Inaccessible Data Research Articles

Algorithms for Verifying the Correctness of Interval Expert Evaluations

Expert probability is used for decision-making, allows you to assess risks, and is especially useful in conditions of limited and inaccessible objective data. At the same time, there may be contradictions between the estimates, which may complicate analysis and decision-making. This paper describes an algorithm that checks the compliance of expert estimates using probability intervals. One of the sections is devoted to the development of an evaluation method for a new formula based on the original evaluation system. The article describes the user interface that provides interaction with the developed algorithms.

SFDA-T: A novel source-free domain adaptation method with strong generalization ability for fault diagnosis

Currently, source free domain adaptation (SFDA) methods are employed to address the issue of inaccessible source domain data (SDD) in transfer learning. However, existing SFDA methods often suffer from overfitting to specific domains, leading to poor generalization ability in the target domain. To address these challenges, this paper proposes a novel SFDA method named SFDA-T for fault diagnosis. Specifically, a Transformer-CNN-based feature extractor is constructed, to mine the transferable feature knowledge of faults in the SDD. The approach reduces the overfitting of the model to domain-specific information and improves model’s generalization ability. In addition, the feature attention loss is designed to calculate attention weights of the sample features to increase the model’s attention to the crucial feature regions in the target domain. A source similarity guided exponential loss is developed to guide target samples based on the decision boundaries of the source domain, facilitating cluster alignment of target sample categories and expanding distances between different categories. Furthermore, a self-training pseudo-labeling constraint is employed to reduce the effect of incorrect label matching and further constrain the model. The results of the experiments on gearboxes and bearings indicate that the proposed method achieves high fault diagnosis accuracy while effectively decoupling from SDD.

Diagnostic Accuracy of Artificial Intelligence-Based Systems for Detecting Diabetic Retinopathy: A Systematic Review

Diabetic retinopathy (DR) represents a leading cause of blindness worldwide, early detection is critical to prevent vision loss. However, traditional screening methods, which rely on human experts, prove to be costly and time-consuming. The systematic review aims to assess the validity of artificial intelligence (AI) as a screening tool for detecting DR among diabetic patients. A systematic literature search was performed of the following databases: PubMed, Scopus, CINAHL, and Web of Science. The last date of our search was January 31, 2024. We included all observational studies, including cohort, case–control and cross-sectional studies and evaluated their quality using the Joanna Briggs Institute tool. We included diagnostic test accuracy studies evaluating the use of AI algorithms for DR screening in patients with diabetes. Studies were excluded if they exclusively assessed diagnostic accuracy for DR that did not use AI algorithms as a diagnostic tool and studies with incomplete or inaccessible data. Thirteen studies with sample sizes ranging from 69 to 1378 participants, reported good sensitivity of AI for detecting visually threatening DR (VTDR). The lowest sensitivity was 89.2%, and the highest was 100%. In terms of specificity, Any DR exhibited higher specificity compared to RDR and VTDR, ranging from 80.2% to 100%. The sensitivity and specificity of the Artificial Intelligence (AI)-based tools available for DR screening was considered acceptable, especially in detecting VTDR and Any DR, was regarded as good. These results implied the potential usefulness of these tools for DR screening in settings with limited resources. However, further high-quality comparative studies were deemed necessary to evaluate their effectiveness in real-world clinical settings.

Guidelines on the selection and inventory of social life cycle assessment indicators: a case study on flexible plastic packaging in the European circular economy

Abstract Purpose Emerging technologies are addressing current challenges to shift from a linear to a circular economy. However, the consideration of social aspects in this context is limited, and the prioritization of indicators appears arbitrary in the absence of clear and robust selection criteria. Following social life cycle assessment (S-LCA) principles, the aim of this paper is to guide the selection and inventory of social indicators based on a case study on evolving flexible packaging within the European circular economy. Methods To achieve the objective, the study involves several steps, including conducting a systematic literature review to identify key stakeholders, impact subcategories, indicators, and inventories for circular flexible plastic packaging. Multi-criteria decision analysis (MCDA) is applied to preselect a set of indicators, followed by stakeholder engagement in prioritizing impact indicators through participatory methods. Subsequently, a data collection procedure was established. Results and discussion This paper presents a materiality ranking for 19 social indicators tailored to the emerging circular flexible packaging. The result is a prioritization of nine of these indicators, including Existence of record of proof of age, Percentage of workers who are paid a living wage or above, and Existence of certified environmental management system. These nine indicators form the basis for the collection of respective inventory data for an entry-level assessment. Furthermore, guidelines on 11 procedural steps were formulated based on these findings. Moreover, challenges of synonymity and inconsistency in S-LCA terminology, besides inaccessible inventory data especially in the context of evolving recycling technologies, are identified as substantial barriers in the effort to streamline and perform S-LCA. Conclusions The paper concludes that prioritizing indicators is essential for entry-level and prospective assessments, especially when time and data are limited. Additionally, using primary inventory data to evaluate the foreground system and its direct impacts on stakeholders promotes accountability and corporate social responsibility. Furthermore, a set of procedural steps, from defining the system boundary to preparing the data questionnaire, serves as a valuable resource for practitioners across diverse industries. Recommendations In future research, these nine prioritized indicators can be applied to assess the social performance of various case studies along the value chain of circular flexible plastic packaging, covering food and non-food applications. In addition, the methodological guidelines for selecting and prioritizing indicators can be replicated across multiple industry sectors. Moreover, research effort should be directed towards establishing a methodological framework for prospective S-LCA in the context of emerging technologies.

Affine Collaborative Normalization: A shortcut for adaptation in medical image analysis

The paradigm of “pretraining-then-finetuning” (PT-FT) has been extensively explored to enhance the performance of clinical applications with limited annotations. A major impediment to applying such workflow to various medical imaging modalities is the lack of parameter-free approaches boosting the transferability against notable domain shifts. The success of normalization techniques in domain adaptation provides a promising solution, while inaccessible source data in finetuning (FT) poses new challenges. In this paper, we revisit the Batch Normalization module in PT-FT and propose a unified framework for both fast transferability estimation and transferability-aware finetuning. We discovered that the vanilla FT suffers from the issue that feature patterns of corresponding channels could be misaligned across domains. Hence, we present an Affine Collaborative Normalization (AC-Norm) to dynamically recalibrate the channels in the target model according to the cross-domain channel-wise correlations without any source data and extra parameters. AC-Norm provides very competitive results compared to state-of-the-art FT methods in six classification/segmentation tasks. We also demonstrate the capability of AC-Norm in estimating the transferability of pretrained models in only single-step backpropagation. Our code is available at https://github.com/EndoluminalSurgicalVision-IMR/ACNorm.

Predicting the Remaining Useful Life of Lithium-Ion Batteries Using 10 Random Data Points and a Flexible Parallel Neural Network

Accurate Remaining Useful Life (RUL) prediction of lithium batteries is crucial for enhancing their performance and extending their lifespan. Existing studies focus on continuous or relatively sparse datasets; however, continuous and complete datasets are rarely available in practical applications due to missing or inaccessible data. This study attempts to achieve the prediction of lithium battery RUL using random sparse data from only 10 data points, aligning more closely with practical industrial scenarios. Furthermore, we introduce the application of a Flexible Parallel Neural Network (FPNN) for the first time in predicting the RUL of lithium batteries. By combining these two approaches, our tests on the MIT dataset show that by randomly downsampling 10 points per cycle from 10 cycles, we can reconstruct new meaningful features and achieve a Mean Absolute Percentage Error (MAPE) of 2.36% in predicting the RUL. When the input data are limited to the first 10 cycles using the dataset constructed from random downsampling and the FPNN, the predicted RUL MAPE is 0.75%. The method proposed in this study offers an accurate, adaptable, and comprehensible new solution for predicting the RUL of lithium batteries, paving a new research path in the field of battery health monitoring.

Health Data Visualization Literacy Skills of Young Adults with Down Syndrome and the Barriers to Inference-making

As health management becomes more intertwined with data, an individual’s ability to read, interpret, and engage with personal health information in data visualizations is increasingly critical to one’s quality of care. People with Down Syndrome already experience greater health disparities than their typically developing peers. Inaccessible health information and technologies have the potential to magnify inequities further. Inaccessible health data can be an additional barrier to people with Down Syndrome’s ability to adopt and use health systems or devices, make informed decisions about their bodies, and advocate for themselves in health contexts. By examining their underlying data visualization literacy skills, our exploratory study involving ten young adults with Down Syndrome identifies several design opportunities to improve the accessibility of health data visualizations (HDVs) by addressing the cascade of negative effects caused by inference-making barriers in HDVs.

Deep Learning-Based Psoriasis Assessment: Harnessing Clinical Trial Imaging for Accurate Psoriasis Area Severity Index Prediction

Introduction: Image-based machine learning holds great promise for facilitating clinical care; however, the datasets often used for model training differ from the interventional clinical trial-based findings frequently used to inform treatment guidelines. Here, we draw on longitudinal imaging of psoriasis patients undergoing treatment in the Ultima 2 clinical trial (NCT02684357), including 2,700 body images with psoriasis area severity index (PASI) annotations by uniformly trained dermatologists. Methods: An image-processing workflow integrating clinical photos of multiple body regions into one model pipeline was developed, which we refer to as the “One-Step PASI” framework due to its simultaneous body detection, lesion detection, and lesion severity classification. Group-stratified cross-validation was performed with 145 deep convolutional neural network models combined in an ensemble learning architecture. Results: The highest-performing model demonstrated a mean absolute error of 3.3, Lin’s concordance correlation coefficient of 0.86, and Pearson correlation coefficient of 0.90 across a wide range of PASI scores comprising disease classifications of clear skin, mild, and moderate-to-severe disease. Within-person, time-series analysis of model performance demonstrated that PASI predictions closely tracked the trajectory of physician scores from severe to clear skin without systematically over- or underestimating PASI scores or percent changes from baseline. Conclusion: This study demonstrates the potential of image processing and deep learning to translate otherwise inaccessible clinical trial data into accurate, extensible machine learning models to assess therapeutic efficacy.

Data Mesh Meets Blockchain

Effective dataset management is crucial for enterprises to make informed decisions and remain competitive. However, centralized dataset management approaches often result in poor scalability, unclear governance, inaccessible data silos, and duplication of efforts. This paper proposes a distributed blockchain-based framework inspired by the data mesh architecture to address these challenges. Our proposed framework leverages blockchain’s decentralized nature to enable efficient and transparent dataset sharing across enterprise business domains. By turning datasets into digital assets and business domains into peers, our framework utilizes blockchain smart contracts to allow business domains to view, request, and share datasets. In this paper, we describe the details of our framework, and we analyze it from scalability, accessibility, security, and data governance perspectives. To validate our framework, we provide a proof-of-concept implementation with a publicly available source code.

Regulation of neuroinflammation in Alzheimer's disease via nanoparticle-loaded phytocompounds with anti-inflammatory and autophagy-inducing properties.

Alzheimer's disease (AD) is characterized by neuroinflammation linked to amyloid β (Aβ) aggregation and phosphorylated tau (τ) protein in neurofibrillary tangles (NFTs). Key elements in Aβ production and NFT assembly, like γ-secretase and p38 mitogen-activated protein kinase (p38MAPK), contribute to neuroinflammation. In addition, impaired proteosomal and autophagic pathways increase Aβ and τ aggregation, leading to neuronal damage. Conventional neuroinflammation drugs have limitations due to unidirectional therapeutic approaches and challenges in crossing the Blood-Brain Barrier (BBB). Clinical trials for non-steroidal anti-inflammatory drugs (NSAIDs) and other therapeutics remain uncertain. Novel strategies addressing the complex pathogenesis and BBB translocation are needed to effectively tackle AD-related neuroinflammation. The current scenario demands for a much-sophisticated theranostic measures which could be achieved via customized engineering and designing of novel nanotherapeutics. As, these therapeutics functions as a double edge sword, having the efficiency of unambiguous targeting, multiple drug delivery and ability to cross BBB proficiently. Inclusion criteria involve selecting recent, English-language studies from the past decade (2013-2023) that explore the regulation of neuroinflammation in neuroinflammation, Alzheimer's disease, amyloid β, tau protein, nanoparticles, autophagy, and phytocompounds. Various study types, including clinical trials, experiments, and reviews, were considered. Exclusion criteria comprised non-relevant publication types, studies unrelated to Alzheimer's disease or phytocompounds, those with methodological flaws, duplicates, and studies with inaccessible data. In this study, polymeric nanoparticles loaded with specific phytocompounds and coated with an antibody targeting the transferrin receptor (anti-TfR) present on BBB. Thereafter, the engineered nanoparticles with the ability to efficiently traverse the BBB and interact with target molecules within the brain, could induce autophagy, a cellular process crucial for neuronal health, and exhibit potent anti-inflammatory effects. Henceforth, the proposed combination of desired phytocompounds, polymeric nanoparticles, and anti-TfR coating presents a promising approach for targeted drug delivery to the brain, with potential implications in neuroinflammatory conditions such as Alzheimer's disease.

Mimicking Clinical Trials with Synthetic Acute Myeloid Leukemia Patients Using Generative Artificial Intelligence

Data sharing is often hindered by concerns of patient privacy, regulatory aspects, and proprietary interests thereby impeding scientific progress and establishing a gatekeeping mechanism in clinical medicine since obtaining large data sets is costly and time-consuming. We employed two different generative artificial intelligence (AI) technologies: CTAB-GAN+ and Normalizing Flows (NFlow) to synthesize clinical trial data based on pooled patient data from four previous multicenter clinical trials of the German Study Alliance Leukemia (AML96, AML2003, AML60+, SORAML) that enrolled adult patients (n=1606) with acute myeloid leukemia (AML) who received intensive induction therapy. As a generative adversarial network (GAN), CTAB-GAN+ consists of two adversarial networks: a generator producing synthetic samples from random noise and a discriminator aiming to distinguish between real and synthetic samples. The model converges as the discriminator can no longer reliably differentiate between real or synthetic data. Contrastingly, NFlow consists of a sequence of invertible transformations (flows) starting from a simple base distribution and gradually adding complexity to better mirror the training data. Both models were trained on tabular data including demographic, laboratory, molecular genetic and cytogenetic patient variables. Detection of molecular alterations in the original cohort was performed via next-generation sequencing (NGS) using the TruSight Myeloid Sequencing Panel (Illumina, San Diego, CA, USA) with a 5% variant-allele frequency (VAF) mutation calling cut-off. For cytogenetics, standard techniques for chromosome banding and fluorescence-in-situ-hybridization (FISH) were used. Hyperparameter tuning of generative models was conducted using the Optuna Framework. For each model, we used a total of 70 optimization trials to optimize a custom score inspired by TabSynDex which assesses both the resemblance of the synthetic data to real training data and its utility. Pairwise analyses were conducted between the original and both synthetic data sets, respectively. All tests were carried out as two-sided tests using a significance level α of 0.05. Table 1 summarizes baseline patient characteristics and outcome for both synthetic cohorts compared to the original cohort. Firstly, we found both models to adequately represent patient features, albeit that some individual variables showed a statistically significant deviation from the original cohort. It is important to note that for such a large sample size (n=1606 for each cohort), even miniscule differences can be rendered statistically significant notwithstanding any meaningful clinical difference. Interestingly, variables that deviated from the original distribution were different for both models indicating model architecture to play a vital role in sample representation: While CTAB-GAN+ showed significant deviations for both age and sex, NFlow showed significant deviations for AML status. Complete remission rate was similar between original (70.7%, odds ratio [OR]: 2.41) and CTAB-GAN+ (73.7%, OR: 2.81, p=0.059) and NFlow (69.1%, OR: 2.24, p=0.356). For event-free survival (EFS), which was not included as a target in hyperparameter tuning, both networks deviated significantly from the original cohort (original: median 7.2 months, HR: 1.36; CTAB-GAN+: median 12.8 months, HR 0.74, p<0.001; NFlow: median 9.0 months, HR: 0.87, p=0.001). Overall survival (OS) was well represented by NFlow compared to the original cohort, while CTAB-GAN+ showed a significant deviation (original: median 17.5 months, HR: 1.14; CTAB-GAN+: median 19.5 months, HR 0.88, p<0.001; NFlow: median 16.2 months, HR: 1.00, p=0.055). Both models showed an adequate graph representation in Kaplan-Meier analysis (Figure 1). Here, we demonstrate using two different generative AI technologies that synthetic data generation provides an attractive solution to circumvent issues in current standards of data collection and sharing. It effectively allows for bypassing logistical, organizational, and financial burdens, as well as regulatory and ethical concerns. Ultimately, this enables explorative research inquiries into previously inaccessible data sets and offers the prospect of fully synthetic control arms in prospective clinical trials.

Modelling and interpreting evacuation time and exit choice for large-scale ancient architectural complex using machine learning

The intricate spatial structures of large-scale ancient architectural complexes, combined with their complex crowd compositions, hinder evacuation efficiency. However, research on the evacuation of such complexes is limited. Moreover, among the existing evacuation studies, most solely rely on either simulation models or machine learning techniques. Few integrate the strengths of simulation modeling in capturing otherwise inaccessible evacuation data with the advantages of machine learning in addressing complex problems. This study aims to utilize random forest algorithm, based on Pathfinder simulation evacuation data, to model the evacuation time and exit choice of occupants within such complexes. Additionally, we employ tools to interpret the random forest models, analyzing the influence of various factors on evacuation time and exit choice. Specifically, this methodology was applied to examine evacuation time and exit choice of 4800 occupants, comprising both tourists and residents, within the Xidi ancient architecture complex that spans nearly 130,000 square meters. The reliability of the random forest evacuation models is validated using real experimental data. The results indicate that the impact of environmental factors on evacuation time is often nonlinear. Compared to environmental factors, individual factors tend to have a lesser influence on evacuation time. For exit choice, the influence of environmental factors varies in significance at different scales. Individual factors have a notably higher influence on exit choice than on evacuation time. We also discern that evacuation distance affects the degree to which other factors influence exit choice. Overall, this research contributes to a deeper understanding of evacuation time and exit choice in such complexes, offering decision-makers more comprehensive evacuation strategies.

К СЕМИДЕСЯТИЛЕТИЮ СО ДНЯ РОЖДЕНИЯ Н. А. РИМСКОГО-КОРСАКОВА

August 23, 2023 marked the 70th anniversary of the birth of Nikolai Andreevich RimskyKorsakov. Nikolai Andreevich worked at the Institute of Oceanology for almost half a century and went from an engineer to deputy director for scientific work in the field of Marine Engineering, head of the Bottom Hydrolocation Laboratory, chief researcher, professor, doctor of technical sciences. He is a participant in numerous expeditions to various areas of the World Ocean and inland waters of Russia on ships of various classes and different departmental affiliations. Under his leadership, many different projects, government contracts were completed, numerous technical means were developed and built for the multifaceted study of water areas, the use of which made it possible to obtain a large amount of previously inaccessible data on various underwater objects of natural and artificial origin, including objects posing a radiation threat.

A volume of fluid method for three dimensional direct numerical simulations of immiscible droplet collisions

This paper presents an advanced Volume of Fluid (VOF) method that enables performant three dimensional Direct Numerical Simulations (DNS) of the interaction of two immiscible fluids in a gaseous environment with large topology changes, e.g., binary droplet collisions. One of the challenges associated with the introduction of a third immiscible phase into the VOF method is the reconstruction of the phase boundaries near the triple line in arbitrary arrangements. For this purpose, an efficient method based on a Piecewise Linear Interface Calculation (PLIC) is shown. Moreover, the surface force modeling with the robust Continuous Surface Stress (CSS) model was enhanced to treat such three-phase situations with large topology changes and thin films. A consistent scaling of the fluid properties at the interfaces ensures energy conservation.The implementation of these methods in the multi-phase flow solver Free Surface 3D (FS3D) allowed a successful validation. A qualitative comparison of the morphology in binary collisions of immiscible droplets as well as a quantitative comparison regarding the threshold velocities that distinguish different collision regimes shows excellent agreement with experimental results.These simulations enable the evaluation of experimentally inaccessible data like the contributions of kinetic, surface and dissipative energy of both immiscible liquids during the collision process. Furthermore, the comparison against binary collisions of the same liquids highlights similarities and differences between immiscible and equal droplet collisions. Both can support the modeling of the immiscible liquid interaction in the future.

Development and Testing of a Data Capture Device for Use With Clinical Incentive Spirometers: Testing and Usability Study.

The incentive spirometer is a basic and common medical device from which electronic health care data cannot be directly collected. As a result, despite numerous studies investigating clinical use, there remains little consensus on optimal device use and sparse evidence supporting its intended benefits such as prevention of postoperative respiratory complications. The aim of the study is to develop and test an add-on hardware device for data capture of the incentive spirometer. An add-on device was designed, built, and tested using reflective optical sensors to identify the real-time location of the volume piston and flow bobbin of a common incentive spirometer. Investigators manually tested sensor level accuracies and triggering range calibrations using a digital flowmeter. A valid breath classification algorithm was created and tested to determine valid from invalid breath attempts. To assess real-time use, a video game was developed using the incentive spirometer and add-on device as a controller using the Apple iPad. In user testing, sensor locations were captured at an accuracy of 99% (SD 1.4%) for volume and 100% accuracy for flow. Median and average volumes were within 7.5% (SD 6%) of target volume sensor levels, and maximum sensor triggering values seldom exceeded intended sensor levels, showing a good correlation to placement on 2 similar but distinct incentive spirometer designs. The breath classification algorithm displayed a 100% sensitivity and a 99% specificity on user testing, and the device operated as a video game controller in real time without noticeable interference or delay. An effective and reusable add-on device for the incentive spirometer was created to allow the collection of previously inaccessible incentive spirometer data and demonstrate Internet-of-Things use on a common hospital device. This design showed high sensor accuracies and the ability to use data in real-time applications, showing promise in the ability to capture currently inaccessible clinical data. Further use of this device could facilitate improved research into the incentive spirometer to improve adoption, incentivize adherence, and investigate the clinical effectiveness to help guide clinical care.

Shortfalls in tracking data available to inform North American migratory bird conservation

Animal tracking has become an effective way to identify where and when migratory species encounter threats throughout their annual cycle. Yet, untracked or poorly tracked species and undiscoverable or inaccessible data for the species that have been tracked mean that gaps in the knowledge of where and when species occur are still an issue for conservation. These gaps in knowledge of species movements have been termed the “movement shortfall”. Here, we quantify the movement shortfall for North American migratory birds by comprehensively reviewing full annual cycle tracking data and identifying gaps and biases in how, where, and what species are tracked with electronic tracking devices. We found 30 species for which tracking is not feasible given body size constraints, no data for 291 trackable species, and restricted or reduced data accessibility for an additional 59 species. Thus, despite the ability to track most species, the movement shortfall remains a constraint to informing conservation strategies for 56 % of North American migratory bird species. The number increases to 65 % when considering species with restricted or reduced data accessibility, further limiting access to this information. Moreover, 23 % of the tracking data stems from low precision tracking technologies reducing the implementation and effectiveness to conservation actions. A lack of species and population data hinders conservation and biases management decisions, ultimately making inefficient use of conservation resources. We encourage researchers to consider these gaps in their decisions about future tracking efforts, conservation management, and data archiving practices.

The Influence of Ocean Processes on Fine-Scale Changes in the Yellow Sea Cold Water Mass Boundary Area Structure Based on Acoustic Observations

The boundary of Yellow Sea Cold Water Mass (YSCWM) is a key ocean frontal structure influencing the regional ecosystem. Complex oceanic processes such as tidal currents, upwelling, and internal waves influence fine-scale hydrological structures, comprehensively resulting in a significantly highly productive area for plankton and fisheries. However, detailed research requires inaccessible high-resolution data. To investigate the fine-scale and high-frequency effects of oceanic processes on the local hydrological and ecological environment, we conducted comprehensive cruise acoustic observations and intensive station surveys of the hydrological environment around the YSCWM boundary in summer 2021 and 2022, and found that: (1) fine-scale hydrological structures across the YSCWM boundary were directly captured through this specific intensive station observation design; (2) clear zooplankton diel vertical migration (DVM) phenomena match well with the thermocline variation, showing that acoustics are effective indicators that reflect the water mass layering structure in summer in the YS; and (3) the shear excited by internal waves during propagation and flood tides enhances the upward and downward mixing of the water mass near the thermocline, thus thickening and weakening the layer, an effect that will be more pronounced when both are present at the same time, with ebb tide having the opposite effect. Topographically influenced tidal upwelling also causes significant vertical fluctuations in isotherms. This represents a new way of studying the fine-scale hydrodynamic–hydrologic–ecological aspects of key regions through acoustic remote sensing.

Reliability and Efficiency of the CAPRI-3 Metastatic Prostate Cancer Registry Driven by Artificial Intelligence.

Manual data collection is still the gold standard for disease-specific patient registries. However, CAPRI-3 uses text mining (an artificial intelligence (AI) technology) for patient identification and data collection. The aim of this study is to demonstrate the reliability and efficiency of this AI-driven approach. CAPRI-3 is an observational retrospective multicenter cohort registry on metastatic prostate cancer. We tested the patient-identification algorithm and automated data extraction through manual validation of the same patients in two pilots in 2019 and 2022. Pilot one identified 2030 patients and pilot two 9464 patients. The negative predictive value of the algorithm was maximized to prevent false exclusions and reached 94.8%. The completeness and accuracy of the automated data extraction were 92.3% or higher, except for date fields and inaccessible data (images/pdf) (10-88.9%). Additional manual quality control took over 3 h less time per patient than the original fully manual CAPRI registry (105 vs. 300 min). The CAPRI-3 patient-identification algorithm is a sound replacement for excluding ineligible candidates. The AI-driven data extraction is largely accurate and complete, but manual quality control is needed for less reliable and inaccessible data. Overall, the AI-driven approach of the CAPRI-3 registry is reliable and timesaving.

Deep data mining reveals variable abundance and distribution of microbial reproductive manipulators within and among diverse host species.

Bacterial symbionts that manipulate the reproduction of their hosts are important factors in invertebrate ecology and evolution, and are being leveraged for host biological control. Infection prevalence restricts which biological control strategies are possible and is thought to be strongly influenced by the density of symbiont infection within hosts, termed titer. Current methods to estimate infection prevalence and symbiont titers are low-throughput, biased towards sampling infected species, and rarely measure titer. Here we develop a data mining approach to estimate symbiont infection frequencies within host species and titers within host tissues. We applied this approach to screen ~32,000 publicly available sequence samples from the most common symbiont host taxa, discovering 2,083 arthropod and 119 nematode infected samples. From these data, we estimated that Wolbachia infects approximately 44% of all arthropod and 34% of all nematode species, while other reproductive manipulators only infect 1-8% of arthropod and nematode species. Although relative titers within hosts were highly variable within and between arthropod species, a combination of arthropod host species and Wolbachia strain explained approximately 36% of variation in Wolbachia titer across the dataset. To explore potential mechanisms for host control of symbiont titer, we leveraged population genomic data from the model system Drosophila melanogaster. In this host, we found a number of SNPs associated with titer in candidate genes potentially relevant to host interactions with Wolbachia. Our study demonstrates that data mining is a powerful tool to detect bacterial infections and quantify infection intensities, thus opening an array of previously inaccessible data for further analysis in host-symbiont evolution.

Protecting and restoring habitats to benefit freshwater biodiversity

Freshwater biodiversity is under great threat across the globe as evidenced by more severe declines relative to other types of ecosystems. Some of the main stressors responsible for these concerning trends is habitat fragmentation, degradation, and loss stemming from anthropogenic activities, including energy production, urbanization, agriculture, and resource extraction. Habitat protection and restoration both play an integral role in efforts to save freshwater biodiversity and associated ecosystem services from further decline. In this paper, we summarize the sources of threats associated with habitat fragmentation, degradation, and loss and then outline response options to protect and restore freshwater habitats. Specific response options are to legislate the protection of healthy and productive freshwater ecosystems, prioritize habitats for protection and restoration, enact durable protections, conserve habitat in a coordinated and integrated manner, engage in evidence-based restoration using an adaptive management approach, ensure that potential freshwater habitat alterations are mitigated or off-set, and future-proof protection and restoration actions. Such work should be done through a lens that engages and involves local community members. We identify three broad categories of obstacles that could arise during the implementation of the response options outlined: (a) scientific (e.g., inaccessible data or uncertainties), (b) institutional and management (e.g., capacity issues or differing goals across agencies), and (c) social and political (e.g., prioritizing economic development over conservation initiatives). The protection and restoration of habitats is key to Bend the Curve for freshwater biodiversity, with a comprehensive, connected, and coordinated effort of response options needed to protect intact habitats and restore fragmented, degraded, and lost habitats and the biodiversity and ecosystem services that they support.