Subjective Bias Research Articles

School ties and evaluation outcomes: Evidence from the Korean Basketball League

AbstractThis paper investigates whether evaluation outcomes are influenced by private information and subjective biases when appraisers and appraisees are socially connected. In the Korean Basketball League, school ties between referees and players are commonplace because a large proportion of referees were previously basketball players. Using data from six basketball seasons, we analyze the degree to which referees' decisions are affected when players and referees attended the same school. Our results suggest that players who play under referees with school ties are called for fewer fouls. We also find that the results are driven mainly by high school ties.

FEI-YOLO: A Lightweight Soybean Pod-Type Detection Model

Identifying and statistically analyzing soybean pod types are crucial for seed evaluation and yield estimation. Traditional visual assessment by breeding personnel is time-consuming, labor-intensive, and prone to subjective bias, especially with large datasets. Automatic assessment methods usually struggle with the highly confusing pod types with two and three seeds, affecting the model’s identification accuracy. To address these issues, we propose to improve the standard YOLOv5s object detection model to enhance the differentiation between pod types and to boost the model’s efficiency in prediction. To reduce the number of parameters and the computational load, we propose to introduce the FasterNet Block module in the FasterNet model into the original C3 module, leading to improvements in both detection accuracy and speed. To strengthen the feature extraction and representation for specific targets, the Efficient Multi-Scale Attention (EMA) module is incorporated into the C3 module of the backbone network, improving the identification of similar pod types. The Inner-IoU is combined with the CIoU as the loss function to further enhance detection accuracy and generalization. Experiments comparing FEI-YOLO with the baseline YOLOv5s show that FEI-YOLO achieves an mAP@0.5 of 98.6% and an mAP@0.5:0.95 of 81.1%, with improvements of 1.5% and 1.4%, respectively. Meanwhile, the number of parameters is reduced by 13.2%, and FLOPs decreased by 10.8%, demonstrating the model’s effectiveness and efficiency, enabling rapid and accurate identification of soybean pod types from images.

Transcending Individual Traditions: Zhang Taiyan’s Interpretation of Zhuangzi’s Notion of Chengxin

This article examines the nuanced interpretations of chengxin 成心 in the Zhuangzi 莊子, a foundational Daoist text, across different philosophical traditions. Historically, Daoist thinkers like Guo Xiang 郭象 and Cheng Xuanying 成玄英, along with Neo-Confucian scholars such as Lü Huiqing 吕惠卿 and Lin Xiyi 林希逸, have provided differing perspectives on the notion of chengxin. They regard it as either a source of subjective bias or an expression of heavenly principle. The core focus of this study is Zhang Taiyan’s 章太炎 innovative interpretation of chengxin as seeds (bijas) within ālaya-vijñāna (storehouse consciousness), integrating Yogācāra Buddhist concepts with traditional Daoist philosophy. The study conducts a textual analysis of Zhang’s Qiwulun Shi 齊物論釋, comparing his interpretation with earlier commentaries on the Zhuangzi. It is argued that Zhang’s approach is profoundly influenced by the Buddhist doctrine of non-duality, which underscores the interconnectedness of all phenomena. While Zhang’s interpretation has faced criticism from contemporaries and later philosophers for misunderstandings of Yogācāra theory and Kantian philosophy, it employs the method of geyi 格義 (matching concepts), not to impose one tradition over another, but to foster a dialogical process where the insights from each tradition mutually inform and transform one another. This article highlights the interpretative openness of the Zhuangzi, which invites diverse interpretations, and demonstrates how Zhang’s approach enriches this tradition by synthesizing various philosophical and religious frameworks. Zhang’s methodology remains relevant, fostering cross-cultural and inclusive philosophical dialog.

Evaluating OSPE as a Formative Assessment Tool for Medical Students

Introduction. Objective structured practical examination (OSPE) is a globally recognized and widely implemented assessment method used for evaluating students in practical examinations. It requires examinees to complete various tasks within a set time frame, each assessed against predetermined criteria designed to measure practical skills and attitudes. This study aimed to assess the effectiveness of the OSPE as a tool for internal assessment in the subject of Anatomy for first-year MBBS students, comparing its performance against the traditional practical examination (TPE) method. Methods. In this prospective study, an OSPE tool specific to Anatomy was developed and validated for face and content to ensure that the OSPE stations were standardized and well-constructed. The part-leaving test covered the Thorax and Brain chapters using both the TPE and OSPE methods. A Bland-Altman plot was employed to assess the agreement between the quantitative assessments of both methods. Additionally, a 5-point Likert scale was utilized to gauge the perceptions of both students and faculty members regarding the assessment methods. Results. The analysis revealed a significant difference between the scores obtained from OSPE and TPE. The intraclass correlation coefficients for reliability between TPE and OSPE were 0.81 for the Thorax chapter and 0.84 for the Brain chapter. Additionally, the Bland-Altman plots illustrating the scores from both assessment methods indicated low bias and no definite pattern. While no significant outliers were identified for the Brain part, the Thorax part exhibited outliers at a rate of 6.43%, suggesting the presence of low-impact erratic variables. The Chi-square test for student perceptions yielded significant results for the entire questionnaire. In contrast, faculty perceptions showed significant results for all questions except for two. Conclusions. Due to the high intraclass correlation coefficients between TPE and OSPE and the absence of bias, systematic errors, and erratic variables, OSPE is a promising assessment tool. Students perceived that OSPE decreased the chances of failure and that the questions asked in OSPE were more relevant compared to the traditional method. Faculty members also noted that OSPE provided reasonably high transparency and decreased subjective bias compared to TPE. Considering these factors, it can be concluded that OSPE is a viable potential tool to replace TSE.

Emergency Capability Evaluation of Port-Adjacent Oil Storage and Transportation Bases: An Improved Analytic Hierarchy Process Approach

The large-scale storage and stable supply of oil products are essential for national energy security and economic development. As the economy expands and energy demands rise, centralized storage and supply systems become increasingly vital for ensuring the efficiency and reliability of oil product distribution. However, large oil storage depots present substantial safety risks. In the event of fires, explosions, or other accidents, emergency response efforts face stringent demands and challenges. To enhance the emergency response capabilities of oil storage and transportation bases (OSTBs), this paper proposes an innovative approach that integrates the improved analytic hierarchy process (IAHP) with the Entropy Weight Method (EMW) to determine the combined weights of various indices. This approach reduces the subjective bias associated with the traditional analytic hierarchy process (AHP). The emergency response capabilities of OSTBs are subsequently evaluated through fuzzy comprehensive analysis. An empirical study conducted on an OSTB in the Zhoushan archipelago quantitatively assesses its emergency preparedness. The results show that the base excels in pre-incident prevention, demonstrates robust preparedness and response capabilities, and exhibits moderate recovery abilities after incidents. These findings provide a theoretical foundation for reducing the likelihood of accidents, enhancing emergency response efficiency, and mitigating the severity of consequences. Practical recommendations are also offered based on the results.

Perceptions of Dental Faculty and Postgraduate Residents towards Learning Professionalism through DOPS Assessments of Implants

ABSTRACT Objective: One of the essential characteristics that define a career is professionalism, which encompasses commitments, actions, standards, and objectives. Background: The study was aimed to assess the perception of dental faculty and residents towards DOPS as a method to assess professionalism and also the perception of participants towards inclusion of professionalism assessments in the dental curriculum. Method: This descriptive cross sectional study was conducted amongst forty three Prosthodontic postgraduate residents and twenty faculty members from ten different dental colleges in India in the year 2023.Views regarding professionalism were assessed using the Modified Australian Medical Council (AMC) DOPS assessment form. We measured Professionalism score from the DOPS assessments conducted six times within a year of the residents in different dental colleges of India. The resident was evaluated by two different faculty members in order to minimize subjective bias. The scores range from 1 to 5, with 1 representing the least amount of professionalism displayed and 5 representing the best. Data was analyzed using SPSS version 23. Descriptive statistics was done using mean score. Results: DOPS Assessment score of professionalism for Prosthodontic postgraduate residents and feedback from participants, post graduates and faculty at the end of last assessment was considered for conclusion. They showed increase in DOPS score and positive attitude towards dental professionalism. The participants believed that DOPS assessment is an effective way to improve professionalism. Conclusion: Academic dishonesty may continue with unethical behaviors in professional practice of a healthcare professional. Hence, it becomes vital to address academic dishonesty lapses among oral health sciences students to ensure the future safety of patients and improvement in patient care.

Risk assessment method for civil aircraft approach and landing at high plateau based on QAR data

Civil aircraft approach and landing at high altitude is a high-risk stage of plateau flight. In order to effectively implement the risk identification and grade judgment of approach and landing at high altitude, an LSTM-DNN deep learning risk assessment method based on variable fuzzy recognition with weights removed is proposed. This method is first based on the high-altitude flight data recorded by the quick access recorder (QAR), and draws on the civil aircraft flight quality monitoring project specification (FOQA) consultation notice and industry QAR monitoring standards. Combined with the indicator importance analysis and Delphi expert survey, five key monitoring items are extracted as the risk assessment indicators of civil aircraft approach and landing at high altitude, including large heading change during landing, low track, large descent rate during 2000-1000 ft approach, vertical acceleration at touchdown, and large descent rate during 500-50ft approach. Then, in order to overcome the subjective bias of the evaluation indicator weight, the weight of the evaluation indicator is determined by the weight removal method, and the risk grade membership function is constructed based on the variable fuzzy recognition method. Finally, a risk assessment model for civil aircraft approach and landing at high altitude based on LSTM-DNN is established. Taking the Chengdu-Lhasa approach and landing segment as an example, the QAR data was extracted to train and test the risk assessment model, and the results were compared with those of Logistic multivariate regression, support vector machine (SVM) and other evaluation methods. The results show that the average recognition rate of the LSTM-DNN deep learning model based on variable fuzzy recognition with weights removed reaches 94.18%, and the highest can reach 94.79%, which verifies the objective effectiveness of the risk assessment method.

Effect of Level and Frequency of Forward Masker on Auditory Brainstem Response.

Forward masking (FM) is characterized by the perception of a signal being reduced or wholly masked due to a preceding sound (masker) of the same or different frequencies that offers a challenge for the auditory system to resolve. Considering that the off-frequency masker is expected to undergo linear processing compared to the on-frequency masker at the signal place, it reflects the peripheral auditory systems' compressive response. Thus, the present study focused on employing FM electrophysiological analogous such as auditory brainstem responses (ABR) to the behavioral masking experiments to objectively measure the frequency and level of processing in the auditory system, from the periphery to the brainstem level. The study was an observational research on 21 female volunteers. ABR was obtained using a tone-on-tone FM paradigm for 1000- and 4000-Hz probe stimuli. An experiment used two forward maskers, on-frequency and off-frequency, with varying levels from 50 to 70 dB SPL. A progressive shift for Vth peak latency and reduction in response amplitude was observed in proportion to the increase of masker level for both the probe stimuli and the masking experiments. However, ABR responses in neither masking condition were observed to differ between 60 and 70 dB SPL. FM ABR experiments are an assessment tool for estimating frequency and level processing in the auditory system, providing good efficiency, reliability, and less subject bias compared to behavioral measures.

Inertial Measurement Unit-Based Frozen Shoulder Identification from Daily Shoulder Tasks Using Machine Learning Approaches.

Frozen shoulder (FS) is a common shoulder condition accompanied by shoulder pain and a loss of shoulder range of motion (ROM). The typical clinical assessment tools such as questionnaires and ROM measurement are susceptible to subjectivity and individual bias. To provide an objective evaluation for clinical assessment, this study proposes an inertial measurement unit (IMU)-based identification system to automatically identify shoulder tasks whether performed by healthy subjects or FS patients. Two groups of features (time-domain statistical features and kinematic features), seven machine learning (ML) techniques, and two deep learning (DL) models are applied in the proposed identification system. For the experiments, 24 FS patients and 20 healthy subjects were recruited to perform five daily shoulder tasks with two IMUs attached to the arm and the wrist. The results demonstrate that the proposed system using deep learning presented the best identification performance using all features. The convolutional neural network achieved the best identification accuracy of 88.26%, and the multilayer perceptron obtained the best F1 score of 89.23%. Further analysis revealed that the identification performance based on wrist features had a higher accuracy compared to that based on arm features. The system's performance using time-domain statistical features has better discriminability in terms of identifying FS compared to using kinematic features. We demonstrate that the implementation of the IMU-based identification system using ML is feasible for FS assessment in clinical practice.

Integrating Whole Slide Imaging and Artificial Intelligence Provides Higher Accuracy in Thyroid Cytology

Abstract Introduction/Objective The aim of the study is to evaluate the potential advantages of integrating Whole Slide Imaging (WSI) and Artificial Intelligence (AI) to increase efficiency and diagnostic accuracy in thyroid cytology, especially in indeterminate lesions. Methods/Case Report A retrospective database search from 2000-2023 was performed, and the concordance with the pathologists was evaluated based on the manual or AI-based Region of Interest (ROI) identification, the best- suited AI, algorithm design, color enhancement, and combination of AI and Electronic Medical Records (EMR). Results (if a Case Study enter NA) The cohort included 3285 WSI and 121,384 ROI. In 109 WSI with manually diagnosed ROI, Deep learning successfully predicted benign and malignant cases, including 11 indeterminate cases, with zero false positives or negatives for 35% of cases. In the second study with 908 WSI, 4494 ROI were identified with Convolutional Neural Network (CNN). A higher area under the curve (AUC) resulted when EMR and Machine Learning algorithm (MLA) were integrated: EMR: 0.931, MLA: 0932, EMR+ MLA; 0.962, Sensitivity: 92.0%, Specificity 90.5%. In the next study with 145 WSI, the CNN-identified ROI was presented to the same pathologist after 117 days. Perfect concordance was seen between the two methods (K score 0.924, time to diagnosis;81.59 Second). In 360 WSI, YOLOV4 algorithm was trained to detect the ROI, and detection and classification rates were used to evaluate the accuracy. It had a precision of 81.84% (3.16% more than the single detection networks of 78.68%). In the last study with 964 WSI, color augmentation improved the AUC for indeterminate cases from 88.3% to 96.2%. Conclusion CNN is a better-suited AI for whole slide imaging due to its compatibility with image data and ability to automatically detect important features without human supervision. It also removes the component of subjective bias in interpretation. Color augmentation and integration of EMR can overcome the unique challenges in thyroid indeterminate cytology.

A novel approach to the cause of death identification—multi-strategy integration of multi-organ FTIR spectroscopy information using machine learning

Identifying the cause of death has always been a major focus and challenge in forensic practice and research. Traditional techniques for determining the causes of death are time-consuming, labor-intensive, have high professional barriers, and are vulnerable to significant subjective bias. Additionally, most current studies on causes of death are limited to specific organs and single causes. To overcome these challenges, this study utilized simple and rapid fourier transform infrared spectroscopy (FTIR) detection technology, integrating data from six organs—heart, liver, spleen, lung, kidney, and brain. The optimum model for identifying seven different causes of death was determined by evaluating the performance of models developed using the model efficiencies of single-organ (SO), single-organ model fusion (SOMF), multi-organ data fusion (MODF), and multi-organ data model fusion (MODMF) modeling methods. Considering factors such as operational costs, model performance, and model complexity, the MODF artificial neural network (ANN) model was found to be the most suitable choice for constructing a cause of death identification model, with a cross-validation mean accuracy of 0.960 and a test set accuracy of 0.952. The heart and kidney contributed more spectral features to the construction of the cause of death identification model compared to other organs. This study not only demonstrated that data fusion and model fusion are effective strategies for improving model performance but also provided a comprehensive data analysis framework and process for modeling with small sample multi-modal data (multiple organ data). In conclusion, by exploring various approaches, this study offers new solutions for identifying the cause of death.

Revisiting 30 years of advances in firefighter protective clothing: a quantitative assessment and bibliometric analysis

Firefighter protective clothing research has undergone rapid development and garnered extensive attention over the past three decades. However, its trajectory lacks quantitative evaluation due to the complex relationship among humans, clothing, and the environment. Here, we aimed to (1) develop an objective and reproducible evaluation index system for the quantitative assessment of firefighter protective clothing research, (2) delineate the developmental stages and their analytical characteristics over the past three decades through bibliometric analysis, and (3) scrutinize the current status, research hotspots, gaps, and future concerns. Our review of 314 publications from the Web of Science Core database (1994–2023) led to the creation of an evaluation index system that integrates the catastrophe progression method with bibliometric analysis, incorporating 10 indicators across 4 dimensions: productivity, influence, development, and innovation. The bibliometric analysis was conducted using the Bibliometrix R package. The results reveal a rapid increase in firefighter protective clothing studies from 1994 to 2023, comprising two stages: the start-up phase (1994–2009) and the rapid development phase (2010–2023). Thematic evolution over three decades indicates a shift from single specific to composite domains, with emerging themes such as achieving a balance in the interaction between the human body, clothing, and environment signaling future research directions. In addition, we visualized the most relevant countries, institutions, journals, authors, articles, and potential challenges during the last 30 years. This review provides an overview of current evidence and minimizes subjectivity bias compared to traditional manual reviews, thus presenting specific future research questions advancing firefighter protective clothing research.

Research on lunar regolith of the Chang'E-4 landing site: An automated analysis method based on deep learning framework

On January 3, 2019, the Chang'E-4 lander successfully landed within the Von Kármán crater, located in the South P ole-Aitken Basin (SPA) on the farside of the Moon (45.5°S, 177.6°E), marking the first soft landing on the lunar farside. The lander, equipped with the Lunar Penetrating Radar (LPR) system, aimed to provide insights into the structure and evolution of the Moon. Previous research often relied on manually identifying hyperbolic features to analyze the lunar shallow subsurface properties. This inefficient approach may lead to subjective biases, resulting in unstable outcomes. This research constructed an automatic analysis framework by integrating the Swin Transformer with a 3D velocity spectrum, which is then applied to analyze the properties of the Chang'E-4 LPR data. The experimental results indicate that the framework achieved a precision of 98.9 % and a recall of 96.7 % in hyperbolic feature identification, with an F1 of 0.9782 and AP of 94.8 %. Additionally, it has been experimentally validated that the framework can accurately invert hyperbolic features' two-way travel time and velocity. Finally, the framework is applied to analyze the lunar shallow subsurface structure and properties within the landing area of the Chang'E-4 mission.

Twenty years’ empirical research on teacher resilience: a bibliometric analysis

ABSTRACT Teacher resilience has become a prominent and multidimensional subject of interest in recent educational research. Its growing recognition has fuelled a wealth of investigations, necessitating a quantitative science mapping approach for comprehensive understanding of this evolving domain. This paper conducts a bibliometric analysis of empirical teacher resilience research using Web of Science data. The analysis reveals that teacher resilience is both an emerging and well-established research area. Leading academic institutions, primarily in the U.S.A. Australia, and the UK, have spearheaded investigations from psychological and teacher education perspectives. Thematic analysis identifies five core areas: teacher resilience in educational psychology, mindfulness-based programmes to mitigate teacher stress and burnout, teacher resilience in initial teacher education, the interplay between self-efficacy, identity, and emotional intelligence, and the profound impact of teacher resilience on critical factors like retention, attrition, and motivation. This research sheds light on the current state of teacher resilience research and offers insights into future directions. It serves as a valuable resource for scholars, policymakers, and educators navigating this evolving landscape. The combination of bibliometric analysis and thematic content analysis minimises potential subjective bias, contributing to understanding the multifaceted field of teacher resilience and its implications for the future of education.

Managing transit-oriented development: A comparative analysis of expert groups and multi-criteria decision making methods

A key challenge for transport managers and planners in sustainable development is evaluating transit facilities' performance. While Multi-Criteria Decision-Making (MCDM) tools are often used, they can be influenced by experts' subjective biases. This study applies MCDM to assess Mass Rapid Transit (MRT) stations in Jakarta, Indonesia, focusing on Transit-Oriented Development (TOD). The primary goal is to compare stakeholder perspectives and MCDM methods, complemented by a sensitivity analysis and validation with real-world smart card data. The findings reveal significant differences in criteria weighting between Indonesian and non-Indonesian experts, and between academic and non-academic experts, especially in transit connectivity and land use diversity. The study also shows variations in station rankings across different MCDM methods. Sensitivity analysis identifies transit connectivity as the most critical criterion. Simple Additive Weighting (SAW) with linear normalisation aligns well with actual usage data and shows robustness in sensitivity analysis, making it the most reliable method for TOD evaluation. The study highlights the need for continuous TOD performance monitoring and the regular collection of real-world data on ridership and TOD indicators.

Generalization and enhancement of the effects of an active placebo nasal spray on sadness

IntroductionThe placebo effect, i.e., the psychobiological response arising from administering an inert treatment, influences various domains, such as pain perception and emotional regulation. Positive framing might enhance this effect. This study tested whether the effect of an active placebo (mimicking drug side effects to enhance treatment credibility) on is generalized between two different contexts of sadness induction and if positive framing of side effects enhances this effect. MethodsNinety-six healthy participants were randomly assigned to one of three groups: Placebo+positive framing (PPF), Placebo+standard information (PSI), or no treatment control (NTC). Participants underwent a sadness induction protocol during an in-person lab session and a 20-min online follow-up at home six hours later. Primary outcome was self-reported sadness, secondary outcome was self-reported side effects. ResultsBoth the PPF and PSI groups showed a significant decrease in sadness compared to the NTC group after placebo administration during the lab session (p < 0.001) and at follow-up (p < 0.05). At follow-up, only the PPF group did not experience a significant increase in sadness. Positive framing did not improve side effect tolerability. LimitationsSelf-reported measures introduce subjective bias. The sample restriction to healthy volunteers limits generalizability. The six-hour period may not capture clinically relevant long-term effects. ConclusionsThe active placebo nasal spray effectively reduced sadness, with effects persisting for six hours and across different contexts. Positive framing did not enhance side effect tolerability but may have helped maintain effectiveness at follow-up. Further research is needed in clinical populations and to explore long-term effects.

Classifying the clouds of Venus using unsupervised machine learning

Because Venus is completely shrouded by clouds, they play an important role in the planet’s atmospheric dynamics. Studying the various morphological features observed on satellite imagery of the Venusian clouds is crucial to understanding not only the dynamic atmospheric processes, but also interactions between the planet’s surface structures and atmosphere. While attempts at manually categorizing and classifying these features have been made many times throughout Venus’ observational history, they have been limited in scope and prone to subjective bias. We therefore present and investigate an automated, objective, and scalable approach for their classification using unsupervised machine learning that can leverage full datasets of past, ongoing, and future missions.To achieve this, we introduce a novel framework to generate nadir observation patches of Venus’ clouds at fixed consistent scales from satellite imagery data of the Venus Express and Akatsuki missions. Such patches are then divided into classes using an unsupervised machine learning approach that consists of encoding the patch images into feature vectors via a convolutional neural network trained on the patch datasets and subsequently clustering the obtained embeddings using hierarchical agglomerative clustering.We find that our approach demonstrates considerable accuracy when tested against a curated benchmark dataset of Earth cloud categories, is able to identify meaningful classes for global-scale (3000km) cloud features on Venus and can detect small-scale (25km) wave patterns. However, at medium scales (∼500km) challenges are encountered, as available resolution and distinctive features start to diminish and blended features complicate the separation of well defined clusters.

A deep learning approach for deriving wheat phenology from near-surface RGB image series using spatiotemporal fusion

Accurate monitoring of wheat phenological stages is essential for effective crop management and informed agricultural decision-making. Traditional methods often rely on labour-intensive field surveys, which are prone to subjective bias and limited temporal resolution. To address these challenges, this study explores the potential of near-surface cameras combined with an advanced deep-learning approach to derive wheat phenological stages from high-quality, real-time RGB image series. Three deep learning models based on three different spatiotemporal feature fusion methods, namely sequential fusion, synchronous fusion, and parallel fusion, were constructed and evaluated for deriving wheat phenological stages with these near-surface RGB image series. Moreover, the impact of different image resolutions, capture perspectives, and model training strategies on the performance of deep learning models was also investigated. The results indicate that the model using the sequential fusion method is optimal, with an overall accuracy (OA) of 0.935, a mean absolute error (MAE) of 0.069, F1-score (F1) of 0.936, and kappa coefficients (Kappa) of 0.924 in wheat phenological stages. Besides, the enhanced image resolution of 512 × 512 pixels and a suitable image capture perspective, specifically a sensor viewing angle of 40° to 60° vertically, introduce more effective features for phenological stage detection, thereby enhancing the model’s accuracy. Furthermore, concerning the model training, applying a two-step fine-tuning strategy will also enhance the model’s robustness to random variations in perspective. This research introduces an innovative approach for real-time phenological stage detection and provides a solid foundation for precision agriculture. By accurately deriving critical phenological stages, the methodology developed in this study supports the optimization of crop management practices, which may result in improved resource efficiency and sustainability across diverse agricultural settings. The implications of this work extend beyond wheat, offering a scalable solution that can be adapted to monitor other crops, thereby contributing to more efficient and sustainable agricultural systems.

Smart detection of subsurface anomalies: Concept, validation and applications

Naturally formed and engineering-induced subsurface anomalies (e.g., cavities and sinkholes) jeopardize infrastructure safety and hinder urban sustainability. Here we report a smart sensing method to detect subsurface anomalies based on the physical characteristics extracted from the effective signals scattered and reflected directly from these anomalies. Potential anomalies at submeter scales can be interpreted based on a sharp variation of anomaly score relative to the background anomaly score, showing the advantage of overcoming subjective uncertainty and biases involved in the traditional geophysical methods. We find that the use of scattered and reflected waves in an intermediate frequency range is well-suited for sensing deep subsurface infrastructure at high resolutions required for civil structures. We also demonstrate that a fast and reliable detection of subsurface anomalies relies solely on the physical characteristics of seismic data in two field cases, promoting geologic hazard forecast and decision-making effectiveness.

RTG-GNN: A novel rock topology-guided approach for permeability prediction using graph neural networks

Digital core permeability is a crucial factor in rock engineering, reservoir simulation, and underground applications, serving as the foundation for evaluating fluid flow underground. However, predicting digital core permeability using artificial intelligence presents challenges such as high model calculation complexity, strong resource dependence, and low efficiency. To overcome these challenges, this paper presents a novel hybrid enhanced rock topology-guided graph neural network (RTG-GNN) to accurately characterize the pore structure of digital cores and predict permeability. This model takes advantage of the physical attributes of pores and throats, utilizing them as features for nodes and edges. It improves information exchange through a gating mechanism and incorporates an attention mechanism for weighted aggregation of neighboring data, culminating in accurate permeability predictions. Furthermore, a scheme for constructing a comprehensive permeability distribution dataset (CPDD) is proposed for digital cores. This scheme effectively mitigates challenges like inadequate training data, subjective sampling bias, and data drift. Multiple experimental results demonstrate that RTG-GNN surpasses most methods in various performance indicators. Additionally, the RTG-GNN model boasts a lightweight design, occupying only 17% to 44% of the size of mainstream models and requiring just 15% to 38% of GPU resources for training. Remarkably, while maintaining high accuracy and low error rates, the computational load is reduced by a factor ranging from 19.06 to 636.41, the training speed is enhanced by 12.12 to 86.47 times, and the inference speed is increased by 6.76 to 16.14 times. The experiment further confirms that both single core and mixed rock datasets, acquired through the CPDD scheme, exhibit excellent generalization and reliability.