Consensus for Operating Room Multimodal Data Management: Identifying Research Priorities for Data-Driven Surgery.

Introduction: This study aimed to identify research areas that demand attention in multimodal data-driven surgery for improving data management in minimally invasive surgery. Background: New surgical procedures, high-tech equipment, and digital tools are increasingly being introduced, potentially benefiting patients and surgical teams. These innovations have resulted in operating rooms evolving into data-rich environments, which, in turn, requires a thorough understanding of the data pipeline for improved and more intelligent real-time data usage. As this new domain is vast, it is necessary to identify where efforts should be focused on developing seamless and practical data usage. Methods: A modified electronic Delphi approach was used; 53 investigators were divided into the following groups: a research group (n=9) for problem identification and a narrative literature review, a medical and technical expert group (n=14) for validation, and an invited panel (n=30) for two electronic survey rounds. Round 1 focused on a consensus regarding bottlenecks in surgical data science areas and research gaps, while round 2 prioritized the statements from round 1, and a roadmap was created based on the identified essential and very important research gaps. Results: Consensus panelists have identified key research areas, including digitizing operating room (OR) activities, improving data streaming through advanced technologies, uniform protocols for handling multimodal data, and integrating AI for efficiency and safety. The roadmap prioritizes standardizing OR data formats, integrating OR data with patient information, ensuring regulatory compliance, standardizing surgical AI models, and securing data transfers in the next generation of wireless networks. Conclusions: This work is an international expert consensus regarding the current issues and key research targets in the promising field of data-driven surgery, highlighting the research needs of many operating room stakeholders with the aim of facilitating the implementation of novel patient care strategies in minimally invasive surgery.

Inference of alveolar capillary network connectivity from blood flow dynamics.

The intricate lung structure is crucial for gas exchange within the alveolar region. Despite extensive research, questions remain about the connection between capillaries and the vascular tree. We propose a computational approach combining three-dimensional morphological modeling with computational fluid dynamics simulations to explore alveolar capillary network connectivity based on blood flow dynamics.We developed three-dimensional sheet-flow models to accurately represent alveolar capillary morphology and conducted simulations to predict flow velocities and pressure distributions. Our approach leverages functional features to identify plausible system architectures. Given capillary flow velocities and arteriole-to-venule pressure drops, we deduced arteriole connectivity details. Preliminary analyses for non-human species indicate a single alveolus connects to at least two 20 µm arterioles or one 30 µm arteriole. Hence, our approach narrows down potential connectivity scenarios, but a unique solution may not always be expected.Integrating our blood flow model results into our previously published gas exchange application, Alvin, we linked these scenarios to gas exchange efficiency. We found that increased blood flow velocity correlates with higher gas exchange efficiency.Our study provides insights into pulmonary microvasculature structure by evaluating blood flow dynamics, offering a new strategy to explore the morphology-physiology relationship that is applicable to other tissues and organs. Future availability of experimental data will be crucial in validating and refining our computational models and hypotheses.

Device-Less Data-Driven Cardiac and Respiratory Gating Using LAFOV PET Histo Images.

Background: The outstanding capabilities of modern Positron Emission Tomography (PET) to highlight small tumor lesions and provide pathological function assessment are at peril from image quality degradation caused by respiratory and cardiac motion. However, the advent of the long axial field-of-view (LAFOV) scanners with increased sensitivity, alongside the precise time-of-flight (TOF) of modern PET systems, enables the acquisition of ultrafast time resolution images, which can be used for estimating and correcting the cyclic motion. Methods: 0.25 s so-called [18F]FDG PET histo image series were generated in the scope of for detecting respiratory and cardiac frequency estimates applicable for performing device-less data-driven gated image reconstructions. The frequencies of the cardiac and respiratory motion were estimated for 18 patients using Short Time Fourier Transform (STFT) with 20 s and 30 s window segments, respectively. Results: The Fourier analysis provided time points usable as input to the gated reconstruction based on eight equally spaced time gates. The cardiac investigations showed estimates in accordance with the measured pulse oximeter references (p = 0.97) and a mean absolute difference of 0.4 ± 0.3 beats per minute (bpm). The respiratory frequencies were within the expected range of 10-20 respirations per minute (rpm) in 16 out of 18 patients. Using this setup, the analysis of three patients with visible lung tumors showed an increase in tumor SUVmax and a decrease in tumor volume compared to the non-gated reconstructed image. Conclusions: The method can provide signals that were applicable for gated reconstruction of both cardiac and respiratory motion, providing a potential increased diagnostic accuracy.

Data-Driven Power Prediction for Proton Exchange Membrane Fuel Cell Reactor Systems.

Enhancing high-performance proton exchange membrane fuel cell (PEMFC) technology is crucial for the widespread adoption of hydrogen energy, a leading renewable resource. In this research, we introduce an innovative and cost-effective data-driven approach using the BP-AdaBoost algorithm to accurately predict the power output of hydrogen fuel cell stacks. The algorithm's effectiveness was validated with experimental data obtained from an advanced fuel cell testing platform, where the predicted power outputs closely matched the actual results. Our findings demonstrate that the BP-AdaBoost algorithm achieved lower RMSE and MAE, along with higher R2, compared to other models, such as Partial Least Squares Regression (PLS), Support Vector Machine (SVM), and back propagation (BP) neural networks, when predicting power output for electric stacks of the same type. However, the algorithm's performance decreased when applied to electric stacks with varying material compositions, highlighting the need for more sophisticated models to handle such diversity. These results underscore the potential of the BP-AdaBoost algorithm to improve PEMFC efficiency while also emphasizing the necessity for further research to develop models capable of accurately predicting power output across different types of PEMFC stacks.

New Strategies for constructing and analyzing semiconductor photosynthetic biohybrid systems based on ensemble Machine learning Models: Visualizing complex mechanisms and yield prediction.

Photosynthetic biohybrid systems (PBSs) composed of semiconductor-microbial hybrids provide a novel approach for converting light into chemical energy. However, comprehending the intricate interactions between materials and microbes that lead to PBSs with high apparent quantum yields (AQY) is challenging. Machine learning holds promise in predicting these interactions. To address this issue, this study employs ensemble learning (ESL) based on Random Forest, Gradient Boosting Decision Tree, and eXtreme Gradient Boosting to predict AQY of PBSs utilizing a dataset comprising 15 influential factors. The ESL model demonstrates exceptional accuracy and interpretability (R2 value of 0.927), offering insights into the impact of these factors on AQY while facilitating the selection of efficient semiconductors. Furthermore, this research propose that efficient charge carrier separation and transfer at the bio-abiotic interface are crucial for achieving high AQY levels. This research provides guidance for selecting semiconductors suitable for productive PBSs while elucidating mechanisms underlying their enhanced efficiency.

The Role of Artificial Intelligence on Tumor Boards: Perspectives from Surgeons, Medical Oncologists and Radiation Oncologists.

The integration of multidisciplinary tumor boards (MTBs) is fundamental in delivering state-of-the-art cancer treatment, facilitating collaborative diagnosis and management by a diverse team of specialists. Despite the clear benefits in personalized patient care and improved outcomes, the increasing burden on MTBs due to rising cancer incidence and financial constraints necessitates innovative solutions. The advent of artificial intelligence (AI) in the medical field offers a promising avenue to support clinical decision-making. This review explores the perspectives of clinicians dedicated to the care of cancer patients-surgeons, medical oncologists, and radiation oncologists-on the application of AI within MTBs. Additionally, it examines the role of AI across various clinical specialties involved in cancer diagnosis and treatment. By analyzing both the potential and the challenges, this study underscores how AI can enhance multidisciplinary discussions and optimize treatment plans. The findings highlight the transformative role that AI may play in refining oncology care and sustaining the efficacy of MTBs amidst growing clinical demands.

Enhancing flood risk mitigation by advanced data-driven approach.

Flood events in the Sefidrud River basin have historically caused significant damage to infrastructure, agriculture, and human settlements, highlighting the urgent need for improved flood prediction capabilities. Traditional hydrological models have shown limitations in capturing the complex, non-linear relationships inherent in flood dynamics. This study addresses these challenges by leveraging advanced machine learning techniques to develop more accurate and reliable flood estimation models for the region. The study applied Random Forest (RF), Bagging, SMOreg, Multilayer Perceptron (MLP), and Adaptive Neuro-Fuzzy Inference System (ANFIS) models using historical hydrological data spanning 50 years. The methods involved splitting the data into training (50-70 %) and validation sets, processed using WEKA 3.9 software. The evaluation revealed that the nonlinear ensemble RF model achieved the highest accuracy with a correlation of 0.868 and an root mean squared error (RMSE) of 0.104. Both RF and MLP significantly outperformed the linear SMOreg approach, demonstrating the suitability of modern machine learning techniques. Additionally, the ANFIS model achieved an exceptional R-squared accuracy of 0.99. The findings underscore the potential of data-driven models for accurate flood estimating, providing a valuable benchmark for algorithm selection in flood risk management.

Examining the development and utilisation of Community-Based Health Information Systems (CBHIS) in Africa: A Scoping Review.

Introduction: The community-based health information system (CBHIS) is a vital component of the community health system, as it assesses community-level healthcare service delivery and generates data for community health programme planning, monitoring, and evaluation. CBHIS promotes data-driven decision-making, by identifying priority interventions and programs, guiding resource allocation, and contributing to evidence-based policy development. Objective: This scoping review aims to comprehensively examine the use of CBHIS in African countries, focusing on data generation, pathways, utilization of CBHIS data, community accessibility to the data and use of the data to empower communities. Methods: We utilised Arksey and O'Malley's scoping review methodology. We searched eight databases: PubMed, EMBASE, HINARI, Cochrane Library, Web of Science, Scopus, Google Scholar, and grey literature databases (Open Grey and OAIster). We synthesized findings using a thematic approach. Results: Our review included 55 articles from 27 African countries, primarily in Eastern and Southern Africa, followed by West Africa. Most of the studies were either quantitative (42%) or qualitative (33%). Paper-based systems are primarily used for data collection in most countries, but some have adopted electronic/mobile-based systems or both. The data flow for CBHIS varies by country and the tools used for data collection. CBHIS data informs policies, resource allocation, staffing, community health dialogues, and commodity supplies for community health programmes. Community dialogue is the most common approach for community engagement, empowerment, and sharing of CBHIS data with communities. Community empowerment tends towards health promotion activities and health provider-led approaches. Conclusion: CBHIS utilizes both paper-based and electronic-based systems to collect and process data. Nevertheless, most countries rely on paper-based systems. Most of the CBHIS investments have focused on its digitization and enhancing data collection, process, and quality. However, there is a need to shift the emphasis towards enabling data utilisation at the community level and community empowerment.

For community health services and systems to work well, health managers and other data users, including policy and decision-makers, need a community-based health information system (CBHIS) that produces reliable and timely information on how well these services are working and that supports the use of CBHIS data to improve community health service delivery. This scoping review aimed to explore the use of CBHIS in African countries. It focused on data generation, pathways, use of CBHIS data, community data access, and use of CBHIS data to empower communities. The review authors collected and analysed all relevant studies to answer this question and found 55 articles from 27 African countries. The review found that most countries use paper-based information systems for data collection, while some have adopted electronic and digital systems. CBHIS also collects information on human resources, medicines, and supply systems. CBHIS data are used to guide policy development, allocate resources, track commodities supplies, staff for community health programmes and organise community health dialogues. Community dialogue is the most common approach for engaging, empowering, and sharing CBHIS data with communities. Community empowerment involves activities that promote health and health provider-led approaches. There is a need to focus on enabling the use of data at the community level and empowerment.

Forecasting and Predicting Stochastic Agent-Based Model Data with Biologically-Informed Neural Networks.

Collective migration is an important component of many biological processes, including wound healing, tumorigenesis, and embryo development. Spatial agent-based models (ABMs) are often used to model collective migration, but it is challenging to thoroughly predict these models' behavior throughout parameter space due to their random and computationally intensive nature. Modelers often coarse-grain ABM rules into mean-field differential equation (DE) models. While these DE models are fast to simulate, they suffer from poor (or even ill-posed) ABM predictions in some regions of parameter space. In this work, we describe how biologically-informed neural networks (BINNs) can be trained to learn interpretable BINN-guided DE models capable of accurately predicting ABM behavior. In particular, we show that BINN-guided partial DE (PDE) simulations can (1) forecast future spatial ABM data not seen during model training, and (2) predict ABM data at previously-unexplored parameter values. This latter task is achieved by combining BINN-guided PDE simulations with multivariate interpolation. We demonstrate our approach using three case study ABMs of collective migration that imitate cell biology experiments and find that BINN-guided PDEs accurately forecast and predict ABM data with a one-compartment PDE when the mean-field PDE is ill-posed or requires two compartments. This work suggests that BINN-guided PDEs allow modelers to efficiently explore parameter space, which may enable data-driven tasks for ABMs, such as estimating parameters from experimental data. All code and data from our study is available at https://github.com/johnnardini/Forecasting_predicting_ABMs .

Legacy sources determine current water quality: Nitrogen and phosphorus in streams of Australia, China, Sweden and USA.

Waterborne nutrient loads to downstream ecosystems integrate contributions from both active and legacy sources. Effective mitigation of nutrient pollution and eutrophication around the world requires distinction of these, largely unknown, relative load contributions. Here, the active and legacy contributions to nitrogen and phosphorus loads are distinguished in numerous streams and associated hydrological catchments of Australia, China, Sweden, and USA. The legacy contributions overshadow the active ones in all countries during 2005-2020. China and USA, with higher population densities and related overall human-activity levels, also have substantial active contributions. The median values of legacy concentration contributions of total nitrogen range from 321 (in Sweden) to 1850 µg/L (in USA); whereas the active contributions range from 2.2 (in Australia) to 315 µg/L (in USA). In China, nitrogen data are available only for ammonia, with median concentration contributions of 294 µg/L for legacy and 352 µg/L for active sources. For total phosphorus, the median values of legacy concentration contributions range from 28.8 (in Sweden) to 270 µg/L (in USA), while the active ones range from 0.1 (in Australia) to 67.3 µg/L (in USA). For relatively fast mitigation responses, China and USA need to mitigate their current nutrient emissions, while Australia and Sweden need a shift in mitigation focus to targeting their dominant legacy source contributions. The data-driven method testing in this study supports the used source distinction-attribution approach. This enables consistent source identification and tailoring of targeted measures for effective nutrient load mitigation in various regional contexts.

Toward the Integration of Machine Learning and Molecular Modeling for Designing Drug Delivery Nanocarriers.

The pioneering work on liposomes in the 1960s and subsequent research in controlled drug release systems significantly advances the development of nanocarriers (NCs) for drug delivery. This field is evolved to include a diverse array of nanocarriers such as liposomes, polymeric nanoparticles, dendrimers, and more, each tailored to specific therapeutic applications. Despite significant achievements, the clinical translation of nanocarriers is limited, primarily due to the low efficiency of drug delivery and an incomplete understanding of nanocarrier interactions with biological systems. Addressing these challenges requires interdisciplinary collaboration and a deep understanding of the nano-bio interface. To enhance nanocarrier design, scientists employ both physics-based and data-driven models. Physics-based models provide detailed insights into chemical reactions and interactions at atomic and molecular scales, while data-driven models leverage machine learning to analyze large datasets and uncover hidden mechanisms. The integration of these models presents challenges such as harmonizing different modeling approaches and ensuring model validation and generalization across biological systems. However, this integration is crucial for developing effective and targeted nanocarrier systems. By integrating these approaches with enhanced data infrastructure, explainable AI, computational advances, and machine learning potentials, researchers can develop innovative nanomedicine solutions, ultimately improving therapeutic outcomes.

From Then to Now and Beyond: Exploring How Machine Learning Shapes Process Design Problems.

Following the discovery of the least squares method in 1805 by Legendre and later in 1809 by Gauss, surrogate modeling and machine learning have come a long way. From identifying patterns and trends in process data to predictive modeling, optimization, fault detection, reaction network discovery, and process operations, machine learning became an integral part of all aspects of process design and process systems engineering. This is enabled, at the same time necessitated, by the vast amounts of data that are readily available from processes, increased digitalization, automation, increasing computation power, and simulation software that can model complex phenomena that span over several temporal and spatial scales. Although this paper is not a comprehensive review, it gives an overview of the recent history of machine learning models that we use every day and how they shaped process design problems from the recent advances to the exploration of their prospects.

Methanogenic Potential of Sewer Microbiomes and Its Implications for Methane Emission.

The sewer system, despite being a significant source of methane emissions, has often been overlooked in current greenhouse gas inventories due to the limited availability of quantitative data. Direct monitoring in sewers can be expensive or biased due to access limitations and internal heterogeneity of sewer networks. Fortunately, since methane is almost exclusively biogenic in sewers, we demonstrate in this study that the methanogenic potential can be estimated using known sewer microbiome data. By combining data mining techniques and bioinformatics databases, we developed the first data-driven method to analyze methanogenic potentials using a data set containing 633 observations of 53 variables obtained from literature mining. The methanogenic potential in the sewer sediment was around 250-870% higher than that in the wet biofilm on the pipe and sewage water. Additionally, k-means clustering and principal component analysis linked higher methane emission rates (9.72 ± 51.3 kgCO2 eq m-3 d-1) with smaller pipe size, higher water level, and higher potentials of sulfate reduction in the wetted pipe biofilm. These findings exhibit the possibility of connecting microbiome data with biogenic greenhouse gases, further offering insights into new approaches for understanding greenhouse gas emissions from understudied sources.

Tuning and modeling cheese flavor.

Flavor is a major sensory attribute affecting consumers' preference for cheese products. Differences in cheesemaking change the cheese microenvironment, thereby affecting cheese flavor profiles. A framework for tuning cheese flavor is proposed in this study, which depicts the full picture of flavor development and modulation, from manufacturing and ripening factors through the main biochemical pathways to flavor compounds and flavor notes. Taking semi-hard and hard cheeses as examples, this review describes how cheese flavor profiles are affected by milk type and applied treatment, fat and salt content, microbiota composition and microbial interactions, ripening time, temperature, and environmental humidity, together with packaging method and material. Moreover, these factors are linked to flavor profiles through their effects on proteolysis, the further catabolism of amino acids, and lipolysis. Acids, alcohols, ketones, esters, aldehydes, lactones, and sulfur compounds are key volatiles, which elicit fruity, sweet, rancid, green, creamy, pungent, alcoholic, nutty, fatty, and sweaty flavor notes, contributing to the overall flavor profiles. Additionally, this review demonstrates how data-driven modeling techniques can link these influencing factors to resulting flavor profiles. This is done by providing a comprehensive review on the (i) identification of key factors and flavor compounds, (ii) discrimination of cheeses, and (iii) prediction of flavor notes. Overall, this review provides knowledge tools for cheese flavor modulation and sheds light on using data-driven modeling techniques to aid cheese flavor analysis and flavor prediction.

Sensitivity analysis of slope stability based on eXtreme gradient boosting and SHapley Additive exPlanations: An exploratory study.

Slope instability through can cause catastrophic consequences, so slope stability analysis has been a key topic in the field of geotechnical engineering. Traditional analysis methods have shortcomings such as high operational difficulty and time-consuming, for this reason many researchers have carried out slope stability analysis based on AI. However, the current relevant studies only judged the importance of each factor and did not specifically quantify the correlation between factors and slope stability. For this purpose, this paper carried out a sensitivity analysis based on the XGBoost and SHAP. The sensitivity analysis results of SHAP were also validated using GeoStudio software. The selected influence factors included slope height ( H ), slope angle ( ß ), unit weight ( γ ), cohesion ( c ), angle of internal friction ( φ ) and pore water pressure coefficient ( r u ). The results showed that c and γ were the most and least important influential parameters, respectively. GeoStudio simulation results showed a negative correlation between γ , ß , H , r u and slope stability, while a positive correlation between c , φ and slope stability. However, for real data, SHAP misjudged the correlation between γ and slope stability. Because current AI lacked common sense knowledge and, leading SHAP unable to effectively explain the real mechanism of slope instability. For this reason, this paper overcame this challenge based on the priori data-driven approach. The method provided more reliable and accurate interpretation of the results than a real sample, especially with limited or low-quality data. In addition, the results of this method showed that the critical values of c , φ , ß , H , and r u in slope destabilization are 18 Kpa, 28°, 32°, 30 m, and 0.28, respectively. These results were closer to GeoStudio simulations than real samples.

Thermal Metamaterials with Configurable Mechanical Properties.

Thermal metamaterials are typically achieved by mixing different natural materials to realize effective thermal conductivities (ETCs) that conventional materials do not possess. However, the necessity for multifunctional design of metamaterials, encompassing both thermal and mechanical functionalities, is somewhat overlooked, resulting in the fixation of mechanical properties in thermal metamaterials designed within current research endeavors. Thus far, conventional methods have faced challenges in designing thermal metamaterials with configurable mechanical properties because of intricate inherent relationships among the structural configuration, thermal and mechanical properties in metamaterials. Here, a data-driven approach is proposed to design a thermal metamaterial capable of seamlessly achieving thermal functionalities and harnessing the advantages of microstructural diversity to configure its mechanical properties. The designed metamaterial possesses thermal cloaking functionality while exhibiting exceptional mechanical properties, such as load-bearing capacity, shearing strength, and tensile resistance, thereby affording mechanical protection for the thermal metadevice. The proposed approach can generate numerous distinct inverse design candidate topological functional cells (TFCs), designing thermal metamaterials with dramatic improvements in mechanical properties compared to traditional ones, which sets up a novel paradigm for discovering thermal metamaterials with extraordinary mechanical structures. Furthermore, this approach also paves the way for investigating thermal metamaterials with additional physical properties.

In situ Synchrotron X-ray Metrology Boosted by Automated Data Analysis for Real-time Monitoring of Cathode Calcination.

Synchrotron X-ray-based in situ metrology is advantageous for monitoring the synthesis of battery materials, offering high throughput, high spatial and temporal resolution, and chemical sensitivity. However, the rapid generation of massive data poses a challenge to on-site, on-the-fly analysis needed for real-time process monitoring. Here, a weighted lagged cross-correlation (WLCC) similarity approach is presented for automated data analysis, which merges with in situ synchrotron X-ray diffraction metrology to monitor the calcination process of the archetypal nickel-based cathode, LiNiO2. The WLCC approach, incorporating variables that account for peak shifts and width changes associated with structural transformations, enables rapid extraction of phase progression within 10 seconds from tens of diffraction patterns. Details are captured, from initial precursors to intermediates and the final layered LiNiO2, providing information for agile on-site adjustments during experiments and complementing post hoc diffraction analysis by offering insights into early-stage phase nucleation and growth. Expanding this data-powered platform paves the way for real time calcination process monitoring and control, which is pivotal to quality control in battery cathode manufacturing.

Modularized neural network incorporating physical priors for future building energy modeling.

Building energy modeling (BEM) is fundamental for achieving optimized energy control, resilient retrofit designs, and sustainable urbanization to mitigate climate change. However, traditional BEM requires detailed building information, expert knowledge, substantial modeling efforts, and customized case-by-case calibrations. This process must be repeated for every building, thereby limiting its scalability. To address these limitations, we developed a modularized neural network incorporating physical priors (ModNN), which is improved by its model structure incorporating heat balance equations, physically consistent model constraints, and data-driven modular design that can allow for multiple-building applications through model sharing and inheritance. We demonstrated its scalability in four cases: load prediction, indoor environment modeling, building retrofitting, and energy optimization. This approach provides guidance for future BEM by incorporating physical priors into data-driven models without extensive modeling efforts, paving the way for large-scale BEM, energy management, retrofit designs, and buildings-to-grid integration.