Precise Prediction Research Articles

Development and validation of a deep learning-enhanced prediction model for the likelihood of pulmonary embolism

BackgroundPulmonary embolism (PE) is a common and potentially fatal condition. Timely and accurate risk assessment in patients with acute deep vein thrombosis (DVT) is crucial. This study aims to develop a deep learning-based, precise, and efficient PE risk prediction model (PE-Mind) to overcome the limitations of current clinical tools and provide a more targeted risk evaluation solution.MethodsWe analyzed clinical data from patients by first simplifying and organizing the collected features. From these, 37 key clinical features were selected based on their importance. These features were categorized and analyzed to identify potential relationships. Our prediction model uses a convolutional neural network (CNN), enhanced with three custom-designed modules for better performance. To validate its effectiveness, we compared this model with five commonly used prediction models.ResultsPE-Mind demonstrated the highest accuracy and reliability, achieving 0.7826 accuracy and an area under the receiver operating characteristic curve of 0.8641 on the prospective test set, surpassing other models. Based on this, we have also developed a Web server, PulmoRiskAI, for real-time clinician operation.ConclusionThe PE-Mind model improves prediction accuracy and reliability for assessing PE risk in acute DVT patients. Its convolutional architecture and residual modules substantially enhance predictive performance.

A deep ensemble learning framework for glioma segmentation and grading prediction

The segmentation and risk grade prediction of gliomas based on preoperative multimodal magnetic resonance imaging (MRI) are crucial tasks in computer-aided diagnosis. Due to the significant heterogeneity between and within tumors, existing methods mainly rely on single-task approaches, overlooking the inherent correlation between segmentation and grading tasks. Furthermore, the limited availability of glioma grading data presents further challenges. To address these issues, we propose a deep-ensemble learning framework based on multimodal MRI and the U-Net model, which simultaneously performs glioma segmentation and risk grade prediction. We introduce asymmetric convolution and dual-domain attention in the encoder, fully integrating effective information from different modalities, enhancing the extraction of features from critical regions, and constructing a dual-branch decoder that combines spatial features and global semantic information for both segmentation and grading. In addition, we propose a weighted composite adaptive loss function to balance the optimization objectives of the two tasks. Our experimental results on the BraTS dataset demonstrate that our method outperforms state-of-the-art methods, yielding superior segmentation accuracy and precise risk grade prediction.

An IoT Approach Enhanced with AI Capabilities for Predictive Neurology in Smart Homes

<p dir="ltr"><span>The integration of Artificial Intelligence (AI) and Internet of Things (IoT) offers remarkable chances for improving healthcare in smart home environments. In order to anticipate and observe neurological disorders, the study explores the use of IoT devices along with AI analytics. The goal is to identify early signs of any decline in cognitive functions or neurodegenerative conditions such as Alzheimer’s or Parkinson’s. The research emphasises on examining a framework for predictive neurology while it faces challenges associated with real-time data processing and prediction precision. The outcomes should result in a smart home ecosystem development that focuses on healthcare interventions and enhances life quality. This paper explores a multi-layered architecture, including an exploratory phase to analyse user needs, a developmental phase to design and integrate IoT devices, Machine Learning algorithms, and cloud architectures, as well as a validation phase to undergo rigorous testing in simulated environments to evaluate performance metrics.</span></p><div><span><br /></span></div>

Machine learning in shoulder arthroplasty : a systematic review of predictive analytics applications.

Machine learning (ML) holds significant promise in optimizing various aspects of total shoulder arthroplasty (TSA), potentially improving patient outcomes and enhancing surgical decision-making. The aim of this systematic review was to identify ML algorithms and evaluate their effectiveness, including those for predicting clinical outcomes and those used in image analysis. We searched the PubMed, EMBASE, and Cochrane Central Register of Controlled Trials databases for studies applying ML algorithms in TSA. The analysis focused on dataset characteristics, relevant subspecialties, specific ML algorithms used, and their performance outcomes. Following the final screening process, 25 articles satisfied the eligibility criteria for our review. Of these, 60% focused on tabular data while the remaining 40% analyzed image data. Among them, 16 studies were dedicated to developing new models and nine used transfer learning to leverage existing pretrained models. Additionally, three of these models underwent external validation to confirm their reliability and effectiveness. ML algorithms used in TSA demonstrated fair to good performance, as evidenced by the reported metrics. Integrating these models into daily clinical practice could revolutionize TSA, enhancing both surgical precision and patient outcome predictions. Despite their potential, the lack of transparency and generalizability in many current models poses a significant challenge, limiting their clinical utility. Future research should prioritize addressing these limitations to truly propel the field forward and maximize the benefits of ML in enhancing patient care.

Artificial intelligence-assisted precise preoperative prediction of lateral cervical lymph nodes metastasis in papillary thyroid carcinoma via a clinical-CT radiomic combined model.

This study aimed to develop an artificial intelligence-assisted model for the preoperative prediction of lateral cervical lymph node metastasis (LCLNM) in papillary thyroid carcinoma (PTC) using computed tomography (CT) radiomics, providing a new noninvasive and accurate diagnostic tool for PTC patients with LCLNM. This retrospective study included 389 confirmed PTC patients, randomly divided into a training set (n = 272) and an internal validation set (n = 117), with an additional 40 patients from another hospital as an external validation set. Patient demographics were evaluated to establish a clinical model. Radiomic features were extracted from preoperative contrast-enhanced CT images (venous phase) for each patient. Feature selection was performed using analysis of variance and the least absolute shrinkage and selection operator algorithm. We employed support vector machine, random forest (RF), logistic regression, and XGBoost algorithms to build CT radiomic models for predicting LCLNM. A radiomics score (Rad-score) was calculated using a radiomic signature-based formula. A combined clinical-radiomic model was then developed. The performance of the combined model was evaluated using the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA). A total of 1724 radiomic features were extracted from each patient's CT images, with 13 features selected based on nonzero coefficients related to LCLNM. Four clinically relevant factors (age, tumor location, thyroid capsule invasion, and central cervical lymph node metastasis) were significantly associated with LCLNM. Among the algorithms tested, the RF algorithm outperformed the others with five-fold cross-validation on the training set. After integrating the best algorithm with clinical factors, the areas under the ROC curves for the training, internal validation, and external validation sets were 0.910 (95% confidence interval [CI]: 0.729-0.851), 0.876 (95% CI: 0.747-0.911), and 0.821 (95% CI: 0.555-0.802), respectively, with DCA demonstrating the clinical utility of the combined radiomic model. This study successfully established a clinical-CT radiomic combined model for predicting LCLNM, which may significantly enhance surgical decision-making for lateral cervical lymph node dissection in patients with PTC.

Highly Precise Prediction of Micro- and Supra-pKa Based on 3D Descriptors Integrating Non-Covalent Interactions.

Accurate pKa prediction is crucial for understanding proton dissociation in complex molecular systems. However, existing models often face challenges in addressing subtle stereoelectronic effects and conformational flexibility. This study presents H-SPOC, a localized 3D descriptor that captures covalent and non-covalent interactions and incorporates solvent effects to predict site-specific pKavalues accurately. H-SPOC was validated on multiple benchmark datasets, including SAMPL6, SAMPL7, and SAMPL8, where it outperformed state-of-the-art methods. H-SPOC also proved versatile across various applications, including aspirin's non-equilibrium conformations, glycine's microstate distributions, and the stereoelectronic anomalies of Janus Sponge and Meldrum's Acid. It addressed challenging supra-pKapredictions in crystalline environments and accurately correlated pKa with reaction rates, selectivity, tautomerism, and pharmacokinetic properties. With its chemically intuitive design and computational efficiency, H-SPOC provides an efficient framework for rapid and precise micro- and supra-pKapredictions, offering significant potential in drug discovery, catalysis, and materials science.

Highly Precise Prediction of Micro‐ and Supra‐pKa Based on 3D Descriptors Integrating Non‐Covalent Interactions

Accurate pKa prediction is crucial for understanding proton dissociation in complex molecular systems. However, existing models often face challenges in addressing subtle stereoelectronic effects and conformational flexibility. This study presents H‐SPOC, a localized 3D descriptor that captures covalent and non‐covalent interactions and incorporates solvent effects to predict site‐specific pKa values accurately. H‐SPOC was validated on multiple benchmark datasets, including SAMPL6, SAMPL7, and SAMPL8, where it outperformed state‐of‐the‐art methods. H‐SPOC also proved versatile across various applications, including aspirin's non‐equilibrium conformations, glycine's microstate distributions, and the stereoelectronic anomalies of Janus Sponge and Meldrum's Acid. It addressed challenging supra‐pKa predictions in crystalline environments and accurately correlated pKa with reaction rates, selectivity, tautomerism, and pharmacokinetic properties. With its chemically intuitive design and computational efficiency, H‐SPOC provides an efficient framework for rapid and precise micro‐ and supra‐pKa predictions, offering significant potential in drug discovery, catalysis, and materials science.

Lithium-Ion Battery State of Health Estimation Based on Model Interpretable Feature Extraction

Abstract This study proposes a lithium battery State of Health (SOH) estimation method that utilizes model interpretability feature extraction and the Equilibrium Optimizer (EO) algorithm to optimize Temporal convolutional neural networks (TCN), addressing issues of feature collinearity, noise interference, and the challenges of manual model parameter tuning. Initially, the battery's incremental capacity (IC) curve is smoothed using Gaussian filtering, and the health features are extracted from the charging, discharging and IC curve to establish a TCN-based SOH estimation model. Subsequently, the SHAP interpretability method is employed to analyze the contribution of various features to the TCN model's predictions, and the features were further screened based on these contributions; the EO algorithm is used to optimize the TCN model hyperparameters, enhancing the model's prediction performance. Finally, this study builds an experimental platform for ageing tests to validate this method with experimental data and public datasets. The results show that SHAP analysis and the EO algorithm, based on the model's real-time feedback mechanism, significantly improved the accuracy of feature selection and model prediction precision. The method proposed in this study achieves an RMSE of below 1.40% for the SOH prediction of six batteries, reducing it by 66.7% compared to the baseline model.

Prediction of Pellet Durability Index (PDI) in a Commercial Feed Mill Using Multiple Linear Regression with Variable Selection and Dimensionality Reduction.

Pellet quality, measured as Pellet Durability Index (PDI), is an important key performance indicator (KPI) for commercial feed manufacturing, as it can impact both mill efficiency and downstream performance of animals fed the manufactured diets. However, it is an ongoing challenge for the feed industry to control pellet quality, due to the complexity of feed manufacturing and the large number of variables influencing the process. Previous studies have explored prediction of pellet quality using either simple empirical models with a few variables or machine learning models with many variables. The objective of the current study was to develop statistical regression models to predict PDI, and to describe the relationship between pellet quality and 55 available variables based on a dataset with 2691 observations collected from a commercial feed mill. In the current study, the response variable (PDI) was transformed using the Box-Cox approach into the transformed response variable (tPDI), that was more normally distributed. Three multiple regression models were developed based on subsets of variables processed by variable selection and dimensionality reduction methods: Forward Selection, Principal Component Analysis, and Partial Least Squares. The results indicated that Model 1 (Forward Selection with manual removal of sparse variables), built on 9 variables, performed better than the other two models. It exhibited consistent model prediction performance on the training data and testing data, in terms of MAE (1.93 ± 0.063 versus 1.96), RMSPE (2.45 ± 0.079 versus 2.45), and CCC (0.549 ± 0.0273 versus 0.550), with a better prediction precision based on the fit plot. Expanding Temperature (℃), Fat Content (%), and ADF Content (%), and Indoor Humidity (Pelletizer) (%) were identified as more influential than other variables on the transformed response variable (tPDI) in Model 1, based on a behavior analysis. The models developed in the current study can be helpful to feed mills for predicting and comprehending the effect of a number of commonly measured variables on pellet quality in the commercial setting.

A Multi-View Feature-Based Interpretable Deep Learning Framework for Drug-Drug Interaction Prediction.

Drug-drug interactions (DDIs) can result in deleterious consequences when patients take multiple medications simultaneously, emphasizing the critical need for accurate DDI prediction. Computational methods for DDI prediction have garnered recent attention. However, current approaches concentrate solely on single-view features, such as atomic-view or substructure-view features, limiting predictive capacity. The scarcity of research on interpretability studies based on multi-view features is crucial for tracing interactions. Addressing this gap, we present MI-DDI, a multi-view feature-based interpretable deep learning framework for DDI. To fully extract multi-view features, we employ a Message Passing Neural Network (MPNN) to learn atomic features from molecular graphs generated by RDkit, and transformer encoders are used to learn substructure-view embeddings from drug SMILES simultaneously. These atomic-view and substructure-view features are then amalgamated into a holistic drug embedding matrix. Subsequently, an intricately designed interaction module not only establishes a tractable path for understanding interactions but also directly informs the construction of weight matrices, enabling precise and interpretable interaction predictions. Validation on the BIOSNAP dataset and DrugBank dataset demonstrates MI-DDI's superiority. It surpasses the current benchmarks by a substantial average of 3% on BIOSNAP and 1% on DrugBank. Additional experiments underscore the significance of atomic-view information for DDI prediction and confirm that our interaction module indeed learns more effective information for DDI prediction. The source codes are available at https://github.com/ZihuiCheng/MI-DDI .

Dynamic behavior of solitons in nonlinear Schrödinger equations

The primary objective of this study is to derive analytical solutions for a significant system that models the evolution of complex wave fields in nonlinear media. This system extends the framework of nonlinear Schrödinger equations and is pivotal in various physical applications, including optical fibers and Bose-Einstein condensates. By employing advanced analytical techniques such as the Khater II, III (Khat II, Khat III) and Unified (UF) methods, we successfully obtain exact analytical solutions that enhance our understanding of the system’s dynamic behavior. The findings reveal a variety of soliton-like solutions, demonstrating the robustness and effectiveness of the methodologies employed. This research underscores the importance of the system in modeling intricate physical phenomena and offers a novel perspective on its solution space. The originality of this study lies in the innovative application of Khat II, Khat III, and UF methods to this system, providing valuable insights and methodologies for future research endeavors. This work represents a significant contribution to applied mathematics and nonlinear dynamics, emphasizing the physical relevance of the system in representing nonlinear wave interactions. The analytical solutions obtained can facilitate precise control and prediction of wave behaviors in practical applications, thereby advancing our capability to manipulate and harness nonlinear wave phenomena in diverse scientific and engineering contexts.

Localizing hierarchical prediction errors and precisions during an oddball task with volatility: Computational insights and relationship with psychosocial functioning in healthy individuals

Abstract The auditory mismatch negativity (MMN) has been widely used to investigate deficits in early auditory information processing, particularly in psychosis. Predictive coding theories suggest that impairments in sensory learning may arise from disturbances in hierarchical message passing, likely due to aberrant precision-weighting of prediction errors (PEs). This study employed a modified auditory oddball paradigm with varying phases of stability and volatility to disentangle the impact of hierarchical PEs on auditory MMN generation in 43 healthy controls (HCs). Single-trial EEG data were modeled with a hierarchical Bayesian model of learning to identify neural correlates of low-level PEs about tones and high-level PEs about environmental volatility. Our analysis revealed a reduced expression of the auditory MMN in volatile compared to stable phases of the paradigm. Additionally, lower Global Functioning (GF): Social scores were associated with a reduced difference waveform at 332 ms after stimulus presentation across the entire MMN paradigm. Further analysis revealed that this association was present during the volatile phase but not the stable phase of the paradigm. Source reconstruction suggested that the association between the stable difference waveform and psychosocial functioning originated in the left superior temporal gyrus. Finally, we found significant EEG signatures of both low- and high-level PEs and precision ratios. Our findings highlight the value of computational models in understanding the neural mechanisms involved in early auditory information processing and their connection to psychosocial functioning.

Geographical Origin Traceability of Navel Oranges Based on Near-Infrared Spectroscopy Combined with Deep Learning

The quality and price of navel oranges vary depending on their geographical origin, thus providing a financial incentive for origin fraud. To prevent this phenomenon, it is necessary to explore a fast, non-destructive, and precise method for tracing the origin of navel oranges. In this study, a total of 490 Newhall navel oranges were selected from five major production regions in China, and the diffuse reflectance near-infrared spectrum in 4000–10,000 cm−1 were non-invasively collected. We examined seven preprocessing techniques for the spectra, including Savitzky–Golay (SG) smoothing, first derivative (FD), multiplicative scattering correction (MSC), combinations of SG with MSC (SG+MSC), SG with FD (SG+FD), MSC with FD (MSC+FD), and three combined (SG+MSC+FD). A one-dimensional convolutional neural network (1DCNN) deep learning model for geographical origin tracing of navel orange was established, and five machine learning algorithms, i.e., partial least squares discriminant analysis (PLS-DA), linear discriminant analysis (LDA), support vector machine (SVM), random forest (RF), and back-propagation neural network (BPNN), were compared with 1DCNN. The results show that the 1DCNN model based on the SG+FD preprocessing method achieved the optimal performance for the testing set, with prediction accuracy, precision, recall, and F1-score of 97.92%, 98%, 97.95%, and 97.90%, respectively. Therefore, NIRS combined with deep learning has a significant research and application value in the rapid, nondestructive, and accurate geographical origin traceability of agricultural products.

Assessment of prognosis and responsiveness to immunotherapy in colorectal cancer patients based on the level of immune cell infiltration

ObjectiveTo build a new prognostic risk assessment model based on immune cell co-expression networks for predicting overall survival and evaluating the efficacy of immunotherapy for colon cancer patients.MethodsThe Cancer Genome Atlas (TCGA) database was used to obtain mRNA expression profiling data, clinical information, and somatic mutation data from colorectal cancer patients. The degree of tumor immune cell infiltration of the samples was analyzed using the CIBERSORT algorithm. Co-expression of immune-related genes was analyzed using weighted correlation network analysis (WGCNA) and gene modules were identified. Prognosis-related genes were screened and models were constructed using LASSO-Cox analysis. The models were validated by survival analysis. The prognostic potential of the models was quantitatively assessed using Cox regression analysis and the development of column line plots. Immunotherapy sensitivity analysis was performed using CIBERSORT and TIMER algorithms. Gene biofunction analysis was performed using Gene set enrichment analysis (GSEA) and Gene set variation analysis (GSVA). And the chemotherapeutic response to different drugs was assessed.ResultsWe established a novel prognostic model utilizing the WGCNA method, which demonstrated robust predictive accuracy for patient survival. The high-risk subgroup in our model exhibited elevated immune cell infiltration coupled with a higher tumor mutation burden, but the difference in response to immunotherapy was not significant compared to the low-risk group. Furthermore, we identified distinct chemotherapy responses to 39 drugs between these risk subgroups.ConclusionThis study revealed a significant correlation between high levels of immune infiltration and unfavorable prognosis in patients with colon cancer. Furthermore, an accurate prognostic risk prediction model based on the co-expression of relevant genes by immune cells was developed, enabling precise prediction of survival of colon cancer patients. These findings offer valuable insights for accurate prognostication and comprehensive management of individuals diagnosed with colon cancer.

Artificial Intelligence in the Heart of Medicine: Transforming Arrhythmia Care with Intelligent Systems.

At a critical juncture in the ongoing fight against cardiovascular disease (CVD), healthcare professionals are striving for more informed and expedited decisionmaking. Artificial Intelligence (AI) promises to be a guiding light in this endeavor. The diagnosis of coronary artery disease has now become non-invasive and convenient, while wearable devices excel at promptly detecting life-threatening arrhythmias and treatments for heart failure. This study aimed to highlight the applications of AI in cardiology with a particular focus on arrhythmias and its potential impact on healthcare for all through careful implementation and constant research efforts. An extensive search strategy was implemented. The search was conducted in renowned electronic medical databases, including Medline, PubMed, Cochrane Library, and Google Scholar. Artificial Intelligence, cardiovascular diseases, arrhythmias, machine learning, and convolutional neural networks in cardiology were used as keywords for the search strategy. A total of 6876 records were retrieved from different electronic databases. Duplicates (N = 1356) were removed, resulting in 5520 records for screening. Based on predefined inclusion and exclusion criteria, 4683 articles were excluded. Following the full-text screening of the remaining 837 articles, a further 637 were excluded. Ultimately, 200 studies were included in this review. AI represents not just a development but a cutting-edge force propelling the next evolution of cardiology. With its capacity to make precise predictions, facilitate non-invasive diagnosis, and personalize therapies, AI holds the potential to save lives and enhance healthcare quality on a global scale.

A novel RNN architecture to improve the precision of ship trajectory predictions

ABSTRACT Monitoring maritime transport activities is crucial for ensuring the security and safety of people and goods. This type of monitoring often relies on the use of navigation systems such as the Automatic Identification System (AIS). AIS data has been used to support the defense teams when identifying equipment defects, locating suspicious activity, ensuring ship collision avoidance, and detecting hazardous events. In this context, Ship Trajectory Prediction (STP) has been conducted using AIS data to support the estimation of vessel routes and locations, contributing to maritime safety and situational awareness. Currently, the Ornstein-Uhlenbeck (OU) model is considered the state-of-the-art for STP. However, this model can be time-consuming and can only represent a single vessel track. To solve these challenges, Recurrent Neural Network (RNN) models have been applied to STP to allow scalability for large data sets and to capture larger regions or anomalous vessels behavior. This research proposes a new RNN architecture that decreases the prediction error up to 50% for cargo vessels when compared to the OU model. Results also confirm that the proposed Decimal Preservation layer can benefit other RNN architectures developed in the literature by reducing their prediction errors for complex data sets.

Disulfidptosis-related immune patterns predict prognosis and characterize the tumor microenvironment in oral squamous cell carcinoma

BackgroundEstablishing a prognostic risk model based on immunological and disulfidptosis signatures enables precise prognosis prediction of oral squamous cell carcinoma (OSCC).MethodsDifferentially expressed immune and disulfidptosis genes were identified in OSCC and normal tissues. We examined the model’s clinical applicability and its relationship to immune cell infiltration. Additionally, the risk score, ssGSEA, ESTIMATE, and CIBERSORT were used to evaluate the intrinsic molecular subtypes, immunological checkpoints, abundances of tumor-infiltrating immune cell types and proportions between the two risk groups. GO-KEGG and GSVA analyses were performed to identify enriched pathways.ResultsWe analyzed the correlation immune genes based on the 14 disulfidptosis-related genes, and found 379 disulfidptosis-related immune genes (DRIGs). After univariate Cox regression we obtained 30 DRIGs and least absolute shrinkage and selection operator (LASSO) regression to reduce the number of genes to 16. Finally we created a nine-DRIGs risk model, of which four were upregulated and five were downregulated. The analysis results showed that disulfidptosis was tightly related to immune cells, immunological-related pathways, the tumor microenvironment (TME), immune checkpoints, human leukocyte antigen (HLA), and tumor mutational burden (TMB). The nomogram, integrating the risk score and clinical factors, accurately predicted overall survival.ConclusionsThis novel risk model highlights the role of disulfidptosis-related immune genes in OSCC prognosis. With this model, we can more accurately predict the prognosis of patients with OSCC, as well as assess the potential effects of their TME and immunotherapy.

Prediction model establishment of prognosis factors for distant metastasis of hepatocellular carcinoma based on the SEER database.

Distant metastasis in hepatocellular carcinoma (HCC) is an important indicator of poor patient prognosis. Identifying patients who are at high risk of metastasis early on is essential for creating personalized treatment plans, yet currently, there is a scarcity of effective predictive tools. To investigate the effects of different factors on distant metastasis in HCC patients and to establish a clinical prediction model for predicting distant metastasis in HCC patients. Our study retrospectively examined 22,318 patients diagnosed with confirmed HCC from the SEER database. Prognostic factors for developing distant metastases in HCC patients were identified by univariate and multivariate logistic regression analyses. Utilizing data from a multivariate logistic regression analysis, we created a nomogram. Its predictive precision was evaluated by analyzing the calibration curve, the area under the curve (AUC) of the receiver operating characteristic curve, decision curve assessment (DCA), and Kaplan-Meier (KM) curve analysis of overall survival. Finally,the nomogram was visualized with an online calculator. We identified six independent prognostic factors: ethnicity, marital status, tumor size, survival time, surgery, and radiotherapy. The nomogram constructed from these six factors showed good calibration, discrimination, and clinical application value after calibration curve analysis, receiver operating characteristic curve analysis and DCA curve analysis. Besides, KaplanMeier survival curves also demonstrated that this nomogram had predictive accuracy. In this research, a nomogram model was created to accurately predict distant metastasis risk in patients with HCC. This study provides guidance for optimizing individual therapies and making better clinical decisions.

Confinement and wettability-driven dispersed phase hydrodynamics in cross-flow jets at low velocity and density ratios

The evolution characteristics of a low-velocity dispersed phase into continuous shear flow have numerous applications across biomedical devices, chemical processes, water management in fuel cells, spray systems, film deposition, and atomizing devices. The flow characteristics arise from a complex interplay of wettability, hydrodynamics, and interfacial properties, which, when constrained by confined geometries such as those in fuel cells, present a fascinating multiphase-multiphysics problem. This study investigates the impact of the chemical signature of a confined geometry and the velocity ratio between the dispersed and continuous phases on the evolution of the dispersed phase. The footprint and shape of the generated droplet guide the pressure distribution, deformation, and subsequent cross-flow-induced stretching. By systematically analyzing the dynamic effects of capillarity, inertia, air-shear, gravity, viscosity, wettability, and confinement, we classify the fate of a liquid droplet within classical flow regimes: jetting, threading, and dripping. These distinct flow regimes are mapped using classical non-dimensional numbers, and a quasi-universal characteristic is obtained relative to velocity ratios. The findings of this research contribute to precise control and prediction of dispersed-phase hydrodynamics, which play a pivotal role in enhancing the efficiency of fuel cells, droplet generation devices, water harvesting systems, film deposition techniques, coatings, and point-of-care diagnostic devices. The work underscores the relevance of integrating experimental and computational insights for optimizing interface-driven processes in interdisciplinary applications.

Abstract 33: A Novel Imaging Biomarker to Make Precise Outcome Predictions for Patients with Acute Ischemic Stroke

Introduction: Net water uptake (NWU) is a novel biomarker which measures edema and tissue injury from the degree of hypoattenuation on non-contrast CT and may serve as a precision tool for predicting outcomes after acute ischemic stroke (AIS). Using our recently developed algorithm, this study aimed to evaluate the relationship between NWU and post-stroke neurologic outcomes, including language impairment and motor weakness. Methods: Consecutive patients treated for AIS at certified stroke centers in Houston, TX were included. Patients’ precise functional outcomes at hospital discharge were recorded including decreased level of consciousness, presence of language impairment, visual deficit, arm and leg weakness, need for walking assistance, and gastrostomy placement. The primary outcome for this study was the performance of calculated NWU and clinical variables to predict language impairment at discharge. Baseline characteristics were compared, and then univariate and multivariate logistic regression were used to evaluate the association between clinical variables, imaging data, and the precise neurological outcomes. Results: Among 776 patients with AIS, average age was 67.0 +/- 14.8, 47.8% were female, median NIHSS was 10 [5,18], median ASPECTS was 9 [7,10], 42.6% received tPA, and 67.1% had a large vessel occlusion (see Table 1). In univariate logistic regression, higher NWU (OR 1.45, CI 1.30-1.63) and lower ASPECTS (OR 0.68, CI 0.63-0.74) were both significantly associated with higher likelihood of language impairment and other deficits at discharge (see Table 2). Additionally, higher NWU in all ten regions was significantly associated with deficit at discharge. In multivariate logistic regression, certain clinical and imaging variables remained significantly associated as described in Table 3. The ASPECTS and NWU-based regression models were directly compared when predicting language impairment using ROC curve analysis, and areas under the curve were 0.838 vs. 0.851 respectively (p = 0.152 with Delong test, see Figure 1). Conclusion: The novel NWU biomarker was significantly associated with precise post-AIS outcomes at discharge. When controlling for confounders, NWU was non-inferior to ASPECTS. Moving forward, region-based and overall NWU will need to be studied with long-term patient outcomes. Ultimately, this novel and open-access imaging biomarker could be used in the emergency setting to guide treatment decision-making and patient counseling.