Merging Data Research Articles

Multimodal biometric integration: Trends and insights from the past quinquennial

Recent years have seen a notable breakthrough in multimodal biometric systems, which combine numerous biometric modalities for increased security and accuracy. This paper offers a thorough analysis of recent advances in the field, addressing the many multimodal biometric strategies that have been put out by different researchers. Numerous modalities are covered by the evaluation, such as palm print, speech, iris, facial features, palm print, fingerprint, body form, gait, and more. To obtain amazing results in terms of accuracy and security, researchers have used cutting-edge methods like convolutional neural networks (CNNs), support vector machines (SVM), recurrent neural networks (RNNs), deep learning, and optimization algorithms. Many fusion strategies have been investigated to efficiently merge data from many modalities, such as feature-level fusion, decision-level fusion, and score-level fusion. Developments in template protection systems have also addressed security issues related to transmission and storage of biometric data. Even though a lot of the suggested systems have shown great accuracy rates, there are still issues with hardware restrictions, dataset biases, privacy problems, and computational costs. Prospective study avenues encompass investigating more extensive and heterogeneous datasets, crafting resilient fusion algorithms, and amalgamating cutting-edge technologies like deep learning-based biometric cryptosystems and federated learning. All things considered, the literature study demonstrates the enormous potential of multimodal biometric systems in offering safe and dependable authentication solutions for a wide range of applications, from on-line student authentication and proctoring systems to customized healthcare networks.

The Mental Health Consequences of George Floyd's Murder in Minneapolis in Black, Latine, and White Communities.

The high-profile police murder of George Floyd is likely to have an aftermath of negative health consequences, particularly among Black people. Our study evaluates the impact of the murder of Mr. Floyd on mental health in Black, Latine, and white communities in Minneapolis, Minnesota. We constructed a panel dataset merging data from the Minnesota Hospital Association, Minnesota Department of Natural Resources, Minneapolis Police Department, and American Community Survey. First, we specify an overall and racial subgroup autoregressive interrupted time-series design to identify the impact of the murder on rates of mental health hospital discharge at the city-level. We then examine the spatial heterogeneity in the impact of the murder by specifying zip code tabulation area (ZCTA)-level panel models. We find a 0.23 per 1,000 increase in mental health conditions among Black people in the immediate post-murder period, followed by a weekly decline (-.007) in mental health diagnoses. We do not find a substantial rate increase in White or Latine residents. Further, our analyses at the ZCTA-week-level corroborate these findings, while showing that the increase for Black residents was global. These findings speak to the traumatizing effects of police violence and the short- and longer-term public health consequences for Black communities.

ADHD Assessment and Treatment Services in a Sample of U.S. Colleges and Universities.

Up to 2 million college students in the United States have been diagnosed as having attention-deficit hyperactivity disorder (ADHD), a condition associated with negative academic, social, and psychiatric outcomes. The authors investigated the online availability and content of policies governing ADHD services at college clinics. Using a stratified sample of 200 colleges and universities, the authors reviewed clinic websites and invited clinical staff to participate in a survey. They weighted percentages to account for oversampling and used regression modeling to examine associations with policy availability. Only 70 institutions (32%, weighted percentage) provided information about ADHD services online. Institutions with <1,000 students had significantly lower odds of providing information online (adjusted odds ratio [AOR]=0.04, 95% CI=0.01-0.26), as did institutions that accepted >67% of applicants (AOR=0.18, 95% CI=0.07-0.48). After merging data from the Web review and survey, the authors noted that 14% (N=11 of 75 institutions with data available for this variable) facilitated neuropsychological assessments on campus, 49% (N=33 of 72) did not allow stimulant medications to be prescribed, 73% (N=43 of 61) did not offer clinical evaluations for ADHD, and 89% (N=32 of 35) required a neuropsychological assessment to receive prescription stimulants. Information about the assessment and management of ADHD is rarely available online, and ADHD services on U.S. college campuses appear to be limited.

Probabilistic Record Linkage for Families (PRLF): A Discussion of the Development and Validation of this Open-Source Linkage Tool

Linking administrative records across programs can yield person-centered information, including client characteristics, public service trajectories, and outcomes and help to answer policy-related questions. Several solutions are available for undertaking record linkage, producing linkage keys for merging data sources for positively matched pairs of records. In this session, we will demonstrate a new application of the Python RecordLinkage package to family-based record linkages with machine learning algorithms for probability scoring, which we call probabilistic record linkage for families (PRLF). First, we will demonstrate the utility of PRLF with a simulation of administrative records and assess linkage accuracy with variations in match rates and data degradation. Second, we will compare generalized linear model estimates across three record linkage solutions (PRLF, ChoiceMaker, and Link Plus). Findings from the simulation study indicate linkage accuracy is largely influenced by degradation (e.g., missing data fields, erroneous or incomplete values) compared to the proportion of simulated matches between datasets. Results from the methods comparison using real world data indicate that all three solutions, when optimized, provide similar results for researchers. We discuss the strengths of our process, such as the use of ensemble methods, to improve match accuracy. We then will identify caveats of record linkage in the context of administrative data. The tool was developed in Python to allow for researchers to work with open-source software and adjust the basic workflow to fit their linkage needs. We will identify several partnerships where this collaboration has worked successfully and empower attendees with access to this useful tool.

As our population changes, so must we: how race and ethnicity data merging has evolved over time in the Wisconsin Administrative Data Core

The Institute for Research on Poverty (IRP), centered at the University of Wisconsin-Madison in Wisconsin, USA, began collaboration with various Wisconsin state agencies in 1985 to link data between disparate state agencies. These initial efforts have since evolved into the Wisconsin Administrative Data Core (WADC) which is an individual-level population dataset connecting extensive administrative data spanning decades of Wisconsin social services program history. Since its purpose is to facilitate evidence-driven policymaking and support program evaluation, it is essential that our data linking efforts accurately represent our Wisconsin population. Since its inception, the WADC has adapted to changes in how race and ethnicity are collected and reported. While Wisconsin has a racial makeup that is generally less diverse than the national population, the number of non-white and multi-racial identifying residents has increased over time. Furthermore, varying state agency demographic collection and reporting methods have fueled the need for our linking methods to change, too. IRP initially implemented standard data linkage methods using a mixture of probabilistic and deterministic matching of personal identifying information. But as the racial composition of Wisconsin residents changed over time, our linking assumptions induced implicit biases. In response, we updated our matching algorithm to account for the cultural nuances in how personally identifying information is assigned and reported within various racial and ethnic subgroups. Furthermore, we prioritized values from data sources that were self-reported rather than assigned from a third-party. These methodological modifications have improved the WADC’s reliability and representativeness of the population.

The Challenges and Strategies for Cleaning the Routinely Collected Data for Total Knee Replacements (TKR) and Total Hip Replacements (THR) in the Kingdom of Fife

Knee and hip replacement surgeries have revolutionised orthopaedic medicine by providing individuals with surgeries resulting in pain relief, improved mobility and a better quality of life. NHS Fife has been collecting routine data since 1998 on patient demographics, type of surgery, recovery time, follow-ups, revision, and outcomes, including complications. This work reports the challenges and strategies to clean the collected data. Data was available for 12,313 patients, totalising 14,524 THR and TKR surgeries performed in NHS Fife between 1998 and 2021. The original dataset consisted of 34 unique-structured files covering distinct periods and representing five categories: primary surgery, follow-up, infection, mortality, and revision. Data wrangling was done using the programming language R, transforming the raw data into a more functional form. The main elements of the data-cleaning procedure were combining variables representing the same information and converting void/corrupted/purposeless values into missing data, providing a standardisation for the collected information. Then, a rigorous data merging resulted in reshaping this complex collection of files into one single streamlined file for each of the five categories and type of surgery. Patients who had multiple surgeries were also cross-identifying, enabling the characterisation of the epidemiology of multiple lower limb joint replacements. The structured files can now be linked to studies across different datasets like social care, prescriptions, and primary care. It can also be used to develop future machine learning models to assess the risk of multiple primary and revision knee and hip replacements and their outcomes.

Transferability across countries of equations developed using milk mid-infrared spectroscopy to estimate daily body condition score change in dairy cows

Routine milk samples are commonly subjected to spectroscopic analysis within the mid-infrared (MIR) region of the electromagnetic spectrum to estimate macro-constituents of milk like fat, protein, lactose, and urea content. These spectra, however, can also be used to predict other traits, such as daily body condition score (BCS) change. The objective of the present study was to assess the transferability across countries of equations to predict daily body condition score change (ΔBCS) developed using milk MIR data collected in Ireland and in Canada. Body condition was scored on a scale from 1 (emaciated) to 5 (obese) in both countries. A total of 347,254 BCS records from 80,400 Canadian cows were available along with 73,193 BCS records from 6,572 Irish cows. Partial least squares regression (PLSR) and neural networks (NN) were separately used to predict daily ΔBCS. Two scenarios were studied 1) using Canadian and Irish data combined as the calibration data set to predict daily ΔBCS in Canada and in Ireland separately, and 2) Canadian and Irish data used separately to predict daily ΔBCS in each country separately. These prediction methods were applied to data with and without pretreatment (i.e., first derivative of the spectrum) as well as with and without standardizing daily ΔBCS across countries. For all the scenarios investigated, the correlation between actual and predicted daily ΔBCS when calibrated and validated (using cross-validation) in the same country ranged from 0.92 to 0.94, and from 0.85 to 0.87 for the Canadian and Irish data sets, respectively. When the data from Canada and Ireland were combined in the calibration process to predict daily ΔBCS, the correlations between actual and predicted ΔBCS were ≥ 0.90 and ≥ 0.80 for Canadian and Irish daily ΔBCS, respectively indicating no improvement in predictive ability. Predictive performance when calibrated using just Canadian data and validated using just Irish data was poor, and vice versa. Nonetheless, when developing equations for a country for which a limited database (i.e., 100 records) of gold standard and MIR data were available, predictive performance improved when the limited database was supplemented with the large data set from the other country. In general, for some of the investigated scenarios, standardizing the daily ΔBCS data within country before undertaking the calibration improved prediction accuracy. In conclusion, the benefit of merging data from different countries, at least based on the trait (i.e., daily ΔBCS) and countries (i.e., Ireland and Canada) considered in the present study were limited and, in cases, counter-productive.

How competition play a role in dental pricing? A study on French medico-administrative and tax reports dataset

ObjectivesFrench dentists charge additional fees for dental prostheses. This paper aims to provide new information on the determinants of dental price setting and inform public decision-making in the context of the widespread rejection of prosthetic dental care for financial reasons. We focus on the competitive mechanism in the dental prosthetics market and measure the impact of the density of professionals and competitors' prices on the fees charged by dentists. MethodsWe use data merging from an administrative health insurance database and information from tax declarations of French dentists. We test the effect of competitor prices and competition on individual price-setting using instrumental variables. The database obtained included 29,220 dentists. ResultsPractitioners’ prices grow with competitors’ prices (+1€ in competitor prices entails an increase of + 0.37€ in the practitioner's price). Women set lower prices, and having a young child in the household predicts an increase in price of 6.8€ (p-value=0.014). Rural areas present lower fees than urban areas (+11.4€ (p value=0.000)). ConclusionProsthetic prices are strategic complements that are compatible with the application of monopolistic competition in the dental care market. We encourage the regulator to develop competitive mechanisms, for example, through a public offer at moderate prices.

A novel mode shape identification approach for structures having planes with rigid-like behavior

The identification of mode shapes of structures through Operational Modal Analysis (OMA) often requires the application of data merging techniques to compensate for the lack of information on mode shapes scaling factors, which is inherent in OMA. In this paper, we propose a novel mode shape identification approach for structures having planes with rigid-like behavior, such as steel or reinforced concrete buildings with rigid floors. The approach is based on a theoretical model that generalizes the mechanical features of the structures under considerations. We show that the mode shapes of the model can be reconstructed starting from two components, i.e., modal centers of rotation and modal rotations; modal rotations depend on scaling factors of mode shapes, while modal centers of rotation turn out to be invariant with respect to mode shape scaling. Afterwards, we develop a method for identifying modal centers of rotation and modal rotations from experimental data, and then for reconstructing mode shapes. Numerical experiments have been performed to assess the capability of the approach with respect to a structural specimen having known modal properties. Compared with classic merging techniques, our approach enables a significant simplification of the experimental setup and a deeper analysis of mode shapes.

Current trends in radiological diagnostics of the temporal bone pyramid pathologies

Diseases of the middle ear and the temporal bone pyramid remain one of the most common diseases in the structure of otorhinolaryngological pathologies, which have a significant impact on the quality of life of adults and children. Exudative processes, as well as their consequences, become one of the main causes of persistent hearing loss and are characterized by an asymptomatic course with a gradual evolution to chronic forms. It should be noted that in childhood the risk of impaired speech and intellectual development increases, leading to changes in the psycho-emotional status, social maladaptation of the child, and a decrease in the quality of life. Timely and complete diagnosis and high-quality follow-up takes a leading role in preventing the development of conductive hearing loss and other complicated pathologies of the temporal bone pyramid.&lt;br /&gt; Purpose of the study. To study the main predisposing factors, methods of radiological diagnosis and dynamic observation of patients, as well as some modern aspects of the treatment of the most common pathologies of the temporal bone pyramid.&lt;br /&gt; Methodology. We conducted an analytical review of open access sources from scientific databases such as Scopus, PubMed, Google Scholar, Web of Science, DisserCat, CyberLeninka. Key words used for the search: “chronic otitis media”, “cholesteatoma”, “mastoiditis”, “complications”, “magnetic resonance imaging”, “DWI”, “computed tomography”. Publications included in the literature review were full-text articles in English and Russian and dissertations in Russian. Exclusion criteria: brief reports, newspaper articles and personal communications. The search range was 5 years (2018-2023). A total of 108 sources were analyzed, of which 50 met the research objectives and inclusion criteria.&lt;br /&gt; Results and conclusions. In the diagnosis of temporal bone pathologies, the gold standard for diagnosis is computed tomography, but non-ionizing research techniques are increasingly used. Indications for the use of magnetic resonance imaging of the temporal bones are constantly expanding, and techniques for merging data with computed tomography data are being increasingly introduced. As a follow-up tool, magnetic resonance therapy is increasingly becoming a priority technique to reduce patient radiation exposure in the long term. In the treatment of complicated forms of diseases, endoscopic and laser correction techniques are used in foreign practice, helping to reduce the risk of recurrence and postoperative complications.&lt;br /&gt; &lt;b&gt;Key words&lt;/b&gt;: chronic otitis media, cholesteatoma, magnetic resonance imaging, computed tomography, treatment

Discriminative analysis of diabetic retinopathy using cascaded network with Atrous convolution and fundus biomarkers

The proposed discriminative analysis of Diabetic Retinopathy aims to develop a discriminative model to perform binary and multi-grading of DR and DME combined in a single input. The discriminative model consists of a cascade network with novel lightweight CNN and Of-the-self CNN Xception. Furthermore, to strengthen the capability of the system, limited data merging technique is inserted into the training set. This research also investigates the various network architectures and cascade networks with different Of-the-self CNN to find the optimal network. The proposed model is validated on the challenging dataset IDRiD with performance matrix accuracy, FPR, precision, recall, and f1 score. At last, the comparison with related state-of-the-art work reveals the significance of the developed system in Diabetic retinopathy.

Family history of cancer and lung cancer: Utility of big data and artificial intelligence for exploring the role of genetic risk

ObjectivesLung Cancer (LC) is a multifactorial disease for which the role of genetic susceptibility has become increasingly relevant. Our aim was to use artificial intelligence (AI) to analyze differences between patients with LC based on family history of cancer (FHC). Materials and methodsFrom August 2016 to June 2020 clinical information was obtained from Thoracic Tumors Registry (TTR), a nationwide database sponsored by the Spanish Lung Cancer Group. In addition to descriptive statistical analysis, an AI-assisted analysis was performed. The German Technical Information Library supported the merging of data from the electronic medical records and database of the TTR. The results of the AI-assisted analysis were reported using Knowledge Graph, Unified Schema and descriptive and predictive analyses. ResultsAnalyses were performed in two phases: first, conventional statistical analysis including 11,684 patients of those 5,806 had FHC. Median overall survival (OS) for the global population was 23 months (CI 95 %: 21.39–24.61) in patients with FHC versus 21 months (CI 95 %: 19.53–22.48) in patients without FHC (NFHC), p < 0.001. The second AI-assisted analysis included 5,788 patients of those 939 had FHC. 58.48 % of women with FHC had LC. 9.53 % of patients had an EGFR or HER2 mutation or ALK translocation and at least one relative with cancer. A family history of LC was associated with an increased risk of smoking-related LC. Non-smokers with a family history of LC were more likely to have an EGFR mutation in NSCLC. In Bayesian network analysis, 55 % of patients with a family history of LC and never-smokers had an EGFR mutation. ConclusionIn our population, the incidence of LC in patients with a FHC is higher in women and younger patients. FHC is a risk factor and predictor of LC development, especially in people ≤ 50 years. These results were confirmed by conventional statistics and AI-assisted analysis.

Reverse scan body: Complete digital workflow for double full-arch zirconia prostheses.

During intraoral complete-arch digital implant scanning, one of the most technique-sensitive steps is the implant data acquisition and merging of different surface geometry data sets for prototype prosthesis fabrication due to the absence of fixed landmarks. Recently developed extraoral techniques have allowed for an alternative approach for the digital acquisition of implant position in completely edentulous patients. Specifically, extraoral digital scanning of the conversion prosthesis after connecting reverse scan bodies has been proposed as an efficient approach. The reverse scan body protocol digitally simulates the traditional back-pouring technique long utilized in analog workflows. The present article describes a technique for simplifying the digital workflow for the fabrication of passive-fitting definitive prostheses using the reverse scan body concept.

A gene set enrichment analysis for the cancer hallmarks

The "hallmarks of cancer" concept provides a valuable framework for understanding fundamental organizing principles common to various cancers. However, without a consensus gene set for cancer hallmarks, data comparison and integration result in diverse biological interpretations across studies. Therefore, we aimed to form a consensus cancer hallmark gene set by merging data from available mapping resources and establishing a framework for mining these gene sets. By consolidating data from seven projects, 6,763 genes associated with ten cancer hallmarks were identified. A cancer hallmarks enrichment analysis was performed for prognostic genes associated with overall survival across twelve types of solid tumors. "Tissue invasion and metastasis" was most prominent in cancers of the stomach (P = 2.2×10−11), pancreas (P = 4.2×10−9), bladder (P = 3.3×10−8), and ovaries (P = 0.0007), aligning with their heightened potential to spread. "Sustained angiogenesis" was most prominent in squamous cell carcinomas of the lung (P = 2.5×10−7), while "genome instability" showed strong enrichment in lung adenocarcinomas (P = 1.5×10−8) and cancers of the liver (P = 5.5×10−10), pancreas (P = 2.1×10−5), and kidney (P = 0.018). Pancreatic cancers displayed the highest enrichment of hallmarks, emphasizing the disease's complexity, while in melanomas, and cancers of the liver, prostate, and kidney, a single hallmark was enriched among the prognostic markers of survival. Additionally, an online tool (www.cancerhallmarks.com) that allows the identification of cancer-associated hallmarks from new gene sets was established. In summary, our aim of establishing a consensus list of cancer hallmark genes was achieved. Furthermore, the analysis of survival-associated genes revealed a unique pattern of hallmark enrichment with potential pharmacological implications in different tumor types.

Blended Genome Exome (BGE) as a Cost Efficient Alternative to Deep Whole Genomes or Arrays.

Genomic scientists have long been promised cheaper DNA sequencing, but deep whole genomes are still costly, especially when considered for large cohorts in population-level studies. More affordable options include microarrays + imputation, whole exome sequencing (WES), or low-pass whole genome sequencing (WGS) + imputation. WES + array + imputation has recently been shown to yield 99% of association signals detected by WGS. However, a method free from ascertainment biases of arrays or the need for merging different data types that still benefits from deeper exome coverage to enhance novel coding variant detection does not exist. We developed a new, combined, "Blended Genome Exome" (BGE) in which a whole genome library is generated, an aliquot of that genome is amplified by PCR, the exome regions are selected and enriched, and the genome and exome libraries are combined back into a single tube for sequencing (33% exome, 67% genome). This creates a single CRAM with a low-coverage whole genome (2-3x) combined with a higher coverage exome (30-40x). This BGE can be used for imputing common variants throughout the genome as well as for calling rare coding variants. We tested this new method and observed >99% r 2 concordance between imputed BGE data and existing 30x WGS data for exome and genome variants. BGE can serve as a useful and cost-efficient alternative sequencing product for genomic researchers, requiring ten-fold less sequencing compared to 30x WGS without the need for complicated harmonization of array and sequencing data.

Towards New Methodology for Cross-Validation of Clinical Evaluation Scales and Functional MRI in Psychiatry.

Objective biomarkers have been a critical challenge for the field of psychiatry, where diagnostic, prognostic, and theranostic assessments are still based on subjective narratives. Psychopathology operates with idiographic knowledge and subjective evaluations incorporated into clinical assessment inventories, but is considered to be a medical discipline and, as such, uses medical intervention methods (e.g., pharmacological, ECT; rTMS; tDCS) and, therefore, is supposed to operate with the language and methods of nomothetic networks. The idiographic assessments are provisionally "quantified" into "structured clinical scales" to in some way resemble nomothetic measures. Instead of fostering data merging and integration, this approach further encapsulates the clinical psychiatric methods, as all other biological tests (molecular, neuroimaging) are performed separately, only after the clinical assessment has provided diagnosis. Translational cross-validation of clinical assessment instruments and fMRI is an attempt to address the gap. The aim of this approach is to investigate whether there exist common and specific neural circuits, which underpin differential item responses to clinical self-rating scales during fMRI sessions in patients suffering from the two main spectra of mental disorders: schizophrenia and major depression. The current status of this research program and future implications to promote the development of psychiatry as a medical discipline are discussed.

Underwater Mapping and Optimization Based on Multibeam Echo Sounders

Multibeam echo sounders (MBESs) enable extensive underwater environment exploration. However, due to weak correlation between adjacent multibeam sonar data and difficulties in inter-frame feature matching, the resulting underwater mapping accuracy frequently falls short of the desired level. To address this issue, this study presents the development of a multibeam data processing system, which includes functionalities for sonar parameter configuration, data storage, and point cloud conversion. Subsequently, an Iterative Extended Kalman Filter (iEKF) algorithm is employed for odometry estimation, facilitating the initial construction of the point cloud map. To further enhance mapping accuracy, we utilize the Generalized Iterative Closest Point (GICP) algorithm for point cloud registration, effectively merging point cloud data collected at different times from the same location. Finally, real-world lake experiments demonstrate that our method achieves an Absolute Trajectory Error (ATE) of 15.10 m and an average local point cloud registration error of 0.97 m. Furthermore, we conduct measurements on various types of artificial targets. The experimental results indicate that the average location error of the targets calculated by our method is 4.62 m, which meets the accuracy requirements for underwater target exploration.

Merging Data with Modeling: An Example from Fatigue.

It is well known that errors are inevitable in experimental observations, but it is equally unavoidable to eliminate errors in modeling the process leading to the experimental observations. If estimation and prediction are to be done with reasonable accuracy, the accumulated errors must be adequately managed. Research in fatigue is challenging because modeling can be quite complex. Furthermore, experimentation is time-consuming, which frequently yields limited data. Both of these exacerbate the magnitude of the potential error. The purpose of this paper is to demonstrate a procedure that combines modeling with independent experimental data to improve the estimation of the cumulative distribution function (cdf) for fatigue life. Subsequently, the effect of intrinsic error will be minimized. Herein, a simplified fatigue crack growth modeling is used. The data considered are a well-known collection of fatigue lives for an aluminum alloy. For lower applied stresses, the fatigue lives can range over an order of magnitude and up to 107 cycles. For larger applied stresses, the scatter in the lives is considerably reduced. Consequently, modeling must encompass a variety of conditions. The primary conclusion of the effort is that merging independent experimental data with a reasonably acceptable model vastly improves the accuracy of the calibrated cdfs for fatigue life, given the loading conditions. This allows for improved life estimation and prediction. For the aluminum data, the calibrated cdfs are shown to be quite good by using statistical goodness-of-fit tests, stress-life (S-N) analysis, and confidence bounds estimated using the mean square error (MSE) method. A short investigation into the effect of sample size is also included. Thus, the proposed methodology is warranted.

Domain Adaptation from Drilling to Geophysical Data for Mineral Exploration

This study utilizes domain adaptation to enhance the integration of diverse geoscience datasets, aiming to improve the identification of ore bodies. Traditional mineral exploration methods often face challenges in merging different geoscience data types, which leads to models that do not perform well across varying domains. Domain adaptation is a deep learning strategy aimed at adapting a model developed in one domain (source) to perform well in a different domain (target). To adapt models trained on detailed, labeled drilling data (source) to interpret broader, unlabeled geophysical data (target), Domain-Adversarial Neural Networks (DANNs) were applied, chosen for their robust performance in scenarios where the target domain does not provide labels. This approach was indirectly validated through the minimal overlap between regions identified as candidate ore and borehole locations marked as host rocks, with qualitative validation provided by t-Distributed Stochastic Neighbor Embedding (t-SNE) visualizations showing improved data integration across domains.

Outcomes of Concomitant Cardiac Surgical Procedures Performed During Pediatric Lung Transplantation in the United States.

Data on concomitant cardiac surgery (CCS) performed during pediatric lung transplantation (LTx) is limited. Therefore, we conducted a multi-institutional analysis to identify the incidence and outcomes of CCS in pediatric (< 18 years) LTx recipients by merging data (2004-2023) from the United Network for Organ Sharing (UNOS) and Pediatric Health Information System (PHIS) databases. Of the total of 596 pediatric LTx recipients, 87 (15%) underwent CCS. The majority of these cardiac surgeries were atrial septal defect (ASD) closure (90%) followed by aortic arch/descending aortic repair (3%), atrial repair (3%), ventricular septal defect closure (2%), patent ductus arteriosus ligation (2%), and tricuspid valve repair (2%). The median age at LTx was 3years (IQR: 0-12). Pulmonary hypertension (PHT) was the predominant indication for LTx (54%). Survival to discharge was 94% and 5-years survival was 64%. Our findings indicate CCS in children undergoing LTx has acceptable outcomes.