Data Points Research Articles

A Novel Video-Based Methodology for Automated Classification of Dystonia and Choreoathetosis in Dyskinetic Cerebral Palsy During a Lower Extremity Task.

Movement disorders in children and adolescents with dyskinetic cerebral palsy (CP) are commonly assessed from video recordings, however scoring is time-consuming and expert knowledge is required for an appropriate assessment. To explore a machine learning approach for automated classification of amplitude and duration of distal leg dystonia and choreoathetosis within short video sequences. Available videos of a heel-toe tapping task were preprocessed to optimize key point extraction using markerless motion analysis. Postprocessed key point data were passed to a time series classification ensemble algorithm to classify dystonia and choreoathetosis duration and amplitude classes (scores 0, 1, 2, 3, and 4), respectively. As ground truth clinical scoring of dystonia and choreoathetosis by the Dyskinesia Impairment Scale was used. Multiclass performance metrics as well as metrics for summarized scores: absence (score 0) and presence (score 1-4) were determined. Thirty-three participants were included: 29 with dyskinetic CP and 4 typically developing, age 14 years:6 months ± 5 years:15 months. The multiclass accuracy results for dystonia were 77% for duration and 68% for amplitude; for choreoathetosis 30% for duration and 38% for amplitude. The metrics for score 0 versus score 1 to 4 revealed an accuracy of 81% for dystonia duration, 77% for dystonia amplitude, 53% for choreoathetosis duration and amplitude. This methodology study yielded encouraging results in distinguishing between presence and absence of dystonia, but not for choreoathetosis. A larger dataset is required for models to accurately represent distinct classes/scores. This study presents a novel methodology of automated assessment of movement disorders solely from video data.

Prediction of Potential Distribution of Carposina coreana in China under the Current and Future Climate Change.

Carposina coreana is an important pest of Cornus officinalis, distributed in China, Korea, and Japan. In recent years, its damage to C. officinalis has become increasingly serious, causing enormous economic losses in China. This study and prediction of current and future suitable habitats for C. coreana in China can provide an important reference for the monitoring, early warning, prevention, and control of the pest. In this study, the potential distributions of C. coreana in China under current climate and future climate models were predicted using the maximum entropy (MaxEnt) model with ArcGIS software. The distribution point data of C. coreana were screened using the buffer screening method. Nineteen environmental variables were screened using the knife-cut method and variable correlation analysis. The parameters of the MaxEnt model were optimized using the kuenm package in R software. The MaxEnt model, combined with key environmental variables, was used to predict the distribution range of the suitable area for C. coreana under the current (1971-2000) and four future scenarios. The buffer screening method screened data from 41 distribution points that could be used for modeling. The main factors affecting the distribution of C. coreana were precipitation in the driest month (Bio14), precipitation in the warmest quarter (Bio18), precipitation in the coldest quarter (Bio19), the standard deviation of seasonal variation of temperature (Bio4), minimum temperature in the coldest month (Bio6), and average temperature in the coldest quarter (Bio11). The feature class (FC) after the kuenm package optimization was a Q-quadratic T-threshold combination, and the regularization multiplier (RM) was 0.8. The suitable areas for C. coreana under the current climate model were mainly distributed in central China, and the highly suitable areas were distributed in southern Shaanxi, southwestern Henan, and northwestern Hubei. The lowest temperature in the coldest month (Bio6), the average temperature in the coldest quarter (Bio11), and the precipitation in the warmest quarter (Bio18) all had good predictive ability. In future climate scenarios, the boundary of the suitable area for C. coreana in China is expected to shift northward, and thus, most of the future climate scenarios would shift northward.

Thinking Inside the Bounds: Improved Error Distributions for Indifference Point Data Analysis and Simulation Via Beta Regression using Common Discounting Functions

Standard nonlinear regression is commonly used when modeling indifference points due to its ability to closely follow observed data, resulting in a good model fit. However, standard nonlinear regression currently lacks a reasonable distribution-based framework for indifference points, which limits its ability to adequately describe the inherent variability in the data. Software commonly assumes data follow a normal distribution with constant variance. However, typical indifference points do not follow a normal distribution or exhibit constant variance. To address these limitations, this paper introduces a class of nonlinear beta regression models that offers excellent fit to discounting data and enhances simulation-based approaches. This beta regression model can accommodate popular discounting functions. This work proposes three specific advances. First, our model automatically captures non-constant variance as a function of delay. Second, our model improves simulation-based approaches since it obeys the natural boundaries of observable data, unlike the ordinary assumption of normal residuals and constant variance. Finally, we introduce a scale-location-truncation trick that allows beta regression to accommodate observed values of 0 and 1. A comparison between beta regression and standard nonlinear regression reveals close agreement in the estimated discounting rate k obtained from both methods.

Estimating Trends of Sea-Level Fluctuations from Tide Gauge Data Analysis along Bangladesh Coast

The present study tries to determine the nature and magnitudes of relative sea-level to estimate present and future trends of mean sea-level (MSL) changes based on historical tide gauge data of Cox's Bazaar, Charchanga, Hiron Point, Khepupara and Sadarghat stations of Bangladesh. Annual PSMSL tide gauge dataset of Cox's Bazaar station shows 2.72 mm/year sea-level rise while CPA provided 'Metric' dataset of Sadarghat station demonstrate 3.93 mm/year sea-level rise along Chittagong coast. Accordingly, the sea­ levels are supposed to be increased 10.54 cm, 24.11 cm and 37.69 cm than the 1988 MSL (213.70 cm) at Cox's Bazaar by the years 2050, 2100 and 2150, respectively. Moreover, based on long-term PSMSL tide dataset of Charchanga, Hiron Point and Khepupara stations, the estimated rise of sea-levels along the respective coasts are 8.15.mm/year, 3:56 mm/year and 18.66 mm/year, respectively. Finally, 7.4 :mm/year average rate of sea-level rise has been calculated from all the studied datasets under present study. However, disregarding the exceptionally high rate (18.66 m.m/year) at Khepupara station determined average rate of sea level rise along Bangladesh coast is 4.59 mm/year which seems to be consistent with the global average. Consequently, the predicted sea level along Bangladesh coast are approximately 22 cm, 45 cm and 68 cm above their average MSL (210.4 cm) in the years 2050, 2100 and 2150, respectively. The Chittagong Univ. J. B. Sci. 8(1&2): 75-86, 2013

GEUKE: A geographic entities uniformly explicit knowledge embedding model

AbstractKnowledge embedding for geographic knowledge graphs can effectively improve computational efficiency and provide support for knowledge reasoning, knowledge answering and other applications of knowledge graphs. To maintain a more comprehensive understanding of spatial features through knowledge embedding, it is crucial to integrate the representation and computation of various entity types, encompassing points, lines, and polygons. This article proposes a geographic entities uniformly explicit knowledge embedding model (GEUKE). In GEUKE, spatial data of point, line, and polygon‐type geographic entities are expressed in the form of subgraphs, and space embedding is generated using a SubGNN‐based uniform spatial feature encoder. GEUKE improves the energy function in TransE to train spatial feature‐based embedding and structural‐based embedding of geographic entities into a unified vector space. Experimental results show that GEUKE has higher performance than TransE, TransH, TransD, and TransE‐GDR on link prediction and triple classification task. Within the spatial feature embedding process, GEUKE effectively preserves the inherent features of entities, encompassing location, neighborhood, and structural attributes, while simultaneously ensuring a coherent spatial data representation across all three entity types: points, lines, and polygons. By maintaining the spatial features of geographic entities and their interrelations, this capability unleashes the full potential of applications such as knowledge reasoning and geospatial question answering in a manner that is conducive to diverse geospatial scenarios.

Follow-up to Adolescence after Early Peanut Introduction for Allergy Prevention

BackgroundA randomized trial demonstrated consumption of peanut from infancy to age 5 years prevented the development of peanut allergy. An extension of that trial demonstrated the effect persisted after 1 year of peanut avoidance. This follow-up trial examined the durability of peanut tolerance at age 144 months after years of ad libitum peanut consumption.MethodsParticipants from a randomized peanut consumption trial were assessed for peanut allergy following an extended period of eating or avoiding peanuts as desired. The primary end point was the rate of peanut allergy at age 144 months.ResultsWe enrolled 508 of the original 640 participants (79.4%); 497 had complete primary end point data. At age 144 months, peanut allergy remained significantly more prevalent in participants in the original peanut avoidance group than in the original peanut consumption group (15.4% [38 of 246 participants] vs. 4.4% [11 of 251 participants]; P<0.001). Participants in both groups reported avoiding peanuts for prolonged periods of time between 72 and 144 months. Participants at 144 months in the peanut consumption group had levels of Ara h2-specific immunoglobulin E (a peanut allergen associated with anaphylaxis) of 0.03 ± 3.42 kU/l and levels of peanut-specific immunoglobulin G4 of 535.5 ± 4.98 μg/l, whereas participants in the peanut avoidance group had levels of Ara h2-specific immunoglobulin E of 0.06 ± 11.21 kU/l and levels of peanut-specific immunoglobulin G4 of 209.3 ± 3.84 μg/l. Adverse events were uncommon, and the majority were related to the food challenge.ConclusionsPeanut consumption, starting in infancy and continuing to age 5 years, provided lasting tolerance to peanut into adolescence irrespective of subsequent peanut consumption, demonstrating that long-term prevention and tolerance can be achieved in food allergy. (Funded by the National Institute of Allergy and Infectious Diseases and others; ITN070AD, ClinicalTrials.gov number, NCT03546413.)

Fast hardware-aware matrix-free algorithms for higher-order finite-element discretized matrix multivector products on distributed systems

Recent hardware-aware matrix-free algorithms for higher-order finite-element (FE) discretized matrix-vector multiplications reduce floating point operations and data access costs compared to traditional sparse matrix approaches. In this work, we address a critical gap in existing matrix-free implementations which are not well suited for the action of FE discretized matrices on very large number of vectors. In particular, we propose efficient matrix-free algorithms for evaluating FE discretized matrix-multivector products on both multi-node CPU and GPU architectures. To this end, we employ batched evaluation strategies, with the batchsize tailored to underlying hardware architectures, leading to better data locality and enabling further parallelization. On CPUs, we utilize even-odd decomposition, SIMD vectorization, and overlapping computation and communication strategies. On GPUs, we develop strategies to overlap compute with data movement for achieving efficient pipelining and reduced data accesses through the use of GPU-shared memory, constant memory and kernel fusion. Our implementation outperforms the baselines for Helmholtz operator action on 1024 vectors, achieving up to 1.4x improvement on one CPU node and up to 2.8x on one GPU node, while reaching up to 4.4x and 1.5x improvement on multiple nodes for CPUs (3072 cores) and GPUs (24 GPUs), respectively. We further benchmark the performance of the proposed implementation for solving a model eigenvalue problem for 1024 smallest eigenvalue-eigenvector pairs by employing the Chebyshev Filtered Subspace Iteration method, achieving up to 1.5x improvement on one CPU node and up to 2.2x on one GPU node while reaching up to 3.0x and 1.4x improvement on multi-node CPUs (3072 cores) and GPUs (24 GPUs), respectively.

A new method for mapping vegetation structure parameters in forested areas using GEDI data

The spatially continuous mapping of vegetation structure parameters in forested areas serves as a crucial foundation for research in various fields, including ecosystem ecology, climate change, hydrology, and forest management and protection. However, there is a notable scarcity of regional scale data regarding vegetation structure parameters. Therefore, our objective was to fill this data gap by providing a methodology for mapping vegetation structure parameters in forested areas at regional-scale. Global Ecosystem Dynamics Investigation (GEDI) provides precise observational data on vegetation structure parameters across the world. Using GEDI vegetation structure parameter point data, we developed a vegetation state coefficient based on the ratio of plant area index (PAI) and vegetation coverage (COVER). Subsequently, we proposed a method for decomposing vegetation coverage to simulate layered vegetation coverage. Taking Jiangxi Province as a case study, we integrated machine learning models to generate a spatially continuous map of vegetation structure parameters in forested areas, including layered coverage and layered leaf area index, in 2020. The map has a spatial resolution of 30 m and a vertical resolution of 5 m. The model exhibits excellent performance. Among the 140 sets of models, approximately 63 % of them achieved an R2 surpassing 0.6, while 89 % of the models achieved a root mean square error (RMSE) below 0.3. This study can serve as a valuable reference for the decomposition of vegetation structure parameters at the regional scale and provide a more precise depiction of the spatial characteristics of vertical vegetation structure at the regional scale. It contributes by providing data support and methodological guidance for research related to forest structure analysis, resource management, and ecological process simulation at the regional scale.

Research on Electricity Load Forecasting Based on LSTM Networks and BP Networks

The aim of this study is to compare the performance of Long Short-Term Memory (LSTM) network with Back Propagation Neural Network (BP network) for electricity load forecasting. Two prediction methods are used in the study: the first one uses a sliding window approach to predict the load of the next hour using the electricity load data of the past 24 hours and external factors such as temperature and humidity; the second one uses the data of the same time point of the past 24 days to predict the load of the corresponding time point of the next day. The results show that in the first method, the BP network outperforms the LSTM in prediction; while in the second method, both networks perform poorly, but the LSTM network outperforms the BP network in prediction. This finding provides new insights into power load forecasting, which is important for power system efficiency and reliability improvement.

Recreation specialization: Resource selection functions as a predictive tool for protected area recreation management

Widespread and ongoing increases in park and protected area (PPA) visitation presents managers with the imminent challenge of preserving ecological conditions while also maintaining accessibility in the open spaces that are fundamental to the health and wellness needs of society. In the field of recreation resource management, recreation specialization, the selective channeling of interests and abilities into a specific recreational activity, has contributed to a more comprehensive understanding of recreation behavior, site preference, management perceptions and conservation support. Contributing to historical understandings of specialization can inform recreation planning on the diversity of uses occurring in urban-proximate PPAs. In this study, survey and GPS data collected in urban and peri-urban parks in the Nature Reserve of Orange County, CA, USA, were used to classify recreationists into three specialization types based on empirically derived dimensions of involvement, commitment and skill. For each survey participant, thirteen spatio-temporal metrics (STM) were calculated. A principle component analysis (PCA) reduced STMs into 3 factors representing expressions of spatial behavior and a one-way analysis of variance indicated unique patterns between specialization types and time spent recreating, elevation gained, speed traveled and stopping behavior. Additionally, GPS point data were analyzed with an analytical approach adopted from the field of wildlife movement ecology; a resource selection function (RSF). The RSF assisted in quantifying spatial distribution patterns specific to each specialization type across activity types at four park locations and demonstrated a landscape-based statistical analysis of probability of use in relation to change in elevation, distance to starting points and park amenities. Committed hikers and mountain bikers largely demonstrated longer trip durations, more elevation gain, less overall stopping, faster overall speeds and farther total distances, while e-bikers in this study demonstrated unintuitive specialization dynamics; casual specialization types travelling farther from starting points and dispersing more across trail systems. Further spatial results revealed spatial behavior to be inherently complex and influenced by numerous confounding factors (i.e. activity type/bike type, starting points, park topography and trail design). This novel spatial examination of recreation specialization according to activity type and park location, is useful in understanding recreation behavior and park use in a spatial context. This ability is helpful in a predictive managerial context when reviewed in correspondence with historical evidence identifying behaviors with increased potential for ecological impact.

Quantitative measurement of urban spatial vitality by integrating physical built environment and subjective perception dimensions

Urban space vitality is a critical indicator for supporting rational urban spatial planning and updating and formulating sustainable development strategies. However, in many areas (e.g., aging urban areas), there is often a mismatch between the conditions of the physical built environment and its spatial attractiveness. Traditional methods based on physical space design theory often fail to accurately measure the spatial vitality of these areas. Street view images directly reflect the actual construction situation and effectively compensate for the lack of visual, subjective, perception dimension information. This study proposes a novel method that integrates objective and subjective dimensions to measure urban vitality, which is captured by incorporating spatial data of points of interest, building outlines, road networks, and street view images. Then, taking mobile phone signaling data as a source of ground truth validation, we choose Nanjing as a case study to demonstrate that our multidimensional fusion method exhibits higher explanatory power and better alignment with actual conditions by comparing it against single-dimensional methods. The results underscore the importance of integrating subjective and perceptual dimensions in measurements of urban vitality. We believe that the localized samples of the subjective perception survey will further enhance the accuracy and generalizability of this method in the future.

“Reinforcement learning particle swarm optimization based trajectory planning of autonomous ground vehicle using 2D LiDAR point cloud”

The advent of autonomous mobile robots has spurred research into efficient trajectory planning methods, particularly in dynamic environments with varied obstacles. This study focuses on optimizing trajectory planning for an Autonomous Ground Vehicle (AGV) using a novel Reinforcement Learning Particle Swarm Optimization (RLPSO) algorithm. Real-time mobile robot localization and map generation are introduced through the utilization of the Hector-SLAM algorithm within the Robot Operating System (ROS) platform, resulting in the creation of a binary occupancy grid. The present research thoroughly investigates the performance of the RLPSO algorithm, juxtaposed against five established Particle Swarm Optimization (PSO) variants, within the context of four distinct physical environments. The experimental design is tailored to emulate real-world scenarios, encompassing a spectrum of challenges posed by static and dynamic obstacles. The AGV, equipped with LiDAR sensors, navigates through diverse environments characterized by obstacles of different geometries. The RLPSO algorithm dynamically adapts its strategies based on feedback, enabling adaptable trajectory planning while effectively avoiding obstacles. Numerical results obtained from extensive experimentation highlight the algorithm's efficacy. The navigational model's validation is achieved within a MATLAB 2D virtual environment, employing 2D Lidar mapping point data. Transitioning to physical experiments with an AGV, RLPSO continues to demonstrate superior performance, showcasing its potential for real-world applications in autonomous navigation. On average, RLPSO achieves a 10–15 % reduction in path distances and traversal time compared to the following best-performing PSO variant across diverse scenarios. The adaptive nature of RLPSO, informed by feedback from the environment, distinguishes it as a promising solution for autonomous navigation in dynamic settings, with implications for practical implementation in real-world scenarios.

Spatio‐temporal data integration for species distribution modelling in R‐INLA

Abstract Species distribution modelling is a highly used tool for understanding and predicting biodiversity change, and recent work has emphasised the importance of understanding how species distributions change over both time and space. Spatio‐temporal models require large amounts of data spread over time and space, and as such are clear candidates to benefit from model‐based integration of different data sources. However, spatio‐temporal models are highly computationally intensive and integrating different data sources can make this approach even more unfeasible to ecologists. Here we demonstrate how the R‐INLA methodology can be used for model‐based data integration for spatio‐temporally explicit modelling of species distribution change. We demonstrate that this method can be applied to both point and areal data with two contrasting case studies, one using the SPDE approach for modelling spatio‐temporal change in the Gatekeeper butterfly (Pyronia tithonus) across Great Britain and the second using a spatio‐temporal areal model to describe change in caddisfly (Trichoptera) populations across the River Thames catchment. We show that in the caddisfly case study integrating together different data sources led to greater understanding of the change in abundance across the River Thames both seasonally and over 5 years of data. However, in the butterfly case study moving to a spatio‐temporal context exacerbated differences between the data sources and resulted in no greater ecological insight into change in the Gatekeeper population. Our work provides a computationally feasible framework for spatio‐temporally explicit integration of data within SDMs and demonstrates both the potential benefits and the challenges in applying this methodology to real ecological data.

Investigating the spatial-temporal pattern of the trophic state index in Hanoi’s largest urban lake (Vietnam) using Sentinel-2 images

ABSTRACT Urban lakes provide aesthetic, recreational and ecological benefits, making understanding their trophic status vital for effective management. Utilizing Sentinel-2 imagery, we investigated the spatial-temporal patterns of the Trophic State Index (TSI) in Hanoi’s West Lake over a 2-year period. Our study proposed a robust TSI estimation model, incorporating three band ratios: R rs 655 / R rs 490 , R rs 705 / R rs 655 , R rs 655 / R rs 443 , along with 124 in situ point data. The model exhibited high accuracy (the determination coefficient, R 2 = 0.63) and minimal errors (the root mean square errors, RMSE = 1.1 and 0.9). Consequently, West Lake consistently registered as hypereutrophic (TSI > 70), with seasonal variations linked to climate factors, leading to algal blooms and reduced water clarity. Spatial analysis highlighted the influence of sewage drains on TSI, indicating an urgent need for wastewater control. Our study demonstrates the cost-effective utility of Sentinel-2 imagery for monitoring urban lake eutrophication and emphasizes the urgency of managing eutrophication in these vital ecosystems.

Remote sensing estimation of δ15NPN in the Zhanjiang Bay using Sentinel-3 OLCI data based on machine learning algorithm

Remote sensing estimation of δ15NPN in the Zhanjiang Bay using Sentinel-3 OLCI data based on machine learning algorithm

Reg-Tune: A Regression-Focused Fine-Tuning Approach for Profiling Low Energy Consumption and Latency

Fine-tuning deep neural networks is pivotal for creating inference modules that can be suitably imported to edge or field-programmable gate array (FPGA) platforms. Traditionally, exploration of different parameters throughout the layers of deep neural networks has been done using grid search and other brute force techniques. Although these methods lead to the optimal choice of network parameters, the search process can be very time consuming and may not consider deployment constraints across different target platforms. This work addresses this problem by proposing Reg-Tune, a regression-based profiling approach to quickly determine the trend of different metrics in relation to hardware deployment of neural networks on tinyML platforms like FPGAs and edge devices. We start by training a handful of configurations belonging to different combinations of \(\mathcal {NN}\scriptstyle \langle q (quantization),\,s (scaling)\rangle \displaystyle\) or \(\mathcal {NN}\scriptstyle \langle r (resolution),\,s\rangle \displaystyle\) workloads to generate the accuracy values respectively for their corresponding application. Next, we deploy these configurations on the target device to generate energy/latency values. According to our hypothesis, the most energy-efficient configuration suitable for deployment on the target device is a function of the variables q , r , and s . Finally, these trained and deployed configurations and their related results are used as data points for polynomial regression with the variables q , r , and s to realize the trend for accuracy/energy/latency on the target device. Our setup allows us to choose the near-optimal energy-consuming or latency-driven configuration for the desired accuracy from the contour profiles of energy/latency across different tinyML device platforms. To this extent, we demonstrate the profiling process for three different case studies and across two platforms for energy and latency fine-tuning. Our approach results in at least 5.7 \(\times\) better energy efficiency when compared to recent implementations for human activity recognition on FPGA and 74.6% reduction in latency for semantic segmentation of aerial imagery on edge devices compared to baseline deployments.

Development of structural schemes for collecting monitoring and diagnostic data of signaling points of the railway run based on programmable logic controllers

Development of structural schemes for collecting monitoring and diagnostic data of signaling points of the railway run based on programmable logic controllers

Light-Adaptive Human Body Key Point Detection Algorithm Based on Multi-Source Information Fusion.

The identification of key points in the human body is vital for sports rehabilitation, medical diagnosis, human-computer interaction, and related fields. Currently, depth cameras provide more precise depth information on these crucial points. However, human motion can lead to variations in the positions of these key points. While the Mediapipe algorithm demonstrates effective anti-shake capabilities for these points, its accuracy can be easily affected by changes in lighting conditions. To address these challenges, this study proposes an illumination-adaptive algorithm for detecting human key points through the fusion of multi-source information. By integrating key point data from the depth camera and Mediapipe, an illumination change model is established to simulate environmental lighting variations. Subsequently, the fitting function of the relationship between lighting conditions and adaptive weights is solved to achieve lighting adaptation for human key point detection. Experimental verification and similarity analysis with benchmark data yielded R2 results of 0.96 and 0.93, and cosine similarity results of 0.92 and 0.90. With a threshold range of 8, the joint accuracy rates for the two rehabilitation actions were found to be 89% and 88%. The experimental results demonstrate the stability of the proposed method in detecting key points in the human body under changing illumination conditions, its anti-shake ability for human movement, and its high detection accuracy. This method shows promise for applications in human-computer interaction, sports rehabilitation, and virtual reality.

Population data evidence of interdependence of the limbs of hormonal feedback loops.

The fundamental models underlying hormonal physiological regulation and homeostasis remain poorly understood. We aimed to derive quantitative evidence regarding these models from the study of population data of balance points of different parameters and their respective controlling hormones. We studied the slopes of correlations between concentrations of circulating free thyroxine and thyrotropin, calcium and parathyroid hormone, hemoglobin and erythropoietin, and glucose and insulin in such population data, as well as the slopes of the limbs of various feedback loops estimated empirically and by reverse engineering of the population data. We used computer simulations to model the factors that influence the slopes derived from the population data, and then matched these simulations with the empirically derived slopes. Our simulations showed that changes to the population distribution of feedback loop limbs may alter the slopes of correlations within population data in specific ways. Non-random (interdependent) associations of the limbs of feedback loops may also have this effect, as well as producing discrepancies between the slopes of feedback limb loops determined experimentally and the same slopes determined by derivation from population data. Our corresponding empirical findings were consistent with the presence of such interdependence in the free thyroxine/thyrotropin, hemoglobin/erythropoietin, and glucose/insulin systems. The glucose/insulin data provided evidence consistent with increasing interdependence with age in childhood. Our findings therefore provide strong evidence that the interdependence of the limbs of feedback loops is a general feature of endocrine homeostatic regulation. This interdependence potentially bestows evolutionary homeostatic and regulatory advantages.

Early autism diagnosis based on path signature and Siamese unsupervised feature compressor.

Autism spectrum disorder has been emerging as a growing public health threat. Early diagnosis of autism spectrum disorder is crucial for timely, effective intervention and treatment. However, conventional diagnosis methods based on communications and behavioral patterns are unreliable for children younger than 2 years of age. Given evidences of neurodevelopmental abnormalities in autism spectrum disorder infants, we resort to a novel deep learning-based method to extract key features from the inherently scarce, class-imbalanced, and heterogeneous structural MR images for early autism diagnosis. Specifically, we propose a Siamese verification framework to extend the scarce data, and an unsupervised compressor to alleviate data imbalance by extracting key features. We also proposed weight constraints to cope with sample heterogeneity by giving different samples different voting weights during validation, and used Path Signature to unravel meaningful developmental features from the two-time point data longitudinally. We further extracted machine learning focused brain regions for autism diagnosis. Extensive experiments have shown that our method performed well under practical scenarios, transcending existing machine learning methods and providing anatomical insights for autism early diagnosis.