Normative Data of Liquid Gastric Emptying and Small-bowel Transit: A Prospective Cross-sectional Study.

Purpose of the Study: The primary objective was to establish the reference values for liquid gastric emptying and small bowel. The secondary objectives encompassed comparing the anterior view and geometric mean methods, assessing gender differences, and exploring potential correlations with age. Materials and Methods: Thirty-five consecutive healthy participants (28 females and 7 males) with a mean age of 42 ± 11 years (median, 42; range, 23-65) underwent liquid gastric emptying scintigraphy at five intervals (0, 0.5, 1, 2, and 4 h), with optional additional imaging at 24 h. Liquid gastric emptying was evaluated using percent retention and half-time (T1/2). Small-bowel transit was assessed using the index of small-bowel transit (ISBT), calculated as the ratio of terminal ileal reservoir counts to total abdominal counts at 4 h. Reference values were established based on percentiles or mean and standard deviation (SD). Rapid small-bowel transit was determined through visual inspection. Statistical analysis involved paired Samples t-test or Wilcoxon signed-rank test for comparing imaging methods, independent Samples t-test or Mann-Whitney U-test for gender comparison, and Spearman's rank correlation for assessing age-related associations. A 2-tailed P < 0.05 indicated significance. Results: Rapid liquid gastric emptying based on the geometric mean method was defined as percent retention <8% at 30 min, while delayed emptying as percent retention >33%, >20%, and >4% at 1, 2, and 4 h, respectively. The reference range of T1/2 of gastric emptying was 10-60 min. The reference value for small-bowel transit using the geometric mean method was established as ISBT >30% at 4 h, while rapid small-bowel transit was defined as the first visualization of activity in the cecum-ascending colon within 1 h. Parameters for liquid gastric emptying and small-bowel transit were notably higher in the anterior view method compared to the geometric mean method (P ≤ 0.019), except for percent retention at 2 h (P = 0.510). Nevertheless, the obtained reference values, whether based on percentiles or mean and SD, showed no notable variance between the two methods to warrant clinical significance. Gender did not display an impact on liquid gastric emptying or small-bowel transit in either method (P ≥ 0.173), and age demonstrated no significant moderate or strong correlations (Spearman's ρ ≤ 0.397). Conclusion: The study determined reference values for liquid gastric emptying and small-bowel transit through a standard gastric emptying scintigraphy protocol, avoiding additional complex procedures or extended imaging sessions. The established normative data can apply to individuals of both genders aged ≥18 years. While advocating the geometric mean method as the primary choice, the study acknowledges that in busy centers handling multiple studies with limited resources and a single-head gamma camera catering to multiple studies, the anterior view method remains a feasible alternative.

Inhibition of Renin Expression Is Regulated by an Epigenetic Switch From an Active to a Poised State.

BACKGROUND: Renin-expressing cells are myoendocrine cells crucial for the maintenance of homeostasis. Renin is regulated by cAMP, p300 (histone acetyltransferase p300)/CBP (CREB-binding protein), and Brd4 (bromodomain-containing protein 4) proteins and associated pathways. However, the specific regulatory changes that occur following inhibition of these pathways are not clear. METHODS: We treated As4.1 cells (tumoral cells derived from mouse juxtaglomerular cells that constitutively express renin) with 3 inhibitors that target different factors required for renin transcription: H-89-dihydrochloride, PKA (protein kinase A) inhibitor; JQ1, Brd4 bromodomain inhibitor; and A-485, p300/CBP inhibitor. We performed ATAC-seq, single-cell RNA sequencing, CUT&Tag, and chromatin immunoprecipitation sequencing for H3K27ac and p300 binding on biological replicates of treated and control As4.1 cells. RESULTS: In response to each inhibitor, Ren1 expression was significantly reduced and reversible upon washout. Chromatin accessibility at the Ren1 locus did not markedly change but was globally reduced at distal elements. Inhibition of PKA led to significant reductions in H3K27ac and p300 binding specifically within the Ren1 super-enhancer region. Further, we identified enriched TF (transcription factor) motifs shared across each inhibitory treatment. Finally, we identified a set of 9 genes with putative roles across each of the 3 renin regulatory pathways and observed that each displayed differentially accessible chromatin, gene expression, H3K27ac, and p300 binding at their respective loci. CONCLUSIONS: Inhibition of renin expression in cells that constitutively synthesize and release renin is regulated by an epigenetic switch from an active to poised state associated with decreased cell-cell communication and an epithelial-mesenchymal transition. This work highlights and helps define the factors necessary for renin cells to alternate between myoendocrine and contractile phenotypes.

Benchmarking predictive methods for small-angle X-ray scattering from atomic coordinates of proteins using maximum likelihood consensus data.

Stimulated by informal conversations at the XVII International Small Angle Scattering (SAS) conference (Traverse City, 2017), an international team of experts undertook a round-robin exercise to produce a large dataset from proteins under standard solution conditions. These data were used to generate consensus SAS profiles for xylose isomerase, urate oxidase, xylanase, lysozyme and ribonuclease A. Here, we apply a new protocol using maximum likelihood with a larger number of the contributed datasets to generate improved consensus profiles. We investigate the fits of these profiles to predicted profiles from atomic coordinates that incorporate different models to account for the contribution to the scattering of water molecules of hydration surrounding proteins in solution. Programs using an implicit, shell-type hydration layer generally optimize fits to experimental data with the aid of two parameters that adjust the volume of the bulk solvent excluded by the protein and the contrast of the hydration layer. For these models, we found the error-weighted residual differences between the model and the experiment generally reflected the subsidiary maxima and minima in the consensus profiles that are determined by the size of the protein plus the hydration layer. By comparison, all-atom solute and solvent molecular dynamics (MD) simulations are without the benefit of adjustable parameters and, nonetheless, they yielded at least equally good fits with residual differences that are less reflective of the structure in the consensus profile. Further, where MD simulations accounted for the precise solvent composition of the experiment, specifically the inclusion of ions, the modelled radius of gyration values were significantly closer to the experiment. The power of adjustable parameters to mask real differences between a model and the structure present in solution is demonstrated by the results for the conformationally dynamic ribonuclease A and calculations with pseudo-experimental data. This study shows that, while methods invoking an implicit hydration layer have the unequivocal advantage of speed, care is needed to understand the influence of the adjustable parameters. All-atom solute and solvent MD simulations are slower but are less susceptible to false positives, and can account for thermal fluctuations in atomic positions, and more accurately represent the water molecules of hydration that contribute to the scattering profile.

Comparative analysis of integrative classification methods for multi-omics data.

Recent advances in sequencing, mass spectrometry, and cytometry technologies have enabled researchers to collect multiple 'omics data types from a single sample. These large datasets have led to a growing consensus that a holistic approach is needed to identify new candidate biomarkers and unveil mechanisms underlying disease etiology, a key to precision medicine. While many reviews and benchmarks have been conducted on unsupervised approaches, their supervised counterparts have received less attention in the literature and no gold standard has emerged yet. In this work, we present a thorough comparison of a selection of six methods, representative of the main families of intermediate integrative approaches (matrix factorization, multiple kernel methods, ensemble learning, and graph-based methods). As non-integrative control, random forest was performed on concatenated and separated data types. Methods were evaluated for classification performance on both simulated and real-world datasets, the latter being carefully selected to cover different medical applications (infectious diseases, oncology, and vaccines) and data modalities. A total of 15 simulation scenarios were designed from the real-world datasets to explore a large and realistic parameter space (e.g. sample size, dimensionality, class imbalance, effect size). On real data, the method comparison showed that integrative approaches performed better or equally well than their non-integrative counterpart. By contrast, DIABLO and the four random forest alternatives outperform the others across the majority of simulation scenarios. The strengths and limitations of these methods are discussed in detail as well as guidelines for future applications.

Analysis of Microbiome Data.

The microbiome represents a hidden world of tiny organisms populating not only our surroundings but also our own bodies. By enabling comprehensive profiling of these invisible creatures, modern genomic sequencing tools have given us an unprecedented ability to characterize these populations and uncover their outsize impact on our environment and health. Statistical analysis of microbiome data is critical to infer patterns from the observed abundances. The application and development of analytical methods in this area require careful consideration of the unique aspects of microbiome profiles. We begin this review with a brief overview of microbiome data collection and processing and describe the resulting data structure. We then provide an overview of statistical methods for key tasks in microbiome data analysis, including data visualization, comparison of microbial abundance across groups, regression modeling, and network inference. We conclude with a discussion and highlight interesting future directions.

Deciphering the molecular landscape of ionising radiation-induced eye damage with the help of genomic data mining.

Even at low levels, exposure to ionising radiation can lead to eye damage. However, the underlying molecular mechanisms are not yet fully understood. We aimed to address this gap with a comprehensive in silico approach to the issue. For this purpose we relied on the Comparative Toxicogenomics Database (CTD), ToppGene Suite, Cytoscape, GeneMANIA, and Metascape to identify six key regulator genes associated with radiation-induced eye damage (ATM, CRYAB, SIRT1, TGFB1, TREX1, and YAP1), all of which have physical interactions. Some of the identified molecular functions revolve around DNA repair mechanisms, while others are involved in protein binding, enzymatic activities, metabolic processes, and post-translational protein modifications. The biological processes are mostly centred on response to DNA damage, the p53 signalling pathway in particular. We identified a significant role of several miRNAs, such as hsa-miR-183 and hsamiR-589, in the mechanisms behind ionising radiation-induced eye injuries. Our study offers a valuable method for gaining deeper insights into the adverse effects of radiation exposure.

Comparison of Enhanced and Standard Data Sheets on Treatment Fidelity and Data Collection for Tact Training.

Specifically designed data sheets have been recommended to assist with the fidelity of implementation of treatment procedures. The present study extended previous research (e.g., Bottini et al. Behavior Analysis: Research & Practice 21(2), 140-152, 2021; LeBlanc et al. Behavior Analysis in Practice 13(1), 53-62, 2020) by comparing an enhanced data sheet (i.e., the inclusion of randomized targets, prompts for treatment components of securing attending and reinforcement) to a standard data sheet (i.e., targets not preset, no prompts for treatment components) on the fidelity of tact training of features. Ten behavior therapists participated in each condition (n = 20). Participants first watched a brief instructional video explaining the teaching procedure and their assigned data sheet, followed by conducting a treatment session with a confederate serving as a child with autism spectrum disorder. The enhanced data sheet resulted in higher fidelity on multiple variables including randomizing of targets and data collection.

Development of the CDISC Pediatrics User Guide: a CDISC and conect4children collaboration.

Introduction: The conect4children (c4c) project aims to facilitate efficient planning and delivery of paediatric clinical trials. One objective of c4c is data standardization and reuse. Interoperability and reusability of paediatric clinical trial data is challenging due to a lack of standardization. The Clinical Data Interchange Standards Consortium (CDISC) standards that are required or recommended for regulatory submissions in several countries lack paediatric specificity with limited awareness within academic institutions. To address this, c4c and CDISC collaborated to develop the Pediatrics User Guide (PUG) consisting of cross-cutting data items that are routinely collected in paediatric clinical trials, factoring in all paediatric age ranges. Methods and Results: The development of the PUG consisted of six stages. During the scoping phase, subtopics (each containing several clinically relevant concepts) were suggested and debated for inclusion in the PUG. Ninety concepts were selected for the modelling phase. Concept maps describing the Research Topic and representation procedure were developed for the 19 concepts that had no (or partial) previous modelling in CDISC. Next, metadata and implementation examples were developed for concepts. This was followed by a CDISC internal review and a public review. For both these review stages, the feedback comments were either implemented or rejected based on budget, timelines, expert review, and scope. The PUG was published on the CDISC website on February 23, 2023. Discussion: The PUG is a first step in bridging the lack of child specific CDISC standards, particularly within academia. Several academic and industrial partners were involved in the development of the PUG, and c4c has undertaken multiple steps to publicize the PUG within its academic partner organizations - in particular, the European Reference Networks (ERNs) that are developing registries and dictionaries in 24 disease areas. In the long term, continued use of the PUG in paediatric clinical trials will enable the pooling of data from multiple trials, which is particularly important for medical domains with small populations.

Scabies in the Philippines: A Secondary Analysis of Local Patient Registries.

Background: Scabies is the second most common cause of disability among skin diseases in the Philippines as of 2019. There is no large nationwide study describing the epidemiologic profile of scabies in the country. Objective: This study aimed to describe the demographic, seasonal, and geographic profile of scabies in the Philippines. Methods: We compared secondary data of two local patient registries (Philippine Dermatological Society, PDS, 2010 to 2021; and Philippine Pediatric Society, PPS, 2009 to 2021) for reported cases of scabies in the Philippines. We reported the frequency and percentage distribution according to age, sex, month, year, and type of diagnosis, and region. Results: The median annual frequency of scabies cases (mostly outpatient) for PDS (from year 2010) was 4087 (range ([QR], 342-6422 [3271.5]), while it was 183 (range [IQR], 64-234 [96.5]) (all inpatient) for PPS (from year 2009). There was a reduction to one-third (PDS) and one-fourth (PPS) of pre-pandemic numbers during the pandemic years (2020-2021). The peak months for scabies cases were the cooler months: January (median, 12.1% of annual cases; range [IQR], 2.6%-31.4% [3.6%]) to February (median, 10.0% of annual cases; range [IQR], 1.5%-27.8% [2.5%]) based on PDS data, and November (median, 10.0% of annual cases; range [IQR], 0.0%-24.3% [7.0%]) to January (median, 9.0% of annual cases; range [IQR], 0.0%-24.3% [6.6%]) for PPS data. Overall, for PDS, age 1-4 years is the most affected age group (median, PDS, 17.5% of annual cases; range [IQR], 11.9%-25.4% [8.1%]), while it was the less than 1-year-olds (median annual cases, 48.9%; range [IQR], 29.1%-67.3% [13.20%]) among PPS pediatric population aged 0 to 18 years. Males (median, 53.9% of annual cases; range [IQR], 45.0%-67.2% [8.8%]) were more affected than females in PPS. While for PDS during earlier years (prior to 2015), males (median, 51.6% of annual cases from 2010 to 2014; range [IQR], 47.4%-52.9% [0.2%]) were more affected than females. However, males became less affected than females with median, 44.7% of annual cases from 2015 onwards (range [IQR], 43.4%-46.5% [1.2%]). NCR was the region with the highest frequency of cases in PPS (median, 52.6% of annual cases; range [IQR], 22.7%-75.0% [20.4%]). The 2nd most affected regions were Central/Eastern Visayas (34.2%, 2009-2013; range [IQR], 17.9%-54.1% [5.3%]), Bicol region (12%; 2014 to 2018; range [IQR], 17.9%-54.1% [7.4%]), Central Luzon (18%; 2019), Central/Eastern Visayas (29%, 2020), and Northern/Central Mindanao (17%, 2021). Conclusion: Scabies was commonly seen in the younger age group, slightly more in females in the PDS, while slightly more among males in the PPS, in the cooler months of the year, and in the urbanized NCR.

Dataset on carbonation and chloride-induced steel corrosion in cementitious mortars.

Machine learning (ML) has seen success in civil and structural engineering, but its application to forecasting corrosion of steel reinforcement in concrete structures is limited due to small datasets from isolated studies. Moreover, the existing corrosion dataset of reinforced concrete typically lacks sufficient and comprehensive material and environmental information that enables reliable corrosion prediction of reinforced concrete under complex corrosion scenarios. This work aims to bridge the gap by compiling and building a comprehensive corrosion dataset focusing on carbon steel in cementitious mortars. This dataset involves 46 distinct mortar mixtures with embedded steel bars. The samples first underwent accelerated corrosion testing (either by carbonation or chloride contamination), followed by investigating their corrosion behaviours under varying relative humidity (RH) conditions. Corrosion data were obtained during this period, in which all corrosion measurements were conducted in laboratory settings and the results are tabulated in spreadsheet format (.xlsx). The dataset encompasses mixture parameters, material properties, environmental parameters, and electrochemical parameters. This extensive dataset provides valuable corrosion data for training ML models to predict steel corrosion across various corrosion-related variables.

Electronic Health Record-Oriented Knowledge Graph System for Collaborative Clinical Decision Support Using Multicenter Fragmented Medical Data: Design and Application Study.

BACKGROUND: The medical knowledge graph provides explainable decision support, helping clinicians with prompt diagnosis and treatment suggestions. However, in real-world clinical practice, patients visit different hospitals seeking various medical services, resulting in fragmented patient data across hospitals. With data security issues, data fragmentation limits the application of knowledge graphs because single-hospital data cannot provide complete evidence for generating precise decision support and comprehensive explanations. It is important to study new methods for knowledge graph systems to integrate into multicenter, information-sensitive medical environments, using fragmented patient records for decision support while maintaining data privacy and security. OBJECTIVE: This study aims to propose an electronic health record (EHR)-oriented knowledge graph system for collaborative reasoning with multicenter fragmented patient medical data, all the while preserving data privacy. METHODS: The study introduced an EHR knowledge graph framework and a novel collaborative reasoning process for utilizing multicenter fragmented information. The system was deployed in each hospital and used a unified semantic structure and Observational Medical Outcomes Partnership (OMOP) vocabulary to standardize the local EHR data set. The system transforms local EHR data into semantic formats and performs semantic reasoning to generate intermediate reasoning findings. The generated intermediate findings used hypernym concepts to isolate original medical data. The intermediate findings and hash-encrypted patient identities were synchronized through a blockchain network. The multicenter intermediate findings were collaborated for final reasoning and clinical decision support without gathering original EHR data. RESULTS: The system underwent evaluation through an application study involving the utilization of multicenter fragmented EHR data to alert non-nephrology clinicians about overlooked patients with chronic kidney disease (CKD). The study covered 1185 patients in nonnephrology departments from 3 hospitals. The patients visited at least two of the hospitals. Of these, 124 patients were identified as meeting CKD diagnosis criteria through collaborative reasoning using multicenter EHR data, whereas the data from individual hospitals alone could not facilitate the identification of CKD in these patients. The assessment by clinicians indicated that 78/91 (86%) patients were CKD positive. CONCLUSIONS: The proposed system was able to effectively utilize multicenter fragmented EHR data for clinical application. The application study showed the clinical benefits of the system with prompt and comprehensive decision support.

A deep learning approach for overall survival prediction in lung cancer with missing values.

BACKGROUND AND OBJECTIVE: In the field of lung cancer research, particularly in the analysis of overall survival (OS), artificial intelligence (AI) serves crucial roles with specific aims. Given the prevalent issue of missing data in the medical domain, our primary objective is to develop an AI model capable of dynamically handling this missing data. Additionally, we aim to leverage all accessible data, effectively analyzing both uncensored patients who have experienced the event of interest and censored patients who have not, by embedding a specialized technique within our AI model, not commonly utilized in other AI tasks. Through the realization of these objectives, our model aims to provide precise OS predictions for non-small cell lung cancer (NSCLC) patients, thus overcoming these significant challenges. METHODS: We present a novel approach to survival analysis with missing values in the context of NSCLC, which exploits the strengths of the transformer architecture to account only for available features without requiring any imputation strategy. More specifically, this model tailors the transformer architecture to tabular data by adapting its feature embedding and masked self-attention to mask missing data and fully exploit the available ones. By making use of ad-hoc designed losses for OS, it is able to account for both censored and uncensored patients, as well as changes in risks over time. RESULTS: We compared our method with state-of-the-art models for survival analysis coupled with different imputation strategies. We evaluated the results obtained over a period of 6 years using different time granularities obtaining a Ct-index, a time-dependent variant of the C-index, of 71.97, 77.58 and 80.72 for time units of 1 month, 1 year and 2 years, respectively, outperforming all state-of-the-art methods regardless of the imputation method used. CONCLUSIONS: The results show that our model not only outperforms the state-of-the-art's performance but also simplifies the analysis in the presence of missing data, by effectively eliminating the need to identify the most appropriate imputation strategy for predicting OS in NSCLC patients.

Multi-objective extensive hypothesis testing for the estimation of advanced crash frequency models.

Analyzing crash data is a complex and labor-intensive process that requires careful consideration of multiple interdependent modeling aspects, such as functional forms, transformations, likely contributing factors, correlations, and unobserved heterogeneity. Limited time, knowledge, and experience may lead to over-simplified, over-fitted, or misspecified models overlooking important insights. This paper proposes an extensive hypothesis testing framework including a multi-objective mathematical programming formulation and solution algorithms to estimate crash frequency models considering simultaneously likely contributing factors, transformations, non-linearities, and correlated random parameters. The mathematical programming formulation minimizes both in-sample fit and out-of-sample prediction. To address the complexity and non-convexity of the mathematical program, the proposed solution framework utilizes a variety of metaheuristic solution algorithms. Specifically, Harmony Search demonstrated minimal sensitivity to hyperparameters, enabling an efficient search for solutions without being influenced by the choice of hyperparameters. The effectiveness of the framework was evaluated using two real-world datasets and one synthetic dataset. Comparative analyses were performed using the two real-world datasets and the corresponding models published in literature by independent teams. The proposed framework showed its capability to pinpoint efficient model specifications, produce accurate estimates, and provide valuable insights for both researchers and practitioners. The proposed approach allows for the discovery of numerous insights while minimizing the time spent on model development. By considering a broader set of contributing factors, models with varied qualities can be generated. For instance, when applied to crash data from Queensland, the proposed approach revealed that the inclusion of medians on sharp curved roads can effectively reduce the occurrence of crashes, when applied to crash data from Washington, the simultaneous consideration of traffic volume and road curvature resulted in a notable reduction in crash variances but an increase in crash means.

Empirical analysis of trade costs and their impact on carbon emissions in RCEP countries.

Global economic integration and environmental issues have attracted widespread attention in recent years. As one of the world's most significant free trade agreements, the Regional Comprehensive Economic Partnership (RCEP) significantly impacts trade and the environment. However, research on the relationship between trade costs and carbon emissions still needs to be completed. This study explores the relationship between trade costs and carbon emissions within the framework of the Trade Benefit Theory, which posits that trade liberalization and openness generate economic benefits through increased efficiency, technological advancement, and economic growth. This study analyzes panel data from 12 RCEP countries from 2001 to 2014, employing static and dynamic panel models to examine the relationship between trade costs and carbon emissions. The analysis utilizes mixed regression, fixed (random) effects models, and the systematic GMM method. The results indicate that decreases in trade costs are associated with reduced environmental pollution, aligning with the Environmental Kuznets Curve (EKC) hypothesis, which posits an N-shaped relationship between trade costs and carbon emissions. Implementing RCEP facilitates a decrease in trade-related pollution, suggesting that reducing trade costs can help mitigate environmental pollution. Furthermore, the observed N-shaped EKC for trade costs and carbon emissions highlights the potential of RCEP to reduce the impact of trade-related pollution.

Leveraging VQ-VAE tokenization for autoregressive modeling of medical time series.

In this work, we present CodeAR, a medical time series generative model for electronic health record (EHR) synthesis. CodeAR employs autoregressive modeling on discrete tokens obtained using a vector quantized-variational autoencoder (VQ-VAE), which addresses key challenges of accurate distribution modeling and patient privacy preservation in the medical domain. The proposed model is trained with next-token prediction instead of a regression problem for more accurate distribution modeling, where the autoregressive property of CodeAR is useful to capture the inherent causality in time series data. In addition, the compressive property of the VQ-VAE prevents CodeAR from memorizing the original training data, which ensures patient privacy. Experimental results demonstrate that CodeAR outperforms the baseline autoregressive-based and GAN-based models in terms of maximum mean discrepancy (MMD) and Train on Synthetic, Test on Real tests. Our results highlight the effectiveness of autoregressive modeling on discrete tokens, the utility of CodeAR in causal modeling, and its robustness against data memorization.

Improved brain effective connectivity modeling by dynamic Bayesian networks.

BACKGROUND: If brain effective connectivity network modelling (ECN) could be accurately achieved, early diagnosis of neurodegenerative diseases would be possible. It has been observed in the literature that Dynamic Bayesian Network (DBN) based methods are more successful than others. However, DBNs have not been applied easily and tested much due to computational complexity problems in structure learning. NEW METHOD: This study introduces an advanced method for modelling brain ECNs using improved discrete DBN (Improved- dDBN) which addresses the computational challenges previously limiting DBN application, offering solutions that enable accurate and fast structure modeling. RESULTS: The practical data and prior sizes needed for the convergence to the globally correct network structure are proved to be much smaller than the theoretical ones using simulated dDBN data. Besides, Hill Climbing is shown to converge to the true structure at a reasonable iteration step size when the appropriate data and prior sizes are used. Finally, importance of data quantization methods are analysed. COMPARISON WITH EXISTING METHODS: The Improved-dDBN method performs better and robust, when compared to the existing methods for realistic scenarios such as varying graph complexity, various input conditions, noise cases and non-stationary connections. The data used in these tests is the simulated fMRI BOLD time series proposed in the literature. CONCLUSIONS: Improved-dDBN is a good candidate to be used on real datasets to accelerate developments in brain ECN modeling and neuroscience. Appropriate data and prior sizes can be identified based on the approach proposed in this study for global and fast convergence.

Multicenter privacy-preserving model training for deep learning brain metastases autosegmentation.

OBJECTIVES: This work aims to explore the impact of multicenter data heterogeneity on deep learning brain metastases (BM) autosegmentation performance, and assess the efficacy of an incremental transfer learning technique, namely learning without forgetting (LWF), to improve model generalizability without sharing raw data. MATERIALS AND METHODS: A total of six BM datasets from University Hospital Erlangen (UKER), University Hospital Zurich (USZ), Stanford, UCSF, New York University (NYU), and BraTS Challenge 2023 were used. First, the performance of the DeepMedic network for BM autosegmentation was established for exclusive single-center training and mixed multicenter training, respectively. Subsequently privacy-preserving bilateral collaboration was evaluated, where a pretrained model is shared to another center for further training using transfer learning (TL) either with or without LWF. RESULTS: For single-center training, average F1 scores of BM detection range from 0.625 (NYU) to 0.876 (UKER) on respective single-center test data. Mixed multicenter training notably improves F1 scores at Stanford and NYU, with negligible improvement at other centers. When the UKER pretrained model is applied to USZ, LWF achieves a higher average F1 score (0.839) than naive TL (0.570) and single-center training (0.688) on combined UKER and USZ test data. Naive TL improves sensitivity and contouring accuracy, but compromises precision. Conversely, LWF demonstrates commendable sensitivity, precision and contouring accuracy. When applied to Stanford, similar performance was observed. CONCLUSION: Data heterogeneity (e.g., variations in metastases density, spatial distribution, and image spatial resolution across centers) results in varying performance in BM autosegmentation, posing challenges to model generalizability. LWF is a promising approach to peer-to-peer privacy-preserving model training.

A Robust CETSA Data Analysis Automation Workflow for Routine Screening.

The Cellular Thermal Shift Assay (CETSA) enables the study of protein-ligand interactions in a cellular context. It provides valuable information on the binding affinity and specificity of both small and large molecule ligands in a relevant physiological context, hence forming a unique tool in drug discovery. Though high-throughput lab protocols exist for scaling up CETSA, subsequent data analysis and quality control remain laborious and limit experimental throughput. Here, we introduce a scalable and robust data analysis workflow which allows integration of CETSA into routine high throughput screening (HT-CETSA). This new workflow automates data analysis and incorporates quality control (QC), including outlier detection, sample and plate QC, and result triage. We describe the workflow and show its robustness against typical experimental artifacts, show scaling effects, and discuss the impact of data analysis automation by eliminating manual data processing steps.

Using the GAP score as a complement to the new injury severity score in identifying severely injured patients- A registry-based cohort study of adult trauma patients in Sweden.

BACKGROUND: New Injury Severity Score (NISS) and Glasgow Coma Scale, Age and Pressure (GAP) scoring systems have cutoffs to define severe injury and identify high-risk patients. This is important in trauma quality monitoring and improvement. The overall aim was to explore if GAP scoring system can be a complement or an alternative to the traditional NISS scoring system. METHODS: Adults exposed to trauma between 2017 and 2021 were included in the study, using data from The Swedish Trauma Registry. The performance of NISS and GAP scores in predicting mortality, and ICU admissions were assessed using the area under the receiver operator characteristics (AUROC) in all patients and in subgroups (blunt, penetrating trauma and older (≥65 years) trauma patients). Patients were classified as severely injured by NISS >15 as Severely Injured NISS (SIN) or with a high-risk for mortality, by GAP 3-18 as High Risk GAP (HRG). Undertriage was calculated based on the cutoffs HRG and SIN. RESULTS: Overall, 37,017 patients were included. The AUROC (95 % CI) for mortality using NISS was 0.84 (0.83-0.85) and for GAP 0.92 (0.91-0.93) (p-value <0.001), the AUROC (95 % CI) for ICU-admissions was 0.82 (0.82-0.83) using NISS and for GAP 0.70 (0.70-0.71) p-value <0.001, in the overall cohort. In older patients the AUROC (95 % CI) for mortality was 0.76 (0.75-0.78) using NISS and 0.79 (0.78-0.81) using GAP, p-value <0.001. Overall, 8,572 (23.2 %) and 2,908 (7.9 %) were classified as SIN and HRG, respectively, with mortality rates of 13.7 % and 34.3 %. In the HRG group low-energy falls dominated and in the SIN group most patients were exposed to MVCs. In the SIN and HRG groups the rate of Emergency Trauma Interventions according to Utstein guidelines (ETIU) and ICU admission was 14.0 vs 9.5 % and 47.0 vs 62.5 % respectively. CONCLUSION: Our findings suggest that the GAP score and its cutoff 3-18 can be used to define severe trauma as complement to NISS >15 and can be a valuable tool in trauma quality monitoring and improvement. However, both scoring systems were less accurate in predicting mortality for the older trauma patients and should be explored further.