Data-driven Tool Research Articles

The LTAR Cropland Common Experiment at Lower Chesapeake Bay.

The Lower Chesapeake Bay (LCB) Long-Term Agroecosystem Research (LTAR) Common Experiment (CE) located in Beltsville, MD, focuses on research of concern to producers of the major regional crops, which are corn (Zea mays L.), soybean [Glycine max (L.) Merr.], wheat (Triticum aestivum L.), and various forage species. Livestock production in the region includes broiler and laying chickens (Gallus gallus domesticus L.) and dairy and beef cattle (Bos taurus L.). The LCB region is among the most heavily populated in the United States. Urban development pressure is high for both farms and natural areas. The need to restore Chesapeake Bay water quality is a major influence on regional agricultural practices. Conservation practices such as cover cropping, no-till agriculture, and nutrient management planning are more common in the region compared to nationally. However, farmers still face management challenges implementing practices that address water quality and the rise of herbicide-resistant weeds. Researchers at the LCB site recognize the need to protect the Chesapeake and Delaware Bays and maintain farmer profitability. The LCB CE compares a 3-year crop rotation system featuring alternative crop management (cover crop intensification, crop rotation diversification, and integrated weed management [IWM]) with a prevailing 2-year system (no cover crops and no IWM), both under continuous no-tillage, to identify the optimal balance to promote the sustainability of regional cropping systems. The LTAR LCB site provides data-driven tools and solutions to support farmers in the mid-Atlantic region.

Color and Spatial Frequency Provide Functional Signatures of Retinotopic Visual Areas.

Primate vision relies on retinotopically organized cortical parcels defined by representations of hemifield (upper versus lower visual field), eccentricity (fovea versus periphery), and area (V1, V2, V3, V4). Here we test for functional signatures of these organizing principles. We used fMRI to measure responses to gratings varying in spatial frequency, color, and saturation across retinotopically defined parcels in two macaque monkeys, and we developed a Sparse Supervised Embedding (SSE) analysis to identify stimulus features that best distinguish cortical parcels from each other. Constraining the SSE model to distinguish just eccentricity representations of the voxels revealed the expected variation of spatial frequency and S-cone modulation with eccentricity. Constraining the model according to the dorsal-ventral location and retinotopic area of each voxel provided unexpected functional signatures, which we investigated further with standard univariate analyses. Posterior parcels (V1) were distinguished from anterior parcels (V4) by differential responses to chromatic and luminance contrast, especially of low spatial frequency gratings. Meanwhile, ventral parcels were distinguished from dorsal parcels by differential responses to chromatic and luminance contrast, especially of colors that modulate all three cone types. The dorsal-ventral asymmetry not only resembled differences between candidate dorsal and ventral subdivisions of human V4, but also extended to include all retinotopic visual areas, starting in V1 and increasing from V1 to V4. The results provide insight into the functional roles of different retinotopic areas and demonstrate the utility of Sparse Supervised Embedding as a data-driven tool for generating hypotheses about cortical function and behavior.Significance Statement This study demonstrates a new analysis, Sparse Supervised Embedding (SSE), which promises to be useful for visualizing and understanding complex neuroimaging datasets. The paper uses SSE to explore the functional roles of retinotopic visual areas (V1, V2, V3, V4, V3a, MT). The results show that retinotopic areas parcellated by representations for eccentricity and upper/lower visual hemifield have functional signatures, which are defined by unique combinations of responses to color, spatial frequency, and contrast. The functional signatures provide hypotheses for the different roles that the parcels play in vision and help resolve apparent differences between human and macaque visual cortex organization.

Implementing AI-Driven Bed Sensors: Perspectives from Interdisciplinary Teams in Geriatric Care

Sleep is a crucial aspect of geriatric assessment for hospitalized older adults, and implementing AI-driven technology for sleep monitoring can significantly enhance the rehabilitation process. Sleepsense, an AI-driven sleep-tracking device, provides real-time data and insights, enabling healthcare professionals to tailor interventions and improve sleep quality. This study explores the perspectives of an interdisciplinary hospital team on implementing Sleepsense in geriatric hospital care. Using the interpretive description approach, we conducted focus groups with physicians, nurses, care aides, and an activity worker. The Consolidated Framework for Implementation Research (CFIR) informed our thematic analysis to identify barriers and facilitators to implementation. Among 27 healthcare staff, predominantly female (88.89%) and Asian (74.1%) and mostly aged 30–50 years, themes emerged that Sleepsense is perceived as a timesaving and data-driven tool that enhances patient monitoring and assessment. However, barriers such as resistance to change and concerns about trusting the device for patient comfort and safety were noted, while facilitators included training and staff engagement. The CFIR framework proved useful for analyzing implementation barriers and facilitators, suggesting future research should prioritize effective strategies for interdisciplinary team support to enhance innovation adoption and patient outcomes in rehabilitation settings.

Predicting Disease Outbreaks using Machine Learning Models on Public Health Data

Timely prediction of disease outbreaks is crucial for effective public health responses. This study explores the use of machine learning models to forecast outbreaks by analyzing public health data, including epidemiological, demographic, and environmental factors. The goal is to create a real-time prediction framework to aid health authorities in early interventions. Models such as time-series forecasting, Random Forest, and Long Short-Term Memory (LSTM) networks are applied to predict diseases like influenza, dengue, and COVID-19. These models are evaluated using accuracy, precision and recall. External factors like weather, population mobility, and public sentiment are also examined for their role in disease spread. The results highlight key factors driving outbreaks and demonstrate how machine learning can enhance public health surveillance by providing early warning systems. This research contributes to the development of scalable, data-driven tools for outbreak prediction, supporting proactive public health strategies and minimizing future impacts.

Leveraging Machine Learning for Sophisticated Rental Value Predictions: A Case Study from Munich, Germany

There has been very limited research conducted to predict rental prices in the German real estate market using an AI-based approach. From a general perspective, conventional approaches struggle to handle large amounts of data and fail to consider the numerous elements that affect rental prices. The absence of sophisticated, data-driven analytical tools further complicates this situation, impeding stakeholders, such as tenants, landlords, real estate agents, and the government, from obtaining the accurate insights necessary for making well-informed decisions in this area. This paper applies novel machine learning (ML) approaches, including ensemble techniques, neural networks, linear regression (LR), and tree-based algorithms, specifically designed for forecasting rental prices in Munich. To ensure accuracy and reliability, the performance of these models is evaluated using the R2 score and root mean squared error (RMSE). The study provides two feature sets for model comparison, selected by particle swarm optimisation (PSO) and CatBoost. These two feature selection methods identify significant variables based on different mechanisms, such as seeking the optimal solution with an objective function and converting categorical features into target statistics (TSs) to address high-dimensional issues. These methods are ideal for this German dataset, which contains 49 features. Testing the performance of 10 ML algorithms on two sets helps validate the robustness and efficacy of the AI-based approach utilising the PyTorch framework. The findings illustrate that ML models combined with PyTorch-based neural networks (PNNs) demonstrate high accuracy compared to standalone ML models, regardless of feature changes. The improved performance indicates that utilising the PyTorch framework for predictive tasks is advantageous, as evidenced by a statistical significance test in terms of both R2 and RMSE (p-values < 0.001). The integration results display outstanding accuracy, averaging 90% across both feature sets. Particularly, the XGB model, which exhibited the lowest performance among all models in both sets, significantly improved from 0.8903 to 0.9097 in set 1 and from 0.8717 to 0.9022 in set 2 after being combined with the PNN. These results showcase the efficacy of using the PyTorch framework, enhancing the precision and reliability of the ML models in predicting the dynamic real estate market. Given that this study applies two feature sets and demonstrates consistent performance across sets with varying characteristics, the methodology may be applied to other locations. By offering accurate projections, it aids investors, renters, property managers, and regulators in facilitating better decision-making in the real estate sector.

Integrated Workflow Development for Data-Driven Neighborhood-Scale Building Performance Simulation

Abstract As urbanization intensifies, cities are key contributors to energy consumption and carbon emissions, accounting for a significant portion of global energy use and CO2 emissions. This paper introduces a systematic approach to support the development of urban projects with minimized operational carbon footprints through the integration of data-driven building performance simulation (BPS) tools in early-stage design. Emphasizing the necessity for a collaborative effort among designers, policymakers, and other stakeholders, we discuss the evolution of BPS toward incorporating data-driven tools for energy need reduction and informed decision-making. Despite the proliferation of modeling methods and data-related challenges, we present a theoretical workflow, supported by interactions with design firms in the US and European Union (EU) through interviews. This structured approach, demonstrating adaptability and scalability across urban contexts, foregrounds the potential for future data-driven integration in design practices. Grounded in theoretical concepts and preliminary real-world insights, our work emphasizes the transformation of standard activities toward data-driven processes, showcasing the crucial role of practical experience in advancing sustainable, low-carbon urban development.

High-dimensional Biomarker Identification for Scalable and Interpretable Disease Prediction via Machine Learning Models.

Omics data generated from high-throughput technologies and clinical features jointly impact many complex human diseases. Identifying key biomarkers and clinical risk factors is essential for understanding disease mechanisms and advancing early disease diagnosis and precision medicine. However, the high-dimensionality and intricate associations between disease outcomes and omics profiles present significant analytical challenges. To address these, we propose an ensemble data-driven biomarker identification tool, Hybrid Feature Screening (HFS), to construct a candidate feature set for downstream advanced machine learning models. The pre-screened candidate features from HFS are further refined using a computationally efficient permutation-based feature importance test, forming the comprehensive High-dimensional Feature Importance Test (HiFIT) framework. Through extensive numerical simulations and real-world applications, we demonstrate HiFIT's superior performance in both outcome prediction and feature importance identification. An R package implementing HiFIT is available on GitHub ( https://github.com/BZou-lab/HiFIT ).

Selecting the Best Compositions of a Wheelchair Basketball Team: A Data-Driven Approach

Wheelchair basketball, regulated by the International Wheelchair Basketball Federation, is a sport designed for individuals with physical disabilities. This article presents a data-driven tool that effectively determines optimal team lineups based on past performance data and metrics for player effectiveness. Our proposed methodology involves combining a Bayesian longitudinal model with an integer linear problem to optimize the lineup of a wheelchair basketball team. To illustrate our approach, we use real data from a team competing in the Rollstuhlbasketball Bundesliga, namely the Doneck Dolphins Trier. We consider three distinct performance metrics for each player and incorporate uncertainty from the posterior predictive distribution of the longitudinal model into the optimization process. The results demonstrate the tool’s ability to select the most suitable team compositions and to calculate posterior probabilities of several players playing together in the best composition.

A Collection of Components to Design Clinical Dashboards Incorporating Patient-Reported Outcome Measures: Qualitative Study.

A clinical dashboard is a data-driven clinical decision support tool visualizing multiple key performance indicators in a single report while minimizing time and effort for data gathering. Studies have shown that including patient-reported outcome measures (PROMs) in clinical dashboards supports the clinician's understanding of how treatments impact patients' health status, helps identify changes in health-related quality of life at an early stage, and strengthens patient-physician communication. This study aims to determine design components for clinical dashboards incorporating PROMs to inform software producers and users (ie, physicians). We conducted interviews with software producers and users to test preselected design components. Furthermore, the interviews allowed us to derive additional components that are not outlined in existing literature. Finally, we used inductive and deductive coding to derive a guide on which design components need to be considered when building a clinical dashboard incorporating PROMs. A total of 25 design components were identified, of which 16 were already surfaced during the literature search. Furthermore, 9 additional components were derived inductively during our interviews. The design components are clustered in a generic dashboard, PROM-related, adjacent information, and requirements for adoption components. Both software producers and users agreed on the primary purpose of a clinical dashboard incorporating PROMs to enhance patient communication in outpatient settings. Dashboard benefits include enhanced data visualization and improved workflow efficiency, while interoperability and data collection were named as adoption challenges. Consistency in dashboard design components is preferred across different episodes of care, with adaptations only for disease-specific PROMs. Clinical dashboards have the potential to facilitate informed treatment decisions if certain design components are followed. This study establishes a comprehensive framework of design components to guide the development of effective clinical dashboards incorporating PROMs in health care practice.

A Review on Deep Learning Approaches and Optimization Techniques For Political Security Threat Prediction

In an era of complex geopolitical dynamics and evolving security threats, the accurate prediction and proactive management of political security risks are imperative. This article provides a comprehensive review of the application of deep learning methodologies and optimization techniques to enhance political security threat prediction. Beginning with analyzing the dynamic landscape of political security threats, the paper emphasizes the necessity for adaptive, data-driven predictive tools. It then delves into the fundamentals of deep learning, elucidating core principles, notable architectural frameworks, and their diverse applications across domains. Expanding upon this foundation, the study evaluates the suitability of deep learning models for addressing the multifaceted challenges associated with political security threat prediction. To maximize the utility of these models, the article explores optimization techniques encompassing hyperparameter tuning, transfer learning, and ensemble strategies, assessing their effectiveness in fine-tuning predictions and bolstering the resilience of threat prediction systems. This review involved the utilization of four journal databases: IEEE, Science Direct, Association for Computing Machinery (ACM), and SpringerLink. We analyzed and examined 39 articles, paying close attention to the different patterns and techniques found within the chosen research framework. Through a critical synthesis of existing research, this review offers insights into the strengths, limitations, and future directions of deep learning-based political security threat prediction, contributing to the ongoing discourse on leveraging artificial intelligence for safeguarding global stability and security.

A Data-driven Race Strategy Tool for Olympic Sailing Competitions

_ Venue-specific training has, over the years, proven to be a key asset for sailing teams performing at major championships and at the Olympics. A comprehensive understanding of the environmental features and possible weather scenarios in an outdoor sport like sailing can provide athletes with a significant strategic advantage. At the same time, GPS tracking is becoming a readily available technology that athletes use both in training and during races to analyse their own and their competitor's performance. This work couples environmental and meteorological data with GPS tracks for Olympic sailing classes, linking weather features with strategic decisions on the race track. We propose a greedy algorithm to search for an optimal route based on weather forecasts to present the best strategy prior to the race. Our results show the potential of this approach to provide valuable decision support for athletes in Olympic sailing competitions, demonstrated for the 470 Olympic class. Keywords Olympic sailing; path planning; data-driven methods; meteorology; A* algorithm

A portable application framework for energy management and information systems (EMIS) solutions using Brick semantic schema

This paper introduces a portable framework for developing, scaling and maintaining energy management and information systems (EMIS) applications using an ontology-based approach. Key contributions include an interoperable layer based on Brick schema, the formalization of application constraints pertaining metadata and data requirements, and a field demonstration. The framework allows for querying metadata models, fetching data, preprocessing, and analyzing data, thereby offering a modular and flexible workflow for application development. Its effectiveness is demonstrated through a case study involving the development and implementation of a data-driven anomaly detection tool for the photovoltaic systems installed at the Politecnico di Torino, Italy. During eight months of testing, the framework was used to tackle practical challenges including: (i) developing a machine learning-based anomaly detection pipeline, (ii) replacing data-driven models during operation, (iii) optimizing model deployment and retraining, (iv) handling critical changes in variable naming conventions and sensor availability (v) extending the pipeline from one system to additional ones.

412 Awardee Talk: Implementing student-centered teaching and learning strategies in undergraduate animal science

Abstract Student-centered teaching and learning strategies enhance active engagement, foster critical thinking, and promote the practical application of knowledge gained from undergraduate animal science courses. The scholarship of teaching and learning (SOTL) has evolved within the animal science community, with a recent focus on adopting evidence-based teaching activities in college classrooms. We discuss documented SOTL approaches utilized in our program for the development of student-centered teaching and learning. In addition to traditional internships and student research projects, capstone projects utilizing a design thinking approach to solving animal industry problems have proven significantly effective. This problem-based learning program enables students to develop industry networks, understand problems within the industry, and develop creative ways to address the problems. Non-conventional and innovative designs of course assignments can foster student interest and enhance learning experiences. For example, by incorporating engaging and novel resources such as animal industry webinars and podcasts as classroom materials, instructors can reinforce course content and captivate students’ attention to the topic. Specifically, students have appreciated the opportunity to learn from experts not available locally at the university. Developing and utilizing real-world case studies exposes students to farm practices and related decision-making. Involving students in competitive team activities like the Dairy Challenge and Academic Quadrathlon contests develops marketable skills for students while training them to implement the knowledge acquired in the classrooms on a commercial farm. Field trips, short courses, and summer schools provide hands-on training opportunities, enhancing students’ exposure to the industry and opportunities after graduation. Students have identified that animal science topics are relevant beyond the classroom walls in everyday life including grocery store and kitchen, as demonstrated by a significant increase in enrollment of non-animal science major students in the animal science courses. The student-centered programs like Cross-training students in data-driven decision support tools help to develop strong day-one competency of graduates through training and networking within and outside the discipline. With the advent of new technologies and tools, training programs surrounding data management and interpretation should integrate systems knowledge across plant science, animal science, and computer science disciplines. These interdisciplinary and cross-disciplinary training activities promote the exchange of ideas while developing systems-level understanding. Furthermore, initiatives like Collaborative Online International Learning (COIL) provide students with exposure to diverse thoughts and situations, broadening their horizons and perspectives. For students not able to participate in study abroad and international programs due to time and financial constraints, COIL delivers cost-effective virtual opportunities to develop global understanding and cultural competency. By incorporating these student-centered teaching and learning strategies, undergraduate animal science programs can effectively prepare students for real-world challenges for the next generation of agriculture leaders.

Understanding Data-Driven Analytic Decision Making on Air Quality Monitoring an Empirical Study

Air quality monitoring is increasingly relying on data-driven analytic decision-making tools to provide accurate and timely information, forming the background of this study. The objective is to understand the factors influencing the adoption and usage behavior of these tools using the Unified Theory of Acceptance and Use of Technology (UTAUT2) model. The method involves incorporating UTAUT2 constructs Performance Expectancy (PE), Effort Expectancy (EE), Social Influence (SI), Facilitating Conditions (FC), Price Value (PV), Hedonic Motivation (HM), and Habit (H), alongside external variables such as Considered Risk (CR) and Considered Trust (CT). Data from 287 respondents were analyzed to assess their impact on Behavior Intention (BI) and Usage Behavior (UB). The results demonstrate that both trust and risk considerations significantly affect user behavior, underscoring the need to address these factors to enhance the adoption of air quality monitoring systems. In conclusion, this research provides valuable insights for developers and policymakers on improving the implementation and acceptance of data-driven technologies in environmental monitoring, thereby contributing to more effective air quality management.

Data-driven models for the prediction of H2 generation through chemical reaction and performance evaluation of on-board hydrogen fuelled sintering tube furnace

Hydrogen is emerging as a sustainable fuel for the future. In this present work, data-driven modelling tools viz. Radial Basis Function Neural Network (RBFNN) and Least Square Fit (LSF) method, are employed to determine the rate of production of H2 gas by the chemical reaction between aluminium and water in the presence of aq. NaOH. Reactions are carried out at NaOH concentrations of 1M–5M, water temperatures 303K–333K. Hydrogen gas obtained at 333K/4M is found to have a yield of ⁓88 % of the theoretical yield. The activation energy of the reaction is found to be 57.62 kJ mol−1. The fitted models are validated with experimental results for two unknown conditions. The correlation coefficient obtained for the RBFNN model is 0.999, which indicates the high reliability of the model. On-board production of H2 gas by the chemical reaction was used for heating a sintering tube furnace. The hot products of the combustion of hydrogen and air at various fuel-air ratios were used to heat the sintering tube furnace. The maximum thermal efficiency obtained for the furnace is 76.22 % at a fuel-air ratio of 1:60 corresponds to a fuel-air equivalent ratio (λ) of 0.57.

Recent Development Trends in Plant Protection UAVs: A Journey from Conventional Practices to Cutting-Edge Technologies—A Comprehensive Review

Uncrewed aerial vehicles (UAVs) for plant protection play a vital role in modern agricultural operations. In recent years, advancements in UAVs and pest control technologies have significantly enhanced operational efficiency. These innovations have addressed historical challenges in agricultural practices by improving automation and precision in managing insect pests, diseases, and weeds. UAVs offer high operational efficiency, wide adaptability to different terrain, and safe applications. The development and demand for these technologies have increased to boost agricultural production. In agricultural settings where conventional machinery struggles to carry out farming operations, UAVs have transformed farming practices by providing high operational efficiency and significant profitability. The integration of UAVs and other smart technologies has driven advancements. The UAV sector has received substantial attention as a convergence of production, service, and delivery, introducing synergy through the presence of several developing areas. The market for this technology is expected to grow in the future. In this comprehensive review, we analyzed an overview of historical research, diverse techniques, the transition from conventional to advanced application, development trends, and operational milestones across diverse cropping systems. We also discussed adoption and subsidy policies. In order to properly understand UAV operational efficiency, we also analyzed and discussed smart atomization systems, spray drift, droplet deposition detection technologies, and the capabilities of related technologies. Additionally, we reviewed the role of software programs, data-driven tools, biodegradable materials, payloads, batteries, sensing technologies, weather, and operational and spraying factors. Regulatory limitations, operating and farmer’s training, economic effects, and guidelines were also acknowledged in this review. This review highlights deficiencies and provides essential knowledge of the use of UAVs for agriculture tasks in different regions. Finally, we examine the urgency of UAV technology implementations in the agricultural sector. In conclusion, we summarize the integration of UAVs and their related technologies with applications and future research prospects, offering directions for follow-up research on the key technologies of UAVs and encouraging the enhancement of agricultural production management in terms of efficiency, accuracy, and sustainability.

Scorecard Approach to Eliminate Onchocerciasis in Venezuela.

In the Americas, onchocerciasis has been eliminated in 11 of 13 endemic foci by mass administration of ivermectin. The remaining at-risk population resides in a contiguous cross-border transmission zone located in the Amazon jungle in northwest Brazil and southern Venezuela, known as the Yanomami Focus Area. Here, we describe the development and implementation of a data-driven tool, called the Scorecard Approach (SCA), for the 393 communities that comprise the Venezuela South Focus. The SCA was first applied in 2018 and is reassessed on an annual basis. This operational strategy seeks to prioritize communities with low ivermectin coverage while taking into account the nature and variation of other epidemiological and logistical variables. Numeric scores are assigned for each factor and added together to yield a composite score for each community that is categorized as high, medium, or low priority. In this way, the SCA serves as a valuable and comprehensive strategy for planning, monitoring, and maximizing programmatic efficiency. In addition, it has allowed the country to face the main challenges of this endemic area: its remoteness, its large areas of territory to cover, the semi-nomadic nature of the Yanomami people, and their continuous cross-border movements. For 2022, the SCA categorized 54 (13.7%), 108 (27.5%), and 231 (58.8%) communities as high, medium, and low priority, respectively. The results presented here show that prioritizing communities at risk and with greatest needs increases the feasibility of interrupting the transmission of onchocerciasis by 2025 in the last endemic focus in the Americas.

Perspectives on Advancing Multimodal Learning in Environmental Science and Engineering Studies.

The environment faces increasing anthropogenic impacts, resulting in a rapid increase in environmental issues that undermine the natural capital essential for human wellbeing. These issues are complex and often influenced by various factors represented by data with different modalities. While machine learning (ML) provides data-driven tools for addressing the environmental issues, the current ML models in environmental science and engineering (ES&E) often neglect the utilization of multimodal data. With the advancement in deep learning, multimodal learning (MML) holds promise for comprehensive descriptions of the environmental issues by harnessing data from diverse modalities. This advancement has the potential to significantly elevate the accuracy and robustness of prediction models in ES&E studies, providing enhanced solutions for various environmental modeling tasks. This perspective summarizes MML methodologies and proposes potential applications of MML models in ES&E studies, including environmental quality assessment, prediction of chemical hazards, and optimization of pollution control techniques. Additionally, we discuss the challenges associated with implementing MML in ES&E and propose future research directions in this domain.

Three-dimensional grid-free sound source localization method based on deep learning

Sound source localization (SSL) technology is a popular method for identifying the locations of noise sources, which serves as a prerequisite for noise control. Deep learning, as a data-driven tool, shows broad perspectives in the field of SSL with its powerful nonlinear fitting ability. The existing deep learning-based SSL methods only provide a two-dimensional (2D) representation of the sound source location and cannot obtain the specific coordinates of the sound source in three-dimensional (3D) space. Although traditional beamforming methods can be directly generalized to 3D scenes in principle, they suffer from the limitations of insufficient vertical resolution and high computational cost. Therefore, a 3D grid-free SSL method (3DGF) informed by deep learning is suggested in this study to enhance the accuracy and computational efficiency of 3D localization. First, the number of data channels is compressed to respect limited memory resources during the training process. Subsequently, a dense convolutional neural network (DenseNet) model is utilized to obtain the 3D spatial coordinates of the sound source using the processed 3D beamforming map as input. Since the coordinates are continuous and are not constrained by the grid of the beamforming map, the grid-free strategy presents more accurate localization results. Then, the effects of the volume of training data and the compression ratio are analyzed, respectively, in simulation, and the localization performance with different signal-to-noise ratios (SNRs) is also tested. Finally, by comparing 3DGF with DAMAS, both simulation and experimental results demonstrate that 3DGF improves the accuracy and efficacy of 3D localization. Meanwhile, its satisfactory generalization ability and robustness against noise highlight its potential for practical applications.

Deep learning for automatic calcium detection in echocardiography

Cardiovascular diseases are the main cause of death in the world and cardiovascular imaging techniques are the mainstay of noninvasive diagnosis. Aortic stenosis is a lethal cardiac disease preceded by aortic valve calcification for several years. Data-driven tools developed with Deep Learning (DL) algorithms can process and categorize medical images data, providing fast diagnoses with considered reliability, to improve healthcare effectiveness. A systematic review of DL applications on medical images for pathologic calcium detection concluded that there are established techniques in this field, using primarily CT scans, at the expense of radiation exposure. Echocardiography is an unexplored alternative to detect calcium, but still needs technological developments. In this article, a fully automated method based on Convolutional Neural Networks (CNNs) was developed to detect Aortic Calcification in Echocardiography images, consisting of two essential processes: (1) an object detector to locate aortic valve – achieving 95% of precision and 100% of recall; and (2) a classifier to identify calcium structures in the valve – which achieved 92% of precision and 100% of recall. The outcome of this work is the possibility of automation of the detection with Echocardiography of Aortic Valve Calcification, a lethal and prevalent disease.