Mild Cognitive Impairment Detection with Machine Learning and Topological Data Analysis Applied to EEG Time-series in Facial Emotion Oddball Paradigm.

We report a novel approach to dementia neurobiomarker development from EEG time series using topological data analysis (TDA) methodology and machine learning (ML) tools in the 'AI for social good' application domain, with possible following application to home-based point of care diagnostics and cognitive intervention monitoring. We propose a new approach to a digital dementia neurobiomarker for early-onset mild cognitive impairment (MCI) prognosis. We report the best median accuracies in a range of upper 85% linear discriminant analysis (LDA), as well above 90% for linear SVM and deep fully connected neural network classifier models in leave-one-out-subject cross-validation, which presents very encouraging results in a binary healthy cognitive aging versus MCI stages using TDA features applied to brainwave time series patterns captured from a four-channel EEG wearable.Clinical relevance- The reported study offers an objective dementia early onset neurobiomarker prospect to replace traditional subjective paper and pencil tests with an application of EEG-wearable-based and topological data analysis machine learning tools in a possibly successive home-based point-of-care environment.

Metagenomic analysis of ecological niche overlap and community collapse in microbiome dynamics.

Species utilizing the same resources often fail to coexist for extended periods of time. Such competitive exclusion mechanisms potentially underly microbiome dynamics, causing breakdowns of communities composed of species with similar genetic backgrounds of resource utilization. Although genes responsible for competitive exclusion among a small number of species have been investigated in pioneering studies, it remains a major challenge to integrate genomics and ecology for understanding stable coexistence in species-rich communities. Here, we examine whether community-scale analyses of functional gene redundancy can provide a useful platform for interpreting and predicting collapse of bacterial communities. Through 110-day time-series of experimental microbiome dynamics, we analyzed the metagenome-assembled genomes of co-occurring bacterial species. We then inferred ecological niche space based on the multivariate analysis of the genome compositions. The analysis allowed us to evaluate potential shifts in the level of niche overlap between species through time. We hypothesized that community-scale pressure of competitive exclusion could be evaluated by quantifying overlap of genetically determined resource-use profiles (metabolic pathway profiles) among coexisting species. We found that the degree of community compositional changes observed in the experimental microbiome was correlated with the magnitude of gene-repertoire overlaps among bacterial species, although the causation between the two variables deserves future extensive research. The metagenome-based analysis of genetic potential for competitive exclusion will help us forecast major events in microbiome dynamics such as sudden community collapse (i.e., dysbiosis).

Machine learning approach for early onset dementia neurobiomarker using EEG network topology features.

Introduction: Modern neurotechnology research employing state-of-the-art machine learning algorithms within the so-called "AI for social good" domain contributes to improving the well-being of individuals with a disability. Using digital health technologies, home-based self-diagnostics, or cognitive decline managing approaches with neuro-biomarker feedback may be helpful for older adults to remain independent and improve their wellbeing. We report research results on early-onset dementia neuro-biomarkers to scrutinize cognitive-behavioral intervention management and digital non-pharmacological therapies. Methods: We present an empirical task in the EEG-based passive brain-computer interface application framework to assess working memory decline for forecasting a mild cognitive impairment. The EEG responses are analyzed in a framework of a network neuroscience technique applied to EEG time series for evaluation and to confirm the initial hypothesis of possible ML application modeling mild cognitive impairment prediction. Results: We report findings from a pilot study group in Poland for a cognitive decline prediction. We utilize two emotional working memory tasks by analyzing EEG responses to facial emotions reproduced in short videos. A reminiscent interior image oddball task is also employed to validate the proposed methodology further. Discussion: The proposed three experimental tasks in the current pilot study showcase the critical utilization of artificial intelligence for early-onset dementia prognosis in older adults.

Facilitative interaction networks in experimental microbial community dynamics.

Facilitative interactions between microbial species are ubiquitous in various types of ecosystems on the Earth. Therefore, inferring how entangled webs of interspecific interactions shift through time in microbial ecosystems is an essential step for understanding ecological processes driving microbiome dynamics. By compiling shotgun metagenomic sequencing data of an experimental microbial community, we examined how the architectural features of facilitative interaction networks could change through time. A metabolic modeling approach for estimating dependence between microbial genomes (species) allowed us to infer the network structure of potential facilitative interactions at 13 time points through the 110-day monitoring of experimental microbiomes. We then found that positive feedback loops, which were theoretically predicted to promote cascade breakdown of ecological communities, existed within the inferred networks of metabolic interactions prior to the drastic community-compositional shift observed in the microbiome time-series. We further applied "directed-graph" analyses to pinpoint potential keystone species located at the "upper stream" positions of such feedback loops. These analyses on facilitative interactions will help us understand key mechanisms causing catastrophic shifts in microbial community structure.

Alternative stable states, nonlinear behavior, and predictability of microbiome dynamics.

BACKGROUND: Microbiome dynamics are both crucial indicators and potential drivers of human health, agricultural output, and industrial bio-applications. However, predicting microbiome dynamics is notoriously difficult because communities often show abrupt structural changes, such as "dysbiosis" in human microbiomes. METHODS: We integrated theoretical frameworks and empirical analyses with the aim of anticipating drastic shifts of microbial communities. We monitored 48 experimental microbiomes for 110 days and observed that various community-level events, including collapse and gradual compositional changes, occurred according to a defined set of environmental conditions. We analyzed the time-series data based on statistical physics and non-linear mechanics to describe the characteristics of the microbiome dynamics and to examine the predictability of major shifts in microbial community structure. RESULTS: We confirmed that the abrupt community changes observed through the time-series could be described as shifts between "alternative stable states" or dynamics around complex attractors. Furthermore, collapses of microbiome structure were successfully anticipated by means of the diagnostic threshold defined with the "energy landscape" analysis of statistical physics or that of a stability index of nonlinear mechanics. CONCLUSIONS: The results indicate that abrupt microbiome events in complex microbial communities can be forecasted by extending classic ecological concepts to the scale of species-rich microbial systems. Video Abstract.

Word-producing brain: Contribution of the left anterior middle temporal gyrus to word production patterns in spoken language.

Vocabulary is based on semantic knowledge. The anterior temporal lobe (ATL) has been considered an essential region for processing semantic knowledge; nonetheless, the association between word production patterns and the structural and functional characteristics of the ATL remains unclear. To examine this, we analyzed over one million words from group conversations among community-dwelling older adults and their multimodal magnetic resonance imaging data. A quantitative index for the word production patterns, namely the exponent ß of Heaps' law, positively correlated with the left anterior middle temporal gyrus volume. Moreover, ß negatively correlated with its resting-state functional connectivity with the precuneus. There was no significant correlation with the diffusion tensor imaging metrics in any fiber. These findings suggest that the vocabulary richness in spoken language depends on the brain status characterized by the semantic knowledge-related brain structure and its activation dissimilarity with the precuneus, a core region of the default mode network.

Dementia Digital Neuro-biomarker Study from Theta-band EEG Fluctuation Analysis in Facial and Emotional Identification Short-term Memory Oddball Paradigm.

An efficient machine learning (ML) implementation in the so-called 'AI for social good' domain shall contribute to dementia digital neuro-biomarker development for early-onset prognosis of a possible cognitive decline. We report encouraging initial developments of wearable EEG-derived theta-band fluctuations examination and a successive classification embracing a time-series complexity examination with a multifractal detrended fluctuation analysis (MFDFA) in the face or emotion video-clip identification short-term oddball memory tasks. We also report findings from a thirty-five elderly volunteer pilot study that EEG responses to instructed to ignore (inhibited) oddball paradigm stimulation results in more informative MFDFA features, leading to better machine learning classification results. The reported pilot project showcases vital social assistance of artificial intelligence (AI) application for an early-onset dementia prognosis. Clinical Relevance- This introduces a candidate for an objective digital neuro-biomarker from theta-band EEG recorded by a wearable for a plausible replacement of biased 'paper & pencil' tests for a mild cognitive impairment (MCI) evaluation.

Chemical-Mediated Microbial Interactions Can Reduce the Effectiveness of Time-Series-Based Inference of Ecological Interaction Networks.

Network-based assessments are important for disentangling complex microbial and microbial-host interactions and can provide the basis for microbial engineering. There is a growing recognition that chemical-mediated interactions are important for the coexistence of microbial species. However, so far, the methods used to infer microbial interactions have been validated with models assuming direct species-species interactions, such as generalized Lotka-Volterra models. Therefore, it is unclear how effective existing approaches are in detecting chemical-mediated interactions. In this paper, we used time series of simulated microbial dynamics to benchmark five major/state-of-the-art methods. We found that only two methods (CCM and LIMITS) were capable of detecting interactions. While LIMITS performed better than CCM, it was less robust to the presence of chemical-mediated interactions, and the presence of trophic competition was essential for the interactions to be detectable. We show that the existence of chemical-mediated interactions among microbial species poses a new challenge to overcome for the development of a network-based understanding of microbiomes and their interactions with hosts and the environment.

Older Adult Mild Cognitive Impairment Prediction from Multiscale Entropy EEG Patterns in Reminiscent Interior Image Working Memory Paradigm.

We discuss the practical employment of a machine learning (ML) technique within AI for a social good application. We present an application for elderly adult dementia onset prognostication. First, the paper explains our encouraging preliminary study results of EEG responses analysis using a signal complexity measure of multiscale entropy (MSE) in reminiscent interior working memory evaluation tasks. Then, we compare shallow and deep learning machine learning models for a digital biomarker of dementia onset detection. The evaluated machine-learning models succeed in the most reliable median accuracies above 80% using random forest and fully connected neural network classifiers in automatic discrimination of normal cognition versus a mild cognitive impairment (MCI) task. The classifier input features consist of MSE patterns only derived from four dry EEG electrodes. Fifteen elderly subjects voluntarily participate in the reported study focusing on EEG-based objective dementia biomarker advancement. The results showcase the essential social advantages of artificial intelligence (AI) application for the dementia prognosis and advance ML for the subsequent use for simple objective EEG-based examination.Clinical relevance- This manuscript introduces an objective biomarker from EEG recorded by a wearable for a plausible replacement of a mild cognitive impairment (MCI) evaluation using usual biased paper and pencil examinations.

Neurotechnology and AI Approach for Early Dementia Onset Biomarker from EEG in Emotional Stimulus Evaluation Task.

We present an efficient utilization of a machine learning (ML) method concentrating on the 'AI for social good' application. We develop a digital dementia biomarker for early-onset dementia forecast. The paper demonstrates encouraging preliminary results of EEG-wearable-based signal analysis and a subsequent classification adopting a signal complexity test of a multifractal detrended fluctuation analysis (MFDFA) in emotional faces working memory training and evaluation tasks. For the digital biomarker of dementia onset detection, we examine shallow- and deep-learning machine learning models. We report the best median accuracies in a range of 90% for random forest and fully connected neural network classifier models in both emotional faces learning and evaluation experimental tasks. In addition, the classifiers are trained in a ten-fold cross-validation regime to discriminate normal versus mild cognitive impairment (MCI) cognition stages using MFDFA patterns from four-channel EEG recordings. Thirty-five volunteer elderly subjects participate in the current study concentrating on simple wearable EEG-based objective dementia biomarker progression. The reported outcomes showcase an essential social benefit of artificial intelligence (AI) employment for early dementia prediction. Furthermore, we improve ML employment for the succeeding application in an uncomplicated and applied EEG-wearable examination.

Cognitive Intervention Through Photo-Integrated Conversation Moderated by Robots (PICMOR) Program: A Randomized Controlled Trial.

Social interaction might prevent or delay dementia, but little is known about the specific effects of various social activity interventions on cognition. This study conducted a single-site randomized controlled trial (RCT) of Photo-Integrated Conversation Moderated by Robots (PICMOR), a group conversation intervention program for resilience against cognitive decline and dementia. In the RCT, PICMOR was compared to an unstructured group conversation condition. Sixty-five community-living older adults participated in this study. The intervention was provided once a week for 12 weeks. Primary outcome measures were the cognitive functions; process outcome measures included the linguistic characteristics of speech to estimate interaction quality. Baseline and post-intervention data were collected. PICMOR contains two key features: 1) photos taken by the participants are displayed and discussed sequentially; and 2) a robotic moderator manages turn-taking to make sure that participants are allocated the same amount of time. Among the primary outcome measures, one of the subcategories of cognitive functions, verbal fluency significantly improved in the intervention group. Among the process outcome measures, a part of the subcategories of linguistic characteristics of speech, the amount of speech and richness of words, proportion of providing topics, questions, and answers in total utterances were larger for the intervention group. This study demonstrated for the first time the positive effects of a robotic social activity intervention on cognitive function in healthy older adults via RCT. The group conversation generated by PICMOR may improve participants' verbal fluency since participants have more opportunity to provide their own topics, asking and answering questions which results in exploring larger vocabularies. PICMOR is available and accessible to community-living older adults. Clinical Trial Registration: UMIN Clinical Trials Registry, identifier UMIN000036667.

Individual Ants Do Not Show Activity-Rest Rhythms in Nest Conditions.

Circadian rhythms, which respond to the day-night cycle on the earth, arise from the endogenous timekeeping system within organisms, called the "biological clock." For accurate circadian rhythms, daily fluctuations in light and temperature are considered one of the important time cues. In social insects, both abiotic and biotic factors (i.e., social interactions) play a significant role in activity-rest rhythm regulation. However, it is challenging to monitor individual activity-rest rhythms in a colony because of the large group size and small body size. Therefore, it is unclear whether individuals in a colony exhibit activity-rest rhythms and how social interactions regulate their activity-rest rhythms in the colony. This study developed an image-based tracking system using 2D barcodes for Diacamma cf. indicum from Japan (a monomorphic ant) and measured the locomotor activities of all colony members under laboratory colony conditions. We also investigated the effect of broods on activity-rest rhythms by removing all broods under colony conditions. Activity-rest rhythms appeared only in isolated ants, not under colony conditions. In addition, workers showed arrhythmic activities after brood removal. These results suggested that a mixture of social interactions, and not light and temperature, induces the loss of activity-rest rhythms. These results contribute to the knowledge of a diverse pattern of circadian activity rhythms in social insects.

Scaling laws in natural conversations among elderly people.

Language is a result of brain function; thus, impairment in cognitive function can result in language disorders. Understanding the aging of brain functions in terms of language processing is crucial for modern aging societies. Previous studies have shown that language characteristics, such as verbal fluency, are associated with cognitive functions. However, the scaling laws in language in elderly people remain poorly understood. In the current study, we recorded large-scale data of one million words from group conversations among healthy elderly people and analyzed the relationship between spoken language and cognitive functions in terms of scaling laws, namely, Zipf's law and Heaps' law. We found that word patterns followed these scaling laws irrespective of cognitive function, and that the variations in Heaps' exponents were associated with cognitive function. Moreover, variations in Heaps' exponents were associated with the ratio of new words taken from the other participants' speech. These results indicate that the exponents of scaling laws in language are related to cognitive processes.

Bipartite network analysis of ant-task associations reveals task groups and absence of colonial daily activity.

Social insects are one of the best examples of complex self-organized systems exhibiting task allocation. How task allocation is achieved is the most fascinating question in behavioural ecology and complex systems science. However, it is difficult to comprehensively characterize task allocation patterns due to behavioural complexity, such as the individual variation, context dependency and chronological variation. Thus, it is imperative to quantify individual behaviours and integrate them into colony levels. Here, we applied bipartite network analyses to characterize individual-behaviour relationships. We recorded the behaviours of all individuals with verified age in ant colonies and analysed the individual-behaviour relationship at the individual, module and network levels. Bipartite network analysis successfully detected the module structures, illustrating that certain individuals performed a subset of behaviours (i.e. task groups). We confirmed age polyethism by comparing age between modules. Additionally, to test the daily rhythm of the executed tasks, the data were partitioned between daytime and nighttime, and a bipartite network was re-constructed. This analysis supported that there was no daily rhythm in the tasks performed. These findings suggested that bipartite network analyses could untangle complex task allocation patterns and provide insights into understanding the division of labour.

Amplitude of circadian rhythms becomes weaken in the north, but there is no cline in the period of rhythm in a beetle.

Many species show rhythmicity in activity, from the timing of flowering in plants to that of foraging behavior in animals. The free-running periods and amplitude (sometimes called strength or power) of circadian rhythms are often used as indicators of biological clocks. Many reports have shown that these traits are highly geographically variable, and interestingly, they often show latitudinal or longitudinal clines. In many cases, the higher the latitude is, the longer the free-running circadian period (i.e., period of rhythm) in insects and plants. However, reports of positive correlations between latitude or longitude and circadian rhythm traits, including free-running periods, the power of the rhythm and locomotor activity, are limited to certain taxonomic groups. Therefore, we collected a cosmopolitan stored-product pest species, the red flour beetle Tribolium castaneum, in various parts of Japan and examined its rhythm traits, including the power and period of the rhythm, which were calculated from locomotor activity. The analysis revealed that the power was significantly lower for beetles collected in northern areas than southern areas in Japan. However, it is worth noting that the period of circadian rhythm did not show any clines; specifically, it did not vary among the sampling sites, despite the very large sample size (n = 1585). We discuss why these cline trends were observed in T. castaneum.

Classifying Mild Cognitive Impairment from Behavioral Responses in Emotional Arousal and Valence Evaluation Task - AI Approach for Early Dementia Biomarker in Aging Societies.

The presented paper discusses a practical application of machine learning (ML) in the so-called 'AI for social good' domain and in particular concerning the problem of a potential elderly adult dementia onset prediction. An increase in dementia cases is producing a significant medical and economic weight in many countries. Approximately 47 million older adults live with a dementia spectrum of neurocognitive disorders, according to an up-to-date statement of the World Health Organization (WHO), and this amount will triple within the next thirty years. This growing problem calls for possible application of AI-based technologies to support early diagnostics for cognitive interventions and a subsequent mental wellbeing monitoring as well as maintenance with so-called 'digital-pharma' or 'beyond a pill' therapeutical strategies. The paper explains our attempt and encouraging preliminary study results of behavioral responses analysis in a facial emotion implicit-short-term-memory learning and evaluation experiment. We present results of various shallow and deep learning machine learning models for digital biomarkers of dementia progress detection and monitoring. The discussed machine-learning models result in median accuracies right below a 90% benchmark using classical shallow and deep learning approaches for automatic discrimination of normal cognition versus a mild cognitive impairment (MCI). The classifier input features consist of an older adult emotional valence and arousal recognition responses, together with reaction times, as well as with self-reported university-level degree education and age, as obtained from a group of 35 older adults participating voluntarily in the reported dementia biomarker development project. The presented results showcase the inherent social benefits of artificial intelligence (AI) utilization for the elderly and establish a step forward to advance machine learning (ML) approaches for the subsequent employment of simple behavioral examination for MCI and dementia onset diagnostics.Clinical relevance- This manuscript establishes a behavioral and cognitive biomarker candidate potentially substituting a Montreal Cognitive Assessment (MoCA) evaluation without a paper and pencil test.

Functional advantages of Lévy walks emerging near a critical point.

A special class of random walks, so-called Lévy walks, has been observed in a variety of organisms ranging from cells, insects, fishes, and birds to mammals, including humans. Although their prevalence is considered to be a consequence of natural selection for higher search efficiency, some findings suggest that Lévy walks might also be epiphenomena that arise from interactions with the environment. Therefore, why they are common in biological movements remains an open question. Based on some evidence that Lévy walks are spontaneously generated in the brain and the fact that power-law distributions in Lévy walks can emerge at a critical point, we hypothesized that the advantages of Lévy walks might be enhanced by criticality. However, the functional advantages of Lévy walks are poorly understood. Here, we modeled nonlinear systems for the generation of locomotion and showed that Lévy walks emerging near a critical point had optimal dynamic ranges for coding information. This discovery suggested that Lévy walks could change movement trajectories based on the magnitude of environmental stimuli. We then showed that the high flexibility of Lévy walks enabled switching exploitation/exploration based on the nature of external cues. Finally, we analyzed the movement trajectories of freely moving Drosophila larvae and showed empirically that the Lévy walks may emerge near a critical point and have large dynamic range and high flexibility. Our results suggest that the commonly observed Lévy walks emerge near a critical point and could be explained on the basis of these functional advantages.

Scoring Species for Synthetic Community Design: Network Analyses of Functional Core Microbiomes.

Constructing biological communities is a major challenge in both basic and applied sciences. Although model synthetic communities with a few species have been constructed, designing systems consisting of tens or hundreds of species remains one of the most difficult goals in ecology and microbiology. By utilizing high-throughput sequencing data of interspecific association networks, we here propose a framework for exploring "functional core" species that have great impacts on whole community processes and functions. The framework allows us to score each species within a large community based on three criteria: namely, topological positions, functional portfolios, and functional balance within a target network. The criteria are measures of each species' roles in maximizing functional benefits at the community or ecosystem level. When species with potentially large contributions to ecosystem-level functions are screened, the framework also helps us design "functional core microbiomes" by focusing on properties of species groups (modules) within a network. When embedded into agroecosystems or human gut, such functional core microbiomes are expected to organize whole microbiome processes and functions. An application to a plant-associated microbiome dataset actually highlighted potential functional core microbes that were known to control rhizosphere microbiomes by suppressing pathogens. Meanwhile, an example of application in mouse gut microbiomes called attention to poorly investigated bacterial species, whose potential roles within gut microbiomes deserve future experimental studies. The framework for gaining "bird's-eye" views of functional cores within networks is applicable not only to agricultural and medical data but also to datasets produced in food processing, brewing, waste water purification, and biofuel production.

Anomalous diffusion on the servosphere: A potential tool for detecting inherent organismal movement patterns.

Tracking animal movements such as walking is an essential task for understanding how and why animals move in an environment and respond to external stimuli. Different methods that implemented image analysis and a data logger such as GPS have been used in laboratory experiments and in field studies, respectively. Recently, animal movement patterns without stimuli have attracted an increasing attention in search for common innate characteristics underlying all of their movements. However, it is difficult to track the movements in a vast and homogeneous environment without stimuli because of space constraints in laboratories or environmental heterogeneity in the field, hindering our understanding of inherent movement patterns. Here, we applied an omnidirectional treadmill mechanism, or a servosphere, as a tool for tracking two-dimensional movements of small animals that can provide both a homogenous environment and a virtual infinite space for walking. To validate the use of our tracking system for assessment of the free-walking behavior, we compared walking patterns of individual pillbugs (Armadillidium vulgare) on the servosphere with that in two types of experimental flat arenas. Our results revealed that the walking patterns on the servosphere showed similar diffusive characteristics to those observed in the large arena simulating an open space, and we demonstrated that our mechanism provides more robust measurements of diffusive properties compared to a small arena with enclosure. Moreover, we showed that anomalous diffusion properties, including Lévy walk, can be detected from the free-walking behavior on our tracking system. Thus, our novel tracking system is useful to measure inherent movement patterns, which will contribute to the studies of movement ecology, ethology, and behavioral sciences.

Optimizing mating encounters by sexually dimorphic movements.

All organisms with sexual reproduction undergo a process of mating, which essentially involves the encounter of two individuals belonging to different sexes. During mate search, both sexes should mutually optimize their encounters, thus raising a question of how they achieve this. Here, we show that a population with sexually dimorphic movement patterns achieves the highest individual mating success under a limited lifespan. Extensive simulations found and analytical approximations corroborated the existence of conditions under which sexual dimorphism in the movement patterns (i.e. how diffusively they move) is advantageous over sexual monomorphism. Mutual searchers with limited lifespans need to balance the speed and accuracy of finding their mates, and dimorphic movements can solve this trade-off. We further demonstrate that the sexual dimorphism can evolve from an initial sexually monomorphic population. Our results emphasize the importance of considering mutual optimization in problems of random search.