Incomplete Data Sets Research Articles

NIpple Position to Pinpoint Localization of Chest Drain Insertion in FEmale Trauma Patients: The NIPPLE-Trial—A Landmark Study

Background: The insertion of chest drains (CD) in trauma patients is a lifesaving, albeit high-risk intervention. Safe insertion of CD in settings where aids like ultrasound are not available relies on a landmark technique defining the so-called triangle of safety. The inferior margin of this triangle is identified by nipple height, which is thought to correspond to the fifth intercostal space (ICS). Training manikins are modeled after a lean male body and oftentimes insinuating jokes fuel uncertainty about the height of the nipple as a reliable landmark in female trauma patients. This study aims to prove that the nipple can be considered a safe landmark for CD insertion in women: even if larger breasts follow the force of gravity, it should not act towards the direction of the feet, but to the sides of the thorax in a flat-lying trauma patient. Methods: An online questionnaire was designed and distributed amongst female surgeons. Epidemiological data (age, nr of pregnancies, history of breastfeeding) was collected. Height, weight, and body mass index (BMI) were reported. Chest circumference and underbust girth were self-measured. The nipple position in relation to the ICS was measured by the women while lying in a supine position. Results: 237 female surgeons completed the questionnaire. Nine questionnaires were excluded due to incomplete data sets. 20 women were excluded due to previous breast surgery. Thus, 208 participants (or 416 nipples) were included in analysis. In supine position, both nipples were located in the 5th (±0.8) intercostal space. Age, BMI, height, weight, and status post-pregnancy had no significant impact on nipple position. Interestingly, a history of breastfeeding correlated with lower nipple position. Conclusions: The nipple is a safe landmark for identifying the correct insertion position of chest drains in supine-lying female trauma patients.

How Much Does Prosthetic Joint Infection and Its Successful Treatment Affect Patient-reported Quality of Life?

Uncomplicated joint replacement improves pain and other patient-reported outcome measures (PROMs) such as joint function scores and quality-of-life measures. However, the overall impact of periprosthetic joint infection (PJI) and its successful treatment on PROMs is poorly defined. In this study, we describe quality-of-life scores using the 12-item Short Form survey, version 2 (SF-12v2), collected as part of a large, prospective, observational study of PJI. (1) Do patients with newly diagnosed PJI have lower quality-of-life scores than the general age-matched population? (2) Are lower quality-of-life scores for patients with PJI sustained for ≥ 12 months? (3) What factors are associated with a good functional outcome at 12 months, defined by achieving a physical component summary (PCS) score of > 50 on the SF-12v2 (that is, above the age-adjusted population mean) or an increase of > 8.9 or more from baseline? The PIANO (Prosthetic joint Infection in Australia and New Zealand, Observational study) cohort was a prospective, longitudinal, multicenter cohort study of 783 patients with newly diagnosed PJI recruited across 27 centers between July 2014 and December 2017. All participants were followed for 2 years. SF-12v2 scores were collected at diagnosis (baseline) and 3, 12, and 24 months after diagnosis. Treatment success was defined as being alive with no clinical or microbiological evidence of infection and no ongoing use of antibiotics for the index joint. After exclusion of 6.6% (52) who had died, 4.2% (33) with PJI of joints other than hips and knees, and those with incomplete data sets (200 [25.5%]), 498 patients had complete SF-12v2 data sets available. At baseline, the median (IQR) PCS score was 37 (30 to 46), which increased to 41 (34 to 49; p < 0.001) at 12 months. Both measures were lower than those for the general age-matched population norm. There was no further improvement in PCS scores between 12 and 24 months. By contrast, the median (IQR) mental component summary score of the SF-12v2 was similar to age-adjusted population norms (48 [37 to 57]) at baseline. Only 40% of patients with PJI achieved a good functional outcome at 12 months. After adjustment for other factors, treatment success of PJI increased the odds of a good functional outcome by 67% (95% CI 11% to 154%; p = 0.02). PJIs have a large effect on the PCS score of the SF-12v2 quality-of-life measure at baseline, which is sustained for at least 2 years. Successful treatment of PJI increases the likelihood of a good functional outcome. These data can be used to set expectations for patients presenting with PJI and can be used to inform future clinical studies in which quality-of-life measures are incorporated into clinical endpoints. Level II, therapeutic study.

Multiple imputation integrated to machine learning: predicting post-stroke recovery of ambulation after intensive inpatient rehabilitation

Good data quality is vital for personalising plans in rehabilitation. Machine learning (ML) improves prognostics but integrating it with Multiple Imputation (MImp) for dealing missingness is an unexplored field. This work aims to provide post-stroke ambulation prognosis, integrating MImp with ML, and identify the prognostic influential factors. Stroke survivors in intensive rehabilitation were enrolled. Data on demographics, events, clinical, physiotherapy, and psycho-social assessment were collected. An independent ambulation at discharge, using the Functional Ambulation Category scale, was the outcome. After handling missingness using MImp, ML models were optimised, cross-validated, and tested. Interpretability techniques analysed predictor contributions. Pre-MImp, the dataset included 54.1% women, 79.2% ischaemic patients, median age 80.0 (interquartile range: 15.0). Post-MImp, 368 non-ambulatory patients on 10 imputed datasets were used for training, 80 for testing. The random forest (the validation best-performing algorithm) obtained 75.5% aggregated balanced accuracy on the test set. The main predictors included modified Barthel index, Fugl-Meyer assessment/motricity index, short physical performance battery, age, Charlson comorbidity index/cumulative illness rating scale, and trunk control test. This is among the first studies applying ML, together with MImp, to predict ambulation recovery in post-stroke rehabilitation. This pipeline reliably exploits the potential of incomplete datasets for healthcare prognosis, identifying relevant predictors.

A novel Tree-augmented Bayesian network for predicting rock weathering degree using incomplete dataset

The precise forecasting of the weathering degree of surrounding rock holds paramount importance for the scientific design and secure execution of tunnel engineering. The apparent features of the surrounding rock serve as critical indicators for evaluating its weathering degree. This paper endeavors to quantify the rock apparent features based on an improved Computer vision model and establish a multi-source heterogeneous dataset encompassing 10 parameters, thereby facilitating data-driven predictions of the weathering degree. Specifically, the rock appearance parameters are quantified and segmented by an improved Tunnel face feature segmentation (TFFSeg) model, which is tailored to the unique characteristics of groundwater, fractures, and interlayers. Concurrently, the TFFSeg model exhibits significantly enhanced performance for these rock features compared to other widely employed Computer vision methods. Subsequently, this multi-source dataset is further enriched by incorporating rock physical and mechanical parameters as well as tunnel design parameters. Nevertheless, the issue of data incompleteness persists within this dataset. To achieve precise prediction of the weathering degree based on this incomplete dataset, a novel Tree-augmented Bayesian network (TAN-BN) is designed, which is capable of learning from incomplete datasets. The predictive outcomes demonstrate that the proposed TAN-BN surpasses other currently utilized meta models and ensemble models, such as ANN, GBRT, and Naive BN. Finally, sensitivity analysis is conducted to determine the importance rankings of the 10 parameters, offering valuable insights for on-site evaluation of the rock weathering degree at the tunnel face.

Utilizing Artificial Intelligence to Reconcile Indeterminate Lupus Anticoagulant Panel Results

Abstract Lupus anticoagulant (LA) is one laboratory criterion for diagnosing antiphospholipid antibody syndrome (APS), a condition associated with hypercoagulability, pregnancy complications, and death. LA diagnosis requires a complex panel, often performed at a specialized laboratory. Ruling out false positive or transient LAs requires ≥ two positive tests ≥ 12 weeks apart. Most LA tests are negative, with small subsets resulting positive or indeterminate. Indeterminates leave clinicians with limited understanding of appropriate clinical follow-up. Thus, the International Society on Thrombosis and Hemostasis discourages weak resulting nomenclature such as “indeterminate.” Our study used artificial intelligence to predict LA indeterminate results as positive or negative, identifying patients requiring further testing. This IRB approved study reviewed nearly 4000 LA tests performed from 2019-2023. Parameters included CBC, thromboelastography, renal/hepatic function assays, coagulation tests, antiphospholipid antibodies and LA tests. LA test components included screening tests (APTT, Dilute Russel Viper Venom Test (DRVVT) screen), mixing studies (1:1 APTT mix, DRVVT mix), phospholipid-dependent confirmation (DRVVT, hexagonal phase, platelet neutralization), and factor VIII. A positive LA test fulfilled 4 criteria: at least one positive screening test, mixing study and phospholipid-dependent assay and factor VIII inhibitor exclusion. An LA panel meeting 2-3 criteria resulted indeterminate. Studies meeting &amp;lt;2 criteria were negative. This analysis was performed using Machine Intelligence Learning Optimizer (MILO) platform which displays the performance of multiple models. MILO was trained with two classifications: positive and negative cases. Data was split into training (100 training positives; 100 negatives) and testing (primary and secondary validation) sets (100 positives; 1388 negatives). LA panels affected by drug interactions (9.4%) or containing incomplete datasets (21%) were excluded. Sensitivity, specificity, negative predictive value (NPV) and positive predictive value (PPV) determined model efficacy. 2903 negative (84%), 318 positive (9.1%), and 246 indeterminate (7.1%) LA results were included from 3467 patients. Only 84 of the 246 indeterminate patients (34%) were retested. Of these, 28 remained indeterminate, 43 became negative and 13 resulted positive. The best performing MILO models were the K-nearest neighbor (KNN) and logistic regression. The ability of these models to determine which indeterminate cases would become positive or negative was evaluated. KNN displayed the highest specificity (0.88) and PPV (0.57). All models had a high NPV and specificity due to substantive exposure to negative cases for the training and testing. Exclusion of some parameters did not improve the performance of the models. However, even the best model (various KNN models) had low sensitivity and PPV due to a small pool of positive training/testing cases and a much smaller group of indeterminate LA with subsequent positive results. Creating models accurately predicting indeterminate cases that may become positive enables appropriate clinical follow-up. Thus, improving model accuracy will require a larger sample size of positive indeterminates with follow-up.

Leveraging Transfer Learning for Predicting Protein-Small Molecule Interactions.

A complex web of intermolecular interactions defines and regulates biological processes. Understanding this web has been particularly challenging because of the sheer number of actors in biological systems: ∼10 4 proteins in a typical human cell offer a plausible 10 8 interactions. This number grows rapidly if we consider metabolites, drugs, nutrients, and other biological molecules. The relative strength of interactions also critically affects these biological processes. However, the small and often incomplete datasets (10 3 -10 4 protein-ligand interactions) traditionally used for binding affinity predictions limit the ability to capture the full complexity of these interactions. To overcome this challenge, we developed Yuel 2, a novel neural network-based approach that leverages transfer learning to address the limitations of small datasets. Yuel 2 is pre-trained on a large-scale dataset to learn intricate structural features and then fine-tuned on specialized datasets like PDBbind to enhance the predictive accuracy and robustness. We show that Yuel 2 predicts multiple binding affinity metrics - Kd, Ki, and IC50 - between proteins and small molecules, offering a comprehensive representation of molecular interactions crucial for drug design and development.

Uncertainty in Fourier Transforms: A Fuzzy Logic Perspective

This study explores the integration of fuzzy logic with Fourier transforms to address the challenges of uncertainty, noise, and imprecision in real-world data. Fuzzy Fourier transforms extend traditional Fourier methods by incorporating fuzzy numbers, allowing for more robust frequency analysis, signal processing, and data reconstruction, particularly in noisy or incomplete datasets. The study examines the mathematical formulation of fuzzy Fourier transforms, computational trade-offs, and their performance advantages in comparison to classical Fourier methods. Real-world applications are discussed, including signal processing, image reconstruction, and time-series forecasting. Furthermore, the research highlights the increased computational complexity associated with fuzzy methods, the challenges of interpreting fuzzy results, and the limitations of handling probabilistic uncertainty. Future directions include the development of multi-dimensional fuzzy Fourier transforms, hybrid models integrating machine learning, and quantum computing applications. These advancements have broad implications for fields such as telecommunications, medical imaging, and financial forecasting, where handling uncertainty is critical.

Is Kinesiophobia Associated With Quality of Life, Level of Physical Activity, and Function in Older Adults With Knee Osteoarthritis?

Beyond knee pain itself, the fear of movement, also known as kinesiophobia, recently has been proposed as a potential factor contributing to disability and functional limitation in patients with knee osteoarthritis (OA). Nevertheless, the available evidence on the association of kinesiophobia with patient-reported outcome measures (PROMs) in knee OA remains limited. Among patients with nonoperatively treated knee OA, we asked: (1) Is kinesiophobia associated with decreased quality of life (QoL), functional outcomes, and physical activity? (2) What are the patient disease and psychosocial demographic factors associated with kinesiophobia? This was a multicenter, cross-sectional study of 406 general orthopaedic patients from two urban, referral-based tertiary hospitals in Singapore under a single healthcare group who received nonoperative treatment for knee OA. Between July 2020 and January 2022, a total of 1541 patients were treated for knee OA nonoperatively. Based on that, 60% (923) of patients were rejected due to refusal to participate in the study, 3% (52) of patients were enrolled but did not show up for their appointments for data collection, and a further 10% (160) had incomplete data sets, leaving 26% (406) for this study's analysis. The mean age of patients was 64 ± 8 years, 69% were women, and 81% were Chinese. The level of kinesiophobia in patients was measured using the Brief Fear of Movement scale, a validated 6-item questionnaire ranging from a score of 6 to 24 to measure kinesiophobia in OA, with higher scores representing higher levels of kinesiophobia. In terms of PROMs, the QoL and functional level of patients were measured using the QoL and activities of daily living (ADL) components of the widely validated 12-item Knee Injury and Osteoarthritis Outcome Score (KOOS-12). The KOOS-12 is a questionnaire consisting of 12 items encompassing three domains (QoL, ADL, and pain), with each item ranging from 0 to 4 and higher scores representing worse outcomes. The University of California, Los Angeles (UCLA) Activity Scale was used to measure the level of physical activity in patients. The UCLA score is a descriptive 10-level activity scale ranging from a score of 1 to 10, with higher scores representing greater physical activity levels. A directed acyclic graph, which is a relationship map used to depict and visualize the confounders between the studied variables, was used to identify the confounders between kinesiophobia and PROMs (QoL, function, and physical activity). An ordinal regression model was used to explore: (1) the association between kinesiophobia (as measured using the Brief Fear of Movement scale) and PROMs (as measured using KOOS QoL, KOOS ADL, and the UCLA Activity Scale), adjusting for key confounders such as age, gender, pain, side of arthritis, OA duration/severity, and psychosocial factors (for example, depression, anxiety, and education levels), and (2) the association between kinesiophobia (Brief Fear of Movement scale) and various patient disease and psychosocial demographic factors. After accounting for confounders, greater kinesiophobia (higher Brief Fear of Movement scores) was associated with lower QoL (KOOS QoL score adjusted IQR OR 0.69 [95% confidence interval (CI) 0.53 to 0.90]; p = 0.007) and lower physical activity (UCLA score adjusted IQR OR 0.68 [95% CI 0.52 to 0.90]; p = 0.007); however, there was no association between kinesiophobia and function (KOOS ADL score adjusted IQR OR 0.90 [95% CI 0.70 to 1.17]; p = 0.45). After adjusting for age, gender, OA duration, pain, and BMI, higher levels of anxiety (Patient Health Questionnaire 2 [PHQ-2] anxiety score adjusted OR 2.49 [95% CI 1.36 to 4.58]; p = 0.003) and depression (PHQ-2 depression score adjusted OR 3.38 [95% CI 1.73 to 6.62]; p < 0.001) were associated with higher levels of kinesiophobia. Education level, OA disease severity, side of arthritis (unilateral versus bilateral), and history of previous injury or surgery on the knee were not associated with kinesiophobia. Clinicians should assess for kinesiophobia and other psychological comorbidities such as depression and anxiety at the point of initial evaluation and subsequent follow-up of knee OA with simple validated tools like the Brief Fear of Movement scale (kinesiophobia) in the clinic. This allows for clinicians to identify high-risk individuals and offer evidence-based treatment such as cognitive behavioral therapies with a multidisciplinary team, including a physical therapist and psychologist, to manage these psychological comorbidities and improve outcomes in patients with knee OA. While kinesiophobia was found to be associated with poorer QoL and physical activity, future studies including larger observational cohort studies should be conducted to determine causal and prognostic relationships between kinesiophobia and outcomes in knee OA. Level III, prognostic study.

Convertible glenoid replacement in the anatomical total shoulder arthroplasty: medium-term results.

The older generation of high thickness metal-backed glenoid prostheses had a high failure rate. The goal of our study was to analyze the medium-term clinical and radiological results of a technically modified generation of metal backed glenoids with 35% decrease in thickness. Thirty-nine patients (43 shoulders) with a mean age of 66.5 years at surgery were examined from two to ten years (mean 71.2 months) using the Constant-Score, Subjective Shoulder Value and X-ray in a.p. and axial view. Indication for surgery included patients with glenoid types B1, B2, or B3 and/or posterior subluxation but intact rotator cuff, or patients who did not want a reverse prosthesis and had high functional demands. Inclusion criteria comprised participation in a clinical and radiological follow-up examination, along with a minimum follow-up duration of 24 months. Additionally, in all cases, the Constant Score and the Subjective Shoulder Value (PROM) had to be determined. Exclusion criteria were an incomplete dataset, implantation of a different prosthesis type, and failure to meet the minimum follow-up time. The absolute Constant Score (CS) improved significantly (p < 0.0001) from 43 ± 13 (range: 20-69) points before to 80 ± 13 (range: 46-98) points after surgery. The only preoperative negative influencing factor on the result was the glenoid protrusion according to Lévigne (p = 0.0150). No significant differences in functional outcomes were found between glenoid types A2,B1,B2 and B3. Radiolucent lines at the interface were observed in a total of four patients (13%). Radiological signs of glenoid osteolysis were seen in six patients (20%).Humeral upward migration was observed in four patients (13%) and posterior shoulder joint decentering occurred in one patient (3%). Implant-specific complications occurred in three cases (7%). Due to a massive rotator cuff tear, two cases (4.7%) had to be converted to a reversed implant, with the primary metal component remaining in place. If the implant-specific and rotator cuff-associated complications are summarized, the rate is 11.7%. All components that required conversion to an reverse implant have so far remained free of complications. Favorable clinical results can be achieved, especially in patients with an eccentric glenoid type. Severe preoperative glenoid erosion (Walch A2, B2, B3) does not appear to have any influence on postoperative functional results. The complication and revision rates are significantly better than in previous studies with conventional metal-backed glenoid components. Conversion of the anatomical glenoid component to a reverse system was always possible and good clinical results were achieved.

DINEOF Interpolation of Global Ocean Color Data: Error Analysis and Masking

Abstract The Data Interpolation Empirical Orthogonal Function (DINEOF) algorithm is used to reconstruct datasets of geophysical and biological variables such as sea surface temperature (SST) and chlorophyll a (Chl a). In this study, we analyze the impact of both the quantity and distribution of missing data on the performance of DINEOF demonstrating how DINEOF plus a connectivity mask can be used for future data reconstruction tasks. We propose an enhanced version of DINEOF (DINEOF+) by adding two steps: 1) Using a 75% threshold of missing data for reconstructing incomplete datasets and 2) masking interpolated points that lack sufficient space–time observations in the original dataset. We successfully apply DINEOF+ to the Ocean Color Climate Change Initiative (OC-CCI) global daily Chl a dataset and validate the results using in situ datasets. We find that the recovery rate varies across ocean basins and years. In oligotrophic waters, the daily data coverage increased by 40%–50% during the period from 2003 to 2020. Using DINEOF+ allows us to obtain a significantly higher temporal resolution of global Chl a data, which will improve understanding of marine phytoplankton dynamics in response to changing environments. Significance Statement We perform an error analysis on the application of DINEOF for reconstructing a global Chl a dataset. The results of this analysis illustrate the impact of missing data—both in terms of quantity and distribution—on the performance of DINEOF. We propose using DINEOF+, an enhanced version of DINEOF that adds an editing step to mask out interpolated points based on the number of surrounding observations in the original input. The performance of DINEOF+ was validated using both simulated and in situ datasets. The results indicate that employing this masking technique effectively reduces biased estimates of missing data. DINEOF+ can be applied to other biogeochemical variables. However, caution is advised when dealing with observations characterized by high variance.

Framework to improve software effort estimation accuracy using novel ensemble rule

This investigation focuses on refining software effort estimation (SEE) to enhance project outcomes amidst the rapid evolution of the software industry. Accurate estimation is a cornerstone of project success, crucial for avoiding budget overruns and minimizing the risk of project failures. The framework proposed in this article addresses three significant issues that are critical for accurate estimation: dealing with missing or inadequate data, selecting key features, and improving the software effort model. Our proposed framework incorporates three methods: the Novel Incomplete Value Imputation Model (NIVIM), a hybrid model using Correlation-based Feature Selection with a meta-heuristic algorithm (CFS-Meta), and the Heterogeneous Ensemble Model (HEM). The combined framework synergistically enhances the robustness and accuracy of SEE by effectively handling missing data, optimizing feature selection, and integrating diverse predictive models for superior performance across varying project scenarios. The framework significantly reduces imputation and feature selection overhead, while the ensemble approach optimizes model performance through dynamic weighting and meta-learning. This results in lower mean absolute error (MAE) and reduced computational complexity, making it more effective for diverse software datasets. NIVIM is engineered to address incomplete datasets prevalent in SEE. By integrating a synthetic data methodology through a Variational Auto-Encoder (VAE), the model incorporates both contextual relevance and intrinsic project features, significantly enhancing estimation precision. Comparative analyses reveal that NIVIM surpasses existing models such as VAE, GAIN, K-NN, and MICE, achieving statistically significant improvements across six benchmark datasets, with average RMSE improvements ranging from 11.05% to 17.72% and MAE improvements from 9.62% to 21.96%. Our proposed method, CFS-Meta, balances global optimization with local search techniques, substantially enhancing predictive capabilities. The proposed CFS-Meta model was compared to single and hybrid feature selection models to assess its efficiency, demonstrating up to a 25.61% reduction in MSE. Additionally, the proposed CFS-Meta achieves a 10% (MAE) improvement against the hybrid PSO-SA model, an 11.38% (MAE) improvement compared to the Hybrid ABC-SA model, and 12.42% and 12.703% (MAE) improvements compared to the hybrid Tabu-GA and hybrid ACO-COA models, respectively. Our third method proposes an ensemble effort estimation (EEE) model that amalgamates diverse standalone models through a Dynamic Weight Adjustment-stacked combination (DWSC) rule. Tested against international benchmarks and industry datasets, the HEM method has improved the standalone model by an average of 21.8% (Pred()) and the homogeneous ensemble model by 15% (Pred()). This comprehensive methodology underscores our model’s contributions to advancing software project management (SPM) through advanced predictive modeling, setting a new benchmark for software engineering effort estimation.

A Contemporary Survey on Multisource Information Fusion for Smart Sustainable Cities: Emerging Trends and Persistent Challenges

The emergence of smart sustainable cities has unveiled a wealth of data sources, each contributing to a vast array of urban applications. At the heart of managing this plethora of data is multisource information fusion (MSIF), a sophisticated approach that not only improves the quality of data collected from myriad sources, including sensors, satellites, social media, and citizen-generated content, but also aids in generating actionable insights crucial for sustainable urban management. Unlike simple data fusion, MSIF excels in harmonizing disparate data sources, effectively navigating through their variability, potential conflicts, and the challenges posed by incomplete datasets. This capability is essential for ensuring the integrity and utility of information, which supports comprehensive insights into urban systems and effective planning. This survey combines hierarchical and multi-dimensional classification to examine how MSIF integrates and analyses diverse datasets, enhancing the operational efficiency and intelligence of urban environments. It highlights the most significant challenges and opportunities presented by MSIF in smart sustainable cities, particularly how it overcomes the limitations of existing approaches in scope and coverage.By considering social, economic, and environmental factors, MSIF offers a multidisciplinary approach that is pivotal for advancing sustainable urban development. Recognized as an essential resource for academics and practitioners, this study promotes a new wave of MSIF innovations aimed at improving the cohesion, efficiency, and sustainability of smart cities.

Composite fault diagnosis of gearbox based on deep graph residual convolutional network

In gearbox systems, a composite fault diagnosis resulting from mutual interference among different components poses a significant challenge. The traditional composite fault diagnosis methods based on conventional signal analyses and feature extractions often suffer from low sensitivity to fault characteristics and difficulty in effectively identifying composite faults. On the other hand, composite fault diagnosis research via deep learning and data-driven approaches typically faces issues such as incomplete training datasets and insufficient exploration of feature correlation information, leading to an underutilization of the fault information. Therefore, this paper proposes a deep graph residual convolutional neural network (DGRCN) based on feature correlation mining for composite fault diagnosis in gearboxes. First, Pearson correlation coefficients are utilized to explore the relationships among features in the traditional feature set, transforming these relationships into a graph-structured feature set. Next, a deep graph residual convolutional network is constructed by integrating deep graph structures into a residual framework. This network globally extracts composite fault subgraph features and explores local feature correlations. Finally, the model is trained via various composite fault datasets under complex working conditions, achieving the diagnosis and identification of composite faults under the constraint of limited samples. The experimental results demonstrate that the proposed method significantly improves composite fault diagnosis accuracy, outperforming commonly used methods in this field.

Using NHS Data to Monitor Trends in the Occurrence of Severe, Food-Induced Allergic Reactions Work Package 1

People with food allergies may experience food allergic reactions due to accidental exposure. These reactions are commonly categorised as non-severe, fatal food anaphylaxis and near-fatal food anaphylaxis. Non-severe allergic reactions to food are more common with an incidence of up to 1,000 times greater than fatal food-related anaphylaxis. However, obtaining accurate data relating to the circumstances under which these reactions occurred is challenging under the current diagnosis coding system used in the National Health Service (NHS). This project addressed two key questions: 1. What are the trends in the occurrence of food hypersensitivity (FHS) reactions and their consequences in terms of healthcare encounters (both to hospital and primary care)? 2. What are the circumstances surrounding severe, life-threatening reactions to food? Approach The researchers used existing NHS datasets (relating to Hospital Admissions, Accident &amp; Emergency visits, Critical Care admissions and Primary care visits) to evaluate healthcare encounters due to food hypersensitivity over the study period. Patient pathways through the healthcare system were also assessed by linking these different datasets. Key Results • Food-induced anaphylaxis represented 29.4% of reported anaphylaxis. admissions, and increased significantly from 1.23 to 4.02 admissions per 100,000 population per annum over the study period. • However, despite an annual increase of 5.7% in hospitalisation for food-induced anaphylaxis between 1998 and 2018, the case fatality rate (proportion of hospital admissions associated with a fatal outcome) more than halved, from 0.7% in 1998 to less than 0.3% in 2018. • 152 deaths were identified during the study period where the cause was very likely to have been food-induced anaphylaxis. • At least 86 (46%) fatalities were triggered by peanut or tree nuts. • Cows’ milk was reported to be the most common cause of fatal anaphylaxis in children aged under 16 years. • Using data from England for the period 2008-2018, the prevalence of food allergy ranged from 4% in preschool-aged children (under 5 years), 1-2% in school-aged children and young people (5 to 20 years) and 0.9% in adults. Although the same methods were used throughout the study period, the statistics presented may be prone to limitations such as miscoding and incomplete datasets. • Prescription of adrenaline auto-injector devices (AAI) are an important risk management intervention in people at risk of food-induced anaphylaxis. However, there is significant under-prescribing of AAI. Data showed that 40% of individuals with prior food-induced anaphylaxis were not prescribed AAI and at least 59% did not have AAI on repeat prescription. • Most healthcare visits for food allergy occurred in general practice. Less than 3% of individuals with a diagnosis of food allergy attended Accident and Emergency during the study period, 2008-2018. Therefore, using hospital data in isolation to analyse patterns of health service utilisation with respect to FHS may not provide a comprehensive overview.

Development of a site-screening method for hydrogen storage purposes and its application to an industrial dataset of Italian hydrocarbon reservoirs

A methodology is presented for the objective and transparent screening of hydrocarbon reservoirs for underground hydrogen storage (UHS), with subsequent testing on a large confidential dataset provided by the energy company Eni. The procedure uses the Analytic Hierarchy Process and the Delphi technique to gather expert opinions and weight 27 screening parameters in terms of: health, safety, and environment; geotechnical performance; and economic performance. A set of scores is produced that characterizes each site in terms of these three categories, as well as a comprehensive site ranking based on their overall suitability for UHS. The results highlight the importance of geotechnical parameters, while the characterization of faults and hydrocarbon type, the onshore or offshore location, the number of wells, and the reservoir architecture yielded the highest individual weights. Potential scores are also estimated for sites with incomplete datasets. Two blind tests evaluated the method's effectiveness against preexisting industrial assessments.

Correction of preferred orientation-induced distortion in cryo-electron microscopy maps.

Reconstruction maps of cryo-electron microscopy (cryo-EM) exhibit distortion when the cryo-EM dataset is incomplete, usually caused by unevenly distributed orientations. Prior efforts had been attempted to address this preferred orientation problem using tilt-collection strategy and modifications to grids or to air-water interfaces. However, these approaches often require time-consuming experiments, and the effect was always protein dependent. Here, we developed a procedure containing removing misaligned particles and an iterative reconstruction method based on signal-to-noise ratio of Fourier component to correct this distortion by recovering missing data using a purely computational algorithm. This procedure called signal-to-noise ratio iterative reconstruction method (SIRM) was applied on incomplete datasets of various proteins to fix distortion in cryo-EM maps and to a more isotropic resolution. In addition, SIRM provides a better reference map for further reconstruction refinements, resulting in an improved alignment, which ultimately improves map quality and benefits model building.

Benchmarking imputation methods for categorical biological data

Abstract Trait datasets are at the basis of a large share of ecology and evolutionary research, being used to infer ancestral morphologies, quantify species extinction risks, or evaluate the functional diversity of biological communities. These datasets, however, are often plagued by missing data, for instance, due to incomplete sampling, limited data and resource availability. Several imputation methods exist to predict missing values and recent studies have explored their performance for continuous traits in biological datasets. However, less is known about the accuracy of these methods for categorical traits. Here we explore the performance of different imputation methods on categorical biological traits combining phylogenetic comparative methods, machine learning and deep learning models. To this end, we develop an open‐source R package, to impute trait data while integrating a simulation framework to evaluate their performance on synthetic datasets. We run a range of simulations under different missing rates, mechanisms, biases and evolutionary models. We propose an integration between phylogenetic comparative methods and machine learning imputation, and an ensemble approach, in which selected imputation methods are combined. Our simulations show that this approach provides the most robust and accurate predictions. We applied our imputation pipeline to an incomplete trait dataset of 1015 elasmobranch species (i.e. sharks, rays and skates) and found a high imputation accuracy of the predictions based on an expert‐based assessment of the missing traits. Overall, our R package facilitates the comparison of multiple imputation methods and allows robust predictions of missing trait values. Our study highlights the benefits of coupling phylogenetic evolutionary models with machine learning inference to augment incomplete biological datasets.

Unsupervised Imputation of Non-Ignorably Missing Data Using Importance-Weighted Autoencoders

Deep Learning (DL) methods have dramatically increased in popularity in recent years. While its initial success was demonstrated in the classification and manipulation of image data, there has been significant growth in the application of DL methods to problems in the biomedical sciences. However, the greater prevalence and complexity of missing data in biomedical datasets present significant challenges for DL methods. Here, we provide a formal treatment of missing data in the context of Variational Autoencoders (VAEs), a popular unsupervised DL architecture commonly used for dimension reduction, imputation, and learning latent representations of complex data. We propose a new VAE architecture, NIMIWAE, that is one of the first to flexibly account for both ignorable and non-ignorable patterns of missingness in input features at training time. Following training, samples can be drawn from the approximate posterior distribution of the missing data can be used for multiple imputation, facilitating downstream analyses on high dimensional incomplete datasets. We demonstrate through statistical simulation that our method outperforms existing approaches for unsupervised learning tasks and imputation accuracy. We conclude with a case study of an EHR dataset pertaining to 12,000 ICU patients containing a large number of diagnostic measurements and clinical outcomes, where many features are only partially observed.

Observational study of missing SOFA score data frequency in RCTs relative to ICU length of stay

The Sequential Organ Failure Assessment, also known as SOFA score, was introduced to assess organ dysfunction of critical ill patients. However, understanding the impact of missing SOFA scores in randomized controlled trials and how this affect the validity and applicability of the SOFA score as a surrogate endpoint for predicting mortality has been a matter of interest. To address this, a secondary analysis of a systematic review was conducted to quantify the relationship between SOFA scores and the prediction of mortality in critically ill adults in randomized controlled trials (RCTs). The systematic review being referred to included 87 RCTs with a total of 12,064 critically ill patients. This analysis focused on missing SOFA score data in relation to the length of stay on the intensive care unit (ICU) and the methods used to handle missing data. SOFA score measurements from the included studies were categorized into three time frames: Early (t ≤ 4 days), Intermediate (t = 5–10 days) and Late (t > 10 days) measurement. Only one study reported a complete data set for calculating the SOFA score for an Early measurement. When considering all methods used to address missing data, 32% of studies still had missing data for Early measurements, and this percentage increased to 64% for Late measurements. These findings suggested that, over time, the number of studies with incomplete data sets has been increasing. The longer a patient is treated on the ICU, the higher the number of missing data which can impact the validity of SOFA score analyses. There was no clear trend towards a specific method for compensating missing data. An exemplary calculation demonstrated that ignoring missing data may lead to an underestimated variability of the treatment effect. This, in turn, could bias the interpretation of study results by policy- and clinical decision-makers. Overall, there are several limitations that need to be considered when using SOFA score as a surrogate endpoint for mortality. When employed as an outcome, the SOFA score is frequently missing and most studies do not adequately describe the amount or nature of missing data, or the methods used to handle missing data in the analysis.

Ultrasound-guided Hydrogel Injection Provides Better Therapeutic Effects After Hand Tendon Surgery Than Intraoperative Injection: A Randomized Controlled Trial.

Hydrogels are used to provide a barrier against peritendinous adhesion formation, but when implanted intraoperatively, they degrade rapidly and aggravate early inflammatory pain. It is uncertain whether clinical efficacy can be improved by avoiding the inflammatory phase when hydrogels are delivered during adhesion formation. (1) Compared with intraoperative hydrogel application, does ultrasound-guided postoperative application result in better total active motion (TAM) at 12 months after tendon injury? (2) Does ultrasound-guided postoperative application of hydrogels result in lower pain, better function, and better satisfaction? This open-label, prospective, single-center, randomized controlled trial was conducted by reparative and reconstructive surgeons at the National Orthopedics Clinical Medical Center, Shanghai, People's Republic of China. Between May 2021 and December 2022, 53% (168 of 317) of patients who met our inclusion criteria were recruited, and 47% (149 of 317) of patients were excluded because of the exclusion criteria. Finally, 84 patients were randomized to the postoperative group to receive ultrasound-guided carboxymethyl chitosan (CMC) hydrogel delayed injection, and 84 patients were randomized to the intraoperative group to receive CMC hydrogel intraoperative application. Another 8% (7 of 84) of patients in the postoperative group and 10% (8 of 84) of patients in the intraoperative group were lost before the minimum study follow-up time of 1 year or had incomplete datasets, leaving 91% (153 of 168) of patients with data for analysis. Data on outcome events were analyzed according to the intention-to-treat principle, which included all patients who underwent randomization. Follow-up visits were completed at 3 weeks, 6 weeks, 3 months, 6 months, and 12 months after tendon repair. The primary outcome was TAM (ie, the sum of the degrees of active metacarpophalangeal joint, proximal interphalangeal joint, and distal interphalangeal joint flexion less the degrees from full extension; minimum clinically important difference [MCID] 20°) at 12 months. Secondary outcomes included pain (measured with a VAS; range 0 to 10, a higher score indicating worse pain; MCID 0.6), Michigan Hand Outcomes Questionnaire activities of daily living (MHQ-ADL) score (range 0 to 100, a higher score indicating better outcomes; MCID 10.1), and MHQ satisfaction (MHQ-SAT) score (range 0 to 100, a higher score indicating better outcomes; MCID 33.0). At 12 months, the ultrasound-guided postoperative injection group had improved TAM (intraoperative 189° [95% CI 179° to 199°] versus postoperative 209° [95% CI 199° to 219°], mean difference 20° [95% CI 6° to 35°]; p = 0.006; the mean difference in the primary outcome fulfilled the MCID value at all time points). At 6 weeks, we found no clinically important difference in VAS pain scores among groups (intraoperative mean ± SD 2.0 ± 1.0 versus postoperative 1.7 ± 1.0, mean difference 0.3 [95% CI 0.1 to 0.7]; p = 0.02); however, at 3 weeks, the VAS pain scores showed clinically important difference among groups (3.6 ± 1.4 versus 2.9 ± 1.2, mean difference 0.7 [95% CI 0.3 to 1.1]; p = 0.001). At 3 months, the ultrasound-guided postoperative injection group had higher MHQ-ADL scores (intraoperative 62 ± 10 versus postoperative 75 ± 10, mean difference 13 [95% CI 11 to 17]; p < 0.001), and the mean difference of MHQ-ADL scores reached the MCID value at all time points. At 3 months, there was no clinically important difference in MHQ-SAT scores between groups (intraoperative 62 ± 8 versus postoperative 70 ± 8, mean difference 8 [95% CI 6 to 11]; p < 0.001). Compared with intraoperative CMC hydrogel injection, postoperative ultrasound-guided injection improved the TAM and function of the affected limb, showed a short-term pain control effect, and did not increase the risk of complications. Clinical trials are needed to confirm the safety and efficacy of ultrasound-guided postoperative injection of CMC hydrogels and to determine the most effective dose and the health and economic benefits of treatment. Level I, therapeutic study.