High-quality Data Research Articles

Artificial Intelligence in Achieving Sustainable Development: Expectations of Undergraduate Students

AbstractWhile there has been ample discussion regarding Artificial Intelligence (AI)’s contributions and challenges on the development agenda at the policy level, little is known about how students translate the potential and barriers of AI in achieving Sustainable Development Goals (SDGs). Drawing upon various qualitative data, including class observation, focus group interviews, and learning activity outcomes generated by 240 students across 7 different majors, this case study explores the expected roles of AI as well as barriers to AI adoption for sustainable development perceived by undergraduate students. The study revealed that students anticipated AI to play diverse roles, including data analyst, a bridge to connect people and resources, and a barrier breaker. On the other hand, students addressed multiple barriers to AI adoption for sustainable development including access to high-quality data and lack of a goal-oriented AI model and skills in AI development and use. These findings suggest how AI can be conceptualized and positioned as a development intervention as well as offer implications on AI-driven interventions for SDGs.

Oxygen optodes on oceanographic moorings: recommendations for deployment and in situ calibration

Increasing interest in the deployment of optical oxygen sensors, or optodes, on oceanographic moorings reflects the value of dissolved oxygen (DO) measurements in studies of physical and biogeochemical processes. Optodes are well-suited for moored applications but require careful, multi-step calibrations in the field to ensure data accuracy. Without a standardized set of protocols, this can be an obstacle for science teams lacking expertise in optode data processing and calibration. Here, we provide a set of recommendations for the deployment and in situ calibration of data from moored optodes, developed from our experience working with a set of 60 optodes deployed as part of the Gases in the Overturning and Horizontal circulation of the Subpolar North Atlantic Program (GOHSNAP). In particular, we detail the correction of drift in moored optodes, which occurs in two forms: (i) an irreversible, time-dependent drift that occurs during both optode storage and deployment and (ii) a reversible and pressure-and-time-dependent drift that is detectable in some optodes deployed at depths greater than 1,000 m. The latter is virtually unidentified in the literature yet appears to cause a low-bias in measured DO on the order of 1 to 3 µmol kg−1 per 1,000 m of depth, appearing as an exponential decay over the first days to months of deployment. Comparisons of our calibrated DO time series against serendipitous mid-deployment conductivity-temperature-depth (CTD)-DO profiles, as well as biogeochemical (BGC)-ARGO float profiles, suggest the protocols described here yield an accuracy in optode-DO of ∼1%, or approximately 2.5 to 3 µmol kg−1. We intend this paper to serve as both documentation of the current best practices in the deployment of moored optodes as well as a guide for science teams seeking to collect high-quality moored oxygen data, regardless of expertise.

Constructing and exploring neuroimaging projects: a survey from clinical practice to scientific research.

Over the past decades, numerous large-scale neuroimaging projects that involved the collection and release of multimodal data have been conducted globally. Distinguished initiatives such as the Human Connectome Project, UK Biobank, and Alzheimer's Disease Neuroimaging Initiative, among others, stand as remarkable international collaborations that have significantly advanced our understanding of the brain. With the advancement of big data technology, changes in healthcare models, and continuous development in biomedical research, various types of large-scale projects are being established and promoted worldwide. For project leaders, there is a need to refer to common principles in project construction and management. Users must also adhere strictly to rules and guidelines, ensuring data safety and privacy protection. Organizations must maintain data integrity, protect individual privacy, and foster stakeholders' trust. Regular updates to legislation and policies are necessary to keep pace with evolving technologies and emerging data-related challenges. CRITICAL RELEVANCE STATEMENT: By reviewing global large-scale neuroimaging projects, we have summarized the standards and norms for establishing and utilizing their data, and provided suggestions and opinions on some ethical issues, aiming to promote higher-quality neuroimaging data development. KEY POINTS: Global neuroimaging projects are increasingly advancing but still face challenges. Constructing and utilizing neuroimaging projects should follow set rules and guidelines. Effective data management and governance should be developed to support neuroimaging projects.

Ultra-long-period Cepheids as Standard Candles from Gaia to Rubin-LSST

An analysis of the properties of ultra-long-period Cepheids (ULPs) could significantly contribute to understanding the Hubble constant tension, e.g., the current discrepancy between determinations based on local distance indicators and those relying on cosmic microwave background measurements. These highly luminous variables are observable beyond 100 Mpc, so if they were confirmed to behave as standard candles, they would allow us a direct measurement of cosmological distances without any secondary distance indicator, thus reducing potential systematic errors in the calibration of the cosmic distance scale. This paper presents an analysis of the largest known sample of 73 ULPs, including 15 objects in nearby galaxies, with new accurate and homogeneous photometry obtained by Gaia DR3, and a new object, in our Galaxy, identified as a long-period variable in Gaia DR3, but recently reclassified as a ULP. The obtained results suggest that, by improving photometric accuracy, the ULP period–Wesenheit relation shows a smaller dispersion than that obtained in literature and is in better agreement with the classical Cepheid (CC) one, supporting the hypothesis that ULPs are the extension of the CCs at higher period, mass, and luminosity. However, to reach this aim, it is necessary to enrich the sample with high-quality data. The Rubin Observatory Legacy Survey of Space and Time (Rubin-LSST) survey offers the possibility to achieve this thanks to its photometric characteristics and time extension. In particular, we will explore the capabilities of the Rubin-LSST survey to recover ULP theoretical light curves by using a new tool called PulsationStarRecovery, built by our group for this type of analysis.

Estimating the vertical profile of water quality variables in reservoirs: Application of remotely sensed data and machine learning techniques.

Water quality assessment and management of reservoirs depend on accurate, large-scale, and continuous monitoring of the vertical profile of Water Quality Variables (WQVs). Remote sensing data have been widely used to retrieve high spatiotemporal water quality data; however, their application has practically been limited to evaluating surface WQVs. In this paper, a novel and efficient approach is introduced for assessing the profile of WQVs in reservoirs that depend on stratification, by taking into account the shape of profile as prior knowledge. First, an appropriate function is fitted to the WQVs vertical profile, and second, the parameters of that function as representative of the WQVs vertical profile are estimated using machine learning techniques. The model's inputs are day, maximum depth of point, and remote sensing data. Finally, PAWN sensitivity analysis is applied to show the extent to which each input influences different parts of the vertical profile. This method is applied in the Wadi Dayqah Reservoir, the largest dam in Oman, to evaluate water temperature, dissolved oxygen, pH, and chlorophyll-a profile. The results show that the predicted profiles are properly representative of in situ measurements, with a mean absolute error of 0.28 °C, 0.25 mg/L, 0.052, and 0.33 μg/L on test data sets of water temperature, dissolved oxygen, pH, and chlorophyll-a, respectively. Finally, PAWN sensitivity analysis illustrates that satellite data not only influence the parameters representing surface WQVs but also contribute to the estimation of other curve parameters.

High-quality sika deer omics data and integrative analysis reveal genic and cellular regulation of antler regeneration.

Antler is the only organ that can fully regenerate annually in mammals. However, the regulatory pattern and mechanism of gene expression and cell differentiation during this process remain largely unknown. Here, we obtain comprehensive assembly and gene annotation of the sika deer (Cervus nippon) genome. Together with large-scale chromatin accessibility and gene expression data, we construct gene regulatory networks involved in antler regeneration, identifying four transcription factors, MYC, KLF4, NFE2L2, and JDP2 with high regulatory activity across whole regeneration process. Comparative studies and luciferase reporter assay suggest the MYC expression driven by a cervid-specific regulatory element might be important for antler regenerative ability. We further develop a model called cTOP which integrates single-cell data with bulk regulatory networks and find PRDM1, FOSL1, BACH1, and NFATC1 as potential pivotal factors in antler stem cell activation and osteogenic differentiation. Additionally, we uncover interactions within and between cell programs and pathways during the regeneration process. These findings provide insights into the gene and cell regulatory mechanisms of antler regeneration, particularly in stem cell activation and differentiation.

Genomic Surveillance of Foodborne Pathogens: Advances and Obstacles

Context: The genomic surveillance of foodborne pathogens in the United States has grown exponentially in the past decade, grounded in a powerful combination of novel sequencing technologies, bioinformatic approaches, data-sharing networks, and metadata harmonization efforts. This practice report examines recent advances in genomic epidemiology as applied to food safety programs and delineates State, Tribal, Local, and Territorial infrastructure necessary for continued life-saving improvements in public health. Program: National databases of foodborne pathogen genomes, along with data sharing and evaluation networks such as GenomeTrakr and PulseNet, have transformed how connections are made among isolates and how root causes of outbreaks are determined, allowing much more timely interventions to protect public health. Freely available bioinformatics tools such as GalaxyTrakr and the National Center for Biotechnology Information Pathogen Detection database have allowed laboratories with limited local computing resources to participate in surveillance efforts and contribute to traceback investigations. Implementation: In this report, we describe advances in genomic epidemiology that have occurred over the past decade and examine obstacles to fully implementing this technology within State, Tribal, Local, and Territorial public health systems. Evaluation: Despite a clear return on investment from governmental expenditures on genomic surveillance of foodborne pathogens, we identify significant obstacles to further sustained progress. These obstacles include workforce gaps, ineffective data sharing, and lack of constitutive and sustained funding. Discussion: Many public health laboratories face major obstacles to widespread and routine adoption of genomic surveillance technologies. While whole genome sequencing has become an integral part of routine public health microbiology, the seamless integration of these protocols into the existing practices, laboratory workflows, and information systems remains challenging. Centralized efforts to address these issues include (1) support through the Food and Drug Administration Laboratory Flexible Funding Model, (2) training and proficiency assessments, (3) open-source, standardized protocols for collecting high-quality genomic data, and (4) open access informatics software.

An historical overview of artificial intelligence for diagnosis of major depressive disorder

The Artificial Intelligence (AI) technology holds immense potential in the realm of automated diagnosis for Major Depressive Disorder (MDD), yet it is not without potential shortcomings. This paper systematically reviews the research progresses of integrating AI technology with depression diagnosis and provides a comprehensive analysis of existing research findings. In this context, we observe that the knowledge-driven first-generation of depression diagnosis methods could only address deterministic issues in structured information, with the selection of depression-related features directly influencing identification outcomes. The data-driven second-generation of depression diagnosis methods achieved automatic learning of features but required substantial high-quality clinical data, and the results were often obtained solely from the black-box models which lack sufficient explainability. In an effort to overcome the limitations of the preceding approaches, the third-generation of depression diagnosis methods combined the strengths of knowledge-driven and data-driven approaches. Through the fusion of information, the diagnostic accuracy is greatly enhanced, but the interpretability remains relatively weak. In order to enhance interpretability and introduce diagnostic criteria, this paper offers a new approach using Large Language Models (LLMs) as AI agents for assisting the depression diagnosis. Finally, we also discuss the potential advantages and challenges associated with this approach. This newly proposed innovative approach has the potential to offer new perspectives and solutions in the diagnosis of depression.

Small-rotative fixed-target serial synchrotron crystallography (SR-FT-SSX) for molecular crystals.

The increasing availability of ultrabright Light Sources is facilitating the study of smaller crystals at faster timescales but with an increased risk of severe X-ray damage, leading to developments in multi-crystal methods such as serial crystallography (SX). SX studies on crystals with small unit cells are challenging as very few reflections are recorded in a single data image, making it difficult to determine the orientation matrix for each crystal and thus preventing the combination of the data from all crystals for structure solution. We herein present a Small-Rotative Fixed-Target Serial Synchrotron Crystallography (SR-FT-SSX) methodology, in which rotation of the serial target through a small diffraction angle at each crystal delivers high-quality data, facilitating ab initio unit cell determination and atomic-scale structure solution. The method is benchmarked using microcrystals of the small-molecule photoswitch sodium nitroprusside dihydrate, obtaining complete data to dmin = 0.6 Å by combining just 66 partial datasets selected against rigorous quality criteria.

Are rates of clinical interventions during pregnancy and childbirth different for refugees and asylum seekers in high-income countries? A scoping review.

Adequate maternity care and appropriate clinical interventions during labour and delivery can reduce adverse perinatal outcomes, but unnecessary interventions may cause harm. While studies have shown that refugees and asylum seekers face important barriers when accessing maternity care, there is a lack of high-quality quantitative data on perinatal health interventions, such as induction of labour or caesarean sections, among refugees and asylum seekers and the findings reported in the literature tend to be inconsistent. Our goal was to examine and synthesise the evidence regarding the rates of intrapartum clinical interventions in women who are refugees and asylum seekers in high-income countries compared to other population groups. We conducted a scoping review of peer-reviewed studies published in English since 2011 that report original quantitative findings regarding intrapartum clinical interventions among refugees and asylum seekers in high-income countries compared to those in non refugee, non asylum seeker populations. We examined reported rates of clinical pain relief, labour induction and augmentation, episiotomies, instrumental deliveries, and caesarean sections. Twenty-five papers were included in the review. Findings indicate that refugees and asylum seekers were less likely to receive pain relief, with 16 out of 20 data points showing unadjusted ORs ranging from 0.20 (CI: 0.10-0.60) to 0.96 (CI: 0.70-1.32). Similarly, findings indicate lower odds of instrumental delivery among refugees and asylum seekers with 14 of 21 data points showing unadjusted ORs between 0.25 (CI: 0.15-0.39) and 0.78 (CI: 0.47-1.30); the remaining papers report no statistically significant difference between groups. There was no discernable trend in rates of labour induction and episiotomies across studies. The studies included in this review suggest that asylum seekers and refugees are less likely to receive clinical pain relief and experience instrumental delivery than non-refugee groups in high-income countries. This review strengthens our understanding of the links between immigration status and maternity care, ultimately informing policy and practice to improve perinatal health and the provision of care for all.

Optimization of Whole-Genome Resequencing Depth for High-Throughput SNP Genotyping in Litopenaeus vannamei

The application of whole-genome resequencing in genetic research is rapidly expanding, yet the impact of sequencing depth on data quality and variant detection remains unclear, particularly in aquaculture species. This study re-sequenced 31 Litopenaeus vannamei (L. vannamei) samples at over 28× sequencing depth using the Illumina NovaSeq system and down-sampled the data to simulate depths from 0.5× to 20×. Results showed that when the sequencing depth was below 10×, the number of SNP identifications increased sharply with the rise in depth, with single nucleotide polymorphisms (SNPs) detected at 10× accounting for approximately 69.16% of those detected at 20×. The genotyping accuracy followed a similar trend to SNP detection results, being approximately 0.90 at 6×. Further analyses showed that the main cause of genotyping errors was the misidentification of heterozygous variants as homozygous variants. Therefore, considering both the quantity and quality of SNPs, a sequencing depth of 10× is recommended for whole-genome studies and genetic mapping, while a depth of 6× is more cost-effective for population structure analysis. This study underscores the importance of selecting optimal sequencing depth to ensure reliable variant detection and high data quality, providing valuable guidance for whole-genome resequencing in shrimp and other aquatic species.

Application of Nexus Learning framework for Personalized Medicine and Drug Development

Biological systems provide an enormous amount of knowledge on development and disease. The pharmaceutical industry is facing new possibilities and challenges with the rise of high-throughput methods to study disease and biology. The aim is to develop drugs based on acceptable therapeutic assumptions. Nexus Learning provides a variety of tools that help improve discovery and decision making using extensive, high-quality data and well-specified queries. Potential applications of NLF can be found across the whole drug development process. Clinical study target validation, prognostic biomarker discovery, and digital pathology data processing are a few examples. There has been a wide variety of applications in terms of technique and environment, with some methods producing reliable forecasts and insights. The primary problem with using NLF is that the findings provided by NL tend to be easy to understand or reproduce, which can limit their value. The collection of complete and methodical high-dimensional data remains an essential need across all domains. By addressing these challenges and increasing understanding about what is required to verify NLF methodologies, we can employ NLF to make data-driven decisions, which can speed up drug discovery and development and minimize failure rates.

Blockchain-Based Intelligent Inventory Management for Hospital Supply Chain (HSC) with Deep Learning Techniques

A medication shortage may have far-reaching consequences, making it an important inventory management concern for healthcare organizations. Intelligent demand forecasting is essential to make cold chain businesses more competitive and save transportation and storage costs. Hence, this paper proposes a blockchain-based intelligent inventory management model (BC-IIMM) with deep learning techniques for hospitals to prevent running out of supplies, run their systems sustainably, and optimize their profits in the hospital supply chain (HSC).Preventing the contamination of pharmaceuticals requires a pharmaceutical cold chain monitoring system that uses IoT devices, blockchain, and cloud computing to track items across their entire supply chain reliably. One other perk is that both doctors and patients may benefit from drug recommendations made by a deep belief network (DBN) model. Further, to construct deep learning (DL), the proposed cold chain monitoring system produces large-scale, high-quality data. Holding cold chain items would become less expensive and cold chain inventory management decisions would be easier with it.Next, they provide a numerical result proving the suggested method works better than the baselines regarding refilling cost and shortage rate.

Numerical simulation and validation of heavy rainfall flood inundation in small watersheds in undocumented mountainous areas

ABSTRACT With global warming and the increase in extreme precipitation events, floods are becoming more frequent in mountainous areas, and the safety of lives and property of people is seriously threatened. However, understanding of the flooding process in uninformative mountainous areas is limited due to the lack of high-quality hydrometeorological data. Hence, this study adopts the MIKE21 model to simulate flood inundation in the Shadai River basin in the Qilian Mountain region of the northern Tibetan Plateau as an example. This validates the model-simulated flow and inundation extent using the flow data obtained from the calculation of the flood trace points, extent of inundation, and high-resolution remote sensing images. The results show that the flash flood inundation mainly occurs at 12:00–01:00 AM on 18 August 2022, and the simulated and actual maximum inundation areas are 7.9 and 9.5 km2, respectively. The fitted F-statistic value is 0.81, and the relative error between the calculated flow rate of the flood trace point and the model-simulated flow rate is 8%, indicating good consistency. Furthermore, an in-depth exploration of the model parameter sensitivity reveals that the use of distributed Manning's roughness coefficient value has higher simulation accuracy.

TMIC-60. BRATS-PATH: ASSESSING HETEROGENEOUS HISTOPATHOLOGIC REGIONS IN GLIOBLASTOMA

Abstract Glioblastoma, the most common malignant primary adult brain tumor, poses significant diagnostic and treatment challenges due to its heterogeneous molecular and micro-environmental profiles. To this end, we organize the BraTS-Path challenge to provide a public benchmarking environment and a comprehensive dataset to develop and validate AI models for identifying distinct histopathologic glioblastoma sub-regions in H&E-stained digitized tissue sections. We identified 188 multi-institutional diagnostic slides of glioblastoma (IDH-wt, Gr.4) cases, from the TCGA-GBM and TCGA-LGG data collections, following their reclassification according to the 2021 WHO classification criteria. Sub-regions were selected according to distinctive morphology of histopathologic features and included aggressive tumor biology and areas consistent with potential treatment effect. Selected sub-region annotations included cellular tumor, geographic necrosis, cortical infiltration, pseudopalisading necrosis, microvascular proliferation, white matter penetration, regions dense with macrophages, leptomeningeal infiltration, and presence of lymphocytes. We obtained 107,340 patches of size 512x512 from the 9 sub-regions. A global network of board-certified expert neuropathologists defined and followed a systematic annotation protocol based on clinical definitions and only delineated sub-regions with high confidence, thus ensuring high-quality standardized data. Each tissue section was assigned to an annotator-approver pair, with the annotator delineating sub-regions and the approver ensuring the consistency of the annotations. By crowdsourcing annotations, the BraTS-Path challenge harnesses the collective expertise of clinical neuropathologists and fosters a collaborative environment to advance the neuro-oncology field. The anticipated developed algorithms are expected to integrate state-of-the-art computational methods, achieving high accuracy in identifying diverse histopathologic features and advancing clinical decision-making processes. The BraTS-Path challenge aims to bridge the gap between research and clinical practice by promoting the development of AI-driven tools for precise tumor characterization. This collaborative effort can significantly enhance our understanding of glioblastoma, improve diagnostic accuracy, and inform treatment strategies, thereby contributing to better patient outcomes.

POTENCIAL TRANSFORMADOR DA INTELIGÊNCIA ARTIFICIAL ​​NA LOGÍSTICA

Artificial intelligence (AI) has become a fundamental tool for many sectors of society, and the transport and logistics sector is no exception. This study aims to review the use of Artificial Intelligence (AI) in the transport and logistics sector, focusing on logistics control towers, routing and telemetry. This is a literature review of articles taken from databases such as Scielo and Google Scholar, published in the last 5 years (2018-2023) that were in the public domain and could be read in full. The results showed that AI has the potential to revolutionize industrial sectors, with applications ranging from route optimization to demand forecasting and efficient warehouse management. However, implementing AI also presents challenges, including the need for large volumes of high-quality data, investment in infrastructure and training.

Theory Development and Crime Prevention Insights From the NCVS

The National Crime Victimization Survey (NCVS) has furnished U.S. crime statistics and high quality data for theory development and policy insights for over 50 years. This article highlights few of these achievements drawing on the development of criminological thinking with regard to measurement, and explanation of victimization and its concentration on a small minority of the population. It overviews lifestyle and the more encompassing environmental criminology theory for victimization, and explanations of multiple victimization, and the international crime drop that relied on NCVS data. Selected crime prevention insight and few recommendations for future research conclude the article.

Pseudo Replay-based Class Continual Learning for Online New Category Anomaly Detection in Advanced Manufacturing

The incorporation of advanced sensors and machine learning techniques has enabled modern manufacturing enterprises to perform data-driven classification-based anomaly detection based on the sensor data collected in manufacturing processes. However, one critical challenge is that newly presented defect category may manifest as the manufacturing process continues, resulting in monitoring performance deterioration of previously trained machine learning models. Hence, there is an increasing need for empowering machine learning models to learn continually. Among all continual learning methods, memory-based continual learning has the best performance but faces the constraints of data storage capacity. To address this issue, this paper develops a novel pseudo replay-based continual learning framework by integrating class incremental learning and oversampling-based data generation. Without storing all the data, the developed framework could generate high-quality data representing previous classes to train machine learning model incrementally when new category anomaly occurs. In addition, it could even enhance the monitoring performance since it also effectively improves the data quality. The effectiveness of the proposed framework is validated in three cases studies, which leverages supervised classification problem for anomaly detection. The experimental results show that the developed method is very promising in detecting novel anomaly while maintaining a good performance on the previous task and brings up more flexibility in model architecture.

Comparative Analysis of Machine Learning, Decision Trees, and K-Nearest Neighbors for Heart Disease Prediction

Abstract. This study investigates the efficacy of Machine Learning (ML), Decision Trees, and K-Nearest Neighbors (KNN) techniques in predicting heart disease, aiming to identify their strengths and limitations. ML models are effective in detecting complex patterns and delivering evolving predictions from large datasets but require high-quality data and can be challenging to interpret. Decision Trees provide clear, understandable decision rules, which is beneficial in clinical settings, yet they are susceptible to overfitting and instability. KNN, valued for its simplicity and flexibility, classifies heart disease based on similarity but struggles with high computational costs and sensitivity to noisy data. Experimental results indicate that each model has distinct advantages: ML excels in pattern recognition, Decision Trees offer interpretability, and KNN handles diverse data effectively. However, each also faces challenges that impact performance, such as data quality issues for ML, overfitting for Decision Trees, and computational demands for KNN. The study suggests that balancing these strengths and weaknesses is crucial for optimizing heart disease prediction models. Future research should explore hybrid approaches that combine these models' advantages while addressing their respective limitations to improve predictive accuracy and practical application in real-world scenarios.

Diving into hot topics of salivary gland carcinoma management—an EORTC young and early career investigator survey

IntroductionRecently, the ASCO and ESMO guidelines on salivary gland carcinomas (SGCs) have been released. However, several crucial points lack strong recommendations due to low or intermediate quality of evidence. To better address these “grey zones” in the guidelines, we conducted a survey among the European Organization for Research and Treatment of Cancer (EORTC) Head and Neck Cancer Group (HNCG) members on behalf of the EORTC young and early career investigators.Materials and methodsThe survey included 29 questions addressing diagnostic and therapeutic issues related to SGC patients and was shared among 539 members of the EORTC HNCG. Responses were collected from December 2022 to March 2023. The primary aim was to evaluate the decision-making criteria guiding physicians in areas with low evidence in SGC guidelines.ResultsWith a response rate of 19%, the survey received input from 102 respondents, mainly medical oncologists (45%). Among those with pathological high-risk features, 35% and 30% of respondents opted for chemoradiotherapy (CRT) in the definitive and adjuvant settings, respectively. For patients with R0 resection of highly aggressive SGC at the pT1–2 stage, 37% proposed a close follow-up, while 38% proposed adjuvant lymph-node field RT. In cases of pT3–4 stage, 48% proposed adjuvant lymph-node field RT in all cases, and 44% proposed it based only on risk factors. The most important factors guiding the decision to give adjuvant RT after salvage surgery for locoregional relapse include previous radiotherapy, margin status, and the presence of extranodal extension. When faced with combined positivity to HER2 and AR, responses regarding the choice of first-line palliative treatment were heterogeneous.ConclusionsInternational guidelines lack strong recommendations in several fields of SGC diagnosis and treatment due to insufficient high-quality data, resulting in heterogeneity in physicians’ treatment decision (e.g., adjuvant lymph-node field RT and their low concordance with guidelines, such as the use of concomitant CRT). The survey demonstrated the need for prospective clinical trial data to address these gaps in the future.