33) in 19 practices in five PCNs, and their patients (n = 35). Three patient cohorts within each participating practice were recruited, based on anticipated higher or lower needs for continuity of care: patients over 65 years with polypharmacy, patients with anxiety or depression and 'working age' adults aged between 18 and 45 years. FINDINGS: Patients and clinicians perceived changes to continuity in PCNs in our study. Larger-scale care provision in PCNs required better care coordination and information-sharing processes, aimed at improving care for 'vulnerable' patients in target groups. However, new working arrangements and ways of delivering care in PCNs undermine HCPs' ability to maintain continuity through ongoing relationships with patients. Patients experience this in terms of reduced availability of their preferred clinician, inefficiencies in care and unfamiliarity of new staff, roles and processes. CONCLUSIONS: New practitioners need to be effectively integrated to support effective team-based care. However, for patients, especially those not deemed 'vulnerable', this may not be sufficient to counter the loss of relationship with their practice. Therefore, caution is required in relation to designating patients as in need of, or not in need of continuity. Rather, continuity for all patients could be maintained through a dynamic understanding of the need for it as fluctuating and situational and by supporting clinicians to provide follow-up care. PATIENT AND PUBLIC INVOLVEMENT (PPI): A PPI group was recruited and consulted during the study for feedback on the study design, recruitment materials and interpretation of findings.

Optimizing hydropower scheduling through accurate power load prediction: A practical case study.

Hydropower stations that are part of the grid system frequently encounter challenges related to the uneven distribution of power generation and associated benefits, primarily stemming from delays in obtaining timely load data. This research addresses this issue by developing a scheduling model that combines power load prediction and dual-objective optimization. The practical application of this model is demonstrated in a real-case scenario, focusing on the Shatuo Hydropower Station in China. In contrast to current models, the suggested model can achieve optimal dispatch for grid-connected hydropower stations even when power load data is unavailable. Initially, the model assesses various prediction models for estimating power load and subsequently incorporates the predictions into the GA-NSGA-II algorithm, specifically an enhanced elite non-dominated sorting genetic algorithm. This integration is performed while considering the proposed objective functions to optimize the discharge flow of the hydropower station. The outcomes reveal that the CNN-GRU model, denoting Convolutional Neural Network-Gated Recursive Unit, exhibits the highest prediction accuracy, achieving R-squared and RMSE (i.e., Root Mean Square Error) values of 0.991 and 0.026, respectively. The variance between scheduling based on predicted load values and actual load values is minimal, staying within 5 (m3/s), showcasing practical effectiveness. The optimized scheduling outcomes in the real case study yield dual advantages, meeting both the demands of ship navigation and hydropower generation, thus achieving a harmonious balance between the two requirements. This approach addresses the real-world challenges associated with delayed load data collection and insufficient scheduling, offering an efficient solution for managing hydropower station scheduling to meet both power generation and navigation needs.

Biomarker potential of competing endogenous RNA networks in Polycystic Ovary Syndrome (PCOS).

Polycystic ovary syndrome (PCOS) is the most common condition affecting women of reproductive age globally. PCOS continues to be the largest contributing factor to female infertility despite significant progress in our knowledge of the molecular underpinnings and treatment of the condition. The fact that PCOS is a very diverse condition makes it one of the key reasons why we haven't been able to overcome it. Non-coding RNAs (ncRNAs) are implicated in the development of PCOS, according to growing evidence. However, it is unclear how the complex regulatory relationships between the many ncRNA types contribute to the growth of this malignancy. Competing endogenous RNA (ceRNA), a recently identified mechanism in the RNA world, suggests regulatory interactions between various RNAs, including long non-coding RNAs (lncRNAs), microRNAs (miRNAs), transcribed pseudogenes, and circular RNAs (circRNAs). Recent studies on PCOS have shown that dysregulation of multiple ceRNA networks (ceRNETs) between these ncRNAs plays crucial roles in developing the defining characteristics of PCOS development. And it is believed that such a finding may open a new door for a deeper comprehension of PCOS's unexplored facets. In addition, it may be able to provide fresh biomarkers and effective therapy targets for PCOS. This review will go over the body of information that exists about the primary roles of ceRNETs before highlighting the developing involvement of several newly found ceRNETs in a number of PCOS characteristics.

Dynamic nesting of Anaplasma marginale in the microbial communities of Rhipicephalus microplus.

Interactions within the tick microbiome involving symbionts, commensals, and tick-borne pathogens (TBPs) play a pivotal role in disease ecology. This study explored temporal changes in the microbiome of Rhipicephalus microplus, an important cattle tick vector, focusing on its interaction with Anaplasma marginale. To overcome limitations inherent in sampling methods relying on questing ticks, which may not consistently reflect pathogen presence due to variations in exposure to infected hosts in nature, our study focused on ticks fed on chronically infected cattle. This approach ensures continuous pathogen exposure, providing a more comprehensive understanding of the nesting patterns of A. marginale in the R. microplus microbiome. Using next-generation sequencing, microbiome dynamics were characterized over 2 years, revealing significant shifts in diversity, composition, and abundance. Anaplasma marginale exhibited varying associations, with its increased abundance correlating with reduced microbial diversity. Co-occurrence networks demonstrated Anaplasma's evolving role, transitioning from diverse connections to keystone taxa status. An integrative approach involving in silico node removal unveils the impact of Anaplasma on network stability, highlighting its role in conferring robustness to the microbial community. This study provides insights into the intricate interplay between the tick microbiome and A. marginale, shedding light on potential avenues for controlling bovine anaplasmosis through microbiome manipulation.

Implementing machine learning techniques for continuous emotion prediction from uniformly segmented voice recordings.

Introduction: Emotional recognition from audio recordings is a rapidly advancing field, with significant implications for artificial intelligence and human-computer interaction. This study introduces a novel method for detecting emotions from short, 1.5 s audio samples, aiming to improve accuracy and efficiency in emotion recognition technologies. Methods: We utilized 1,510 unique audio samples from two databases in German and English to train our models. We extracted various features for emotion prediction, employing Deep Neural Networks (DNN) for general feature analysis, Convolutional Neural Networks (CNN) for spectrogram analysis, and a hybrid model combining both approaches (C-DNN). The study addressed challenges associated with dataset heterogeneity, language differences, and the complexities of audio sample trimming. Results: Our models demonstrated accuracy significantly surpassing random guessing, aligning closely with human evaluative benchmarks. This indicates the effectiveness of our approach in recognizing emotional states from brief audio clips. Discussion: Despite the challenges of integrating diverse datasets and managing short audio samples, our findings suggest considerable potential for this methodology in real-time emotion detection from continuous speech. This could contribute to improving the emotional intelligence of AI and its applications in various areas.

Gender Differences in Social Support Vary by Environment Among Older People With HIV in South Africa and Uganda.

Background and Objectives: Sub-Saharan Africa is home to 3.7 million older adults living with HIV, who experience high rates of comorbid conditions. Formal services other than HIV clinical care are largely unavailable. Overall, women are the mainstay of informal social support networks, and older women with HIV may face burdens due to family caregiving expectations. Thus, it is important to understand the extent of informal support provided to older adults living with HIV, and how this is affected by gender. Research Design and Methods: We examined differences in social networks, needs, social support and caregiving, and perceptions of support adequacy among women and men aged 50 and older living with HIV in Uganda (n = 101) and South Africa (n = 108), mostly rural and suburban populations, respectively. We used multiple regression to determine whether there was an association between gender and the amount of social support received and whether that varied by research site. Results: Men were more likely than women to receive support from a partner. Women were more likely to live with offspring, both providing and receiving care. In South Africa but not Uganda, women received more help from family than men did. There was no gender difference in getting help from friends, but it was more common in Uganda. Living alone was strongly associated with less family help and more help from friends. Discussion and Implications: Older women with HIV in sub-Saharan Africa tend to be more heavily involved in social support exchanges-both providing and receiving care-than their male peers, but place matters. Interdependence is high in rural Uganda, where formal services are scarce and needs exceed resources. Given the projected growth in this population, stronger formal supports are needed for communities and older people with HIV, especially those who live alone.

Toughening Hydrogels with Fibrillar Connected Double Networks.

Biological tissues, such as tendons or cartilage, possess high strength and toughness with very low plastic deformations. In contrast, current strategies to prepare tough hydrogels commonly utilize energy dissipation mechanisms based on physical bonds that lead to irreversible large plastic deformations, thus limiting their load-bearing applications. This article reports a strategy to toughen hydrogels using fibrillar connected double networks (fc-DN), which consist of two distinct but chemically interconnected polymer networks, that is, a polyacrylamide network and an acrylated agarose fibril network. The fc-DN design allows efficient stress transfer between the two networks and high fibril alignment during deformation, both contributing to high strength and toughness, while the chemical crosslinking ensures low plastic deformations after undergoing high strains. The mechanical properties of the fc-DN network can be readily tuned to reach an ultimate tensile strength of 8 MPa and a toughness of above 55 MJ m-3, which is 3 and 3.5 times more than that of fibrillar double network hydrogels without chemical connections, respectively. The application potential of the fc-DN hydrogel is demonstrated as load-bearing damping material for a jointed robotic lander. The fc-DN design provides a new toughening mechanism for hydrogels that can be used for soft robotics or bioelectronic applications.

The Social Connectedness of Older Adults during the COVID-19 Pandemic: Evidence by Survey Mode and Respondent Dementia.

OBJECTIVES: The COVID-19 pandemic has affected many aspects of social life, especially among older adults who may face cognitive impairments. Concerning this combination of circumstances, the study evaluates the degree to which data collection on social connectedness among older adults might be affected by the social complexities of the COVID-19 pandemic. METHOD: We use data from the National Social Life, Health and Aging Project (NSHAP), a nationally representative study of community-dwelling older adults in the U.S., which conducted a special multi-mode COVID study between September 2020 and January 2021, in part to examine social impacts of the COVID-19 pandemic and to assess how alternative survey modes performed during the pandemic. Our final sample includes 2,251 older adults, ages 55 and older. RESULTS: Older adults' social connectedness was adversely affected by the pandemic. People reported a tendency to move toward electronic communication and away from in-person contact. Concomitantly, there is some evidence of survey mode effects that are related to electronic communication. Those who elected to participate on the phone, or the internet disproportionately reported using those means of communication with their social network members. Notably, this pattern was stronger among those who did not suffer from dementia, suggesting cognition effects on survey completion. DISCUSSION: Researchers should remain cognizant of how data on social connections were collected during the COVID-19 pandemic. These findings may indicate the role dementia plays in preventing people from adapting to new social networking realities with alternative means of communication during the pandemic.

New results on bifurcation for fractional-order octonion-valued neural networks involving delays.

This work chiefly explores fractional-order octonion-valued neural networks involving delays. We decompose the considered fractional-order delayed octonion-valued neural networks into equivalent real-valued systems via Cayley-Dickson construction. By virtue of Lipschitz condition, we prove that the solution of the considered fractional-order delayed octonion-valued neural networks exists and is unique. By constructing a fairish function, we confirm that the solution of the involved fractional-order delayed octonion-valued neural networks is bounded. Applying the stability theory and basic bifurcation knowledge of fractional order differential equations, we set up a sufficient condition remaining the stability behaviour and the appearance of Hopf bifurcation for the addressed fractional-order delayed octonion-valued neural networks. To illustrate the justifiability of the derived theoretical results clearly, we give the related simulation results to support these facts. Simultaneously, the bifurcation plots are also displayed. The established theoretical results in this work have important guiding significance in devising and improving neural networks.