Test Dataset Research Articles

Deep learning system for malignancy risk prediction in cystic renal lesions: a multicenter study.

To develop an interactive, non-invasive artificial intelligence (AI) system for malignancy risk prediction in cystic renal lesions (CRLs). In this retrospective, multicenter diagnostic study, we evaluated 715 patients. An interactive geodesic-based 3D segmentation model was created for CRLs segmentation. A CRLs classification model was developed using spatial encoder temporal decoder (SETD) architecture. The classification model combines a 3D-ResNet50 network for extracting spatial features and a gated recurrent unit (GRU) network for decoding temporal features from multi-phase CT images. We assessed the segmentation model using sensitivity (SEN), specificity (SPE), intersection over union (IOU), and dice similarity (Dice) metrics. The classification model's performance was evaluated using the area under the receiver operator characteristic curve (AUC), accuracy score (ACC), and decision curve analysis (DCA). From 2012 to 2023, we included 477 CRLs (median age, 57 [IQR: 48-65]; 173 men) in the training cohort, 226 CRLs (median age, 60 [IQR: 52-69]; 77 men) in the validation cohort, and 239 CRLs (median age, 59 [IQR: 53-69]; 95 men) in the testing cohort (external validation cohort 1, cohort 2, and cohort 3). The segmentation model and SETD classifier exhibited excellent performance in both validation (AUC = 0.973, ACC = 0.916, Dice = 0.847, IOU = 0.743, SEN = 0.840, SPE = 1.000) and testing datasets (AUC = 0.998, ACC = 0.988, Dice = 0.861, IOU = 0.762, SEN = 0.876, SPE = 1.000). The AI system demonstrated excellent benign-malignant discriminatory ability across both validation and testing datasets and illustrated improved clinical decision-making utility. In this era when incidental CRLs are prevalent, this interactive, non-invasive AI system will facilitate accurate diagnosis of CRLs, reducing excessive follow-up and overtreatment. The rising prevalence of CRLs necessitates better malignancy prediction strategies. The AI system demonstrated excellent diagnostic performance in identifying malignant CRL. The AI system illustrated improved clinical decision-making utility.

A novel radiomics-based technique for identifying vulnerable coronary plaques: a follow-up study.

Previous reports have suggested that coronary computed tomography angiography (CCTA)-based radiomics analysis is a potentially helpful tool for assessing vulnerable plaques. We aimed to investigate whether coronary radiomic analysis of CCTA images could identify vulnerable plaques in patients with stable angina pectoris. This retrospective study included patients initially diagnosed with stable angina pectoris. Patients were randomly divided into either the training or test dataset at an 8 : 2 ratio. Radiomics features were extracted from CCTA images. Radiomics models for predicting vulnerable plaques were developed using the support vector machine (SVM) algorithm. The model performance was assessed using the area under the curve (AUC); the accuracy, sensitivity, and specificity were calculated to compare the diagnostic performance using the two cohorts. A total of 158 patients were included in the analysis. The SVM radiomics model performed well in predicting vulnerable plaques, with AUC values of 0.977 and 0.875 for the training and test cohorts, respectively. With optimal cutoff values, the radiomics model showed accuracies of 0.91 and 0.882 in the training and test cohorts, respectively. Although further larger population studies are necessary, this novel CCTA radiomics model may identify vulnerable plaques in patients with stable angina pectoris.

Machine Learning-Enabled Image Classification for Automated Electron Microscopy.

Traditionally, materials discovery has been driven more by evidence and intuition than by systematic design. However, the advent of "big data" and an exponential increase in computational power have reshaped the landscape. Today, we use simulations, artificial intelligence (AI), and machine learning (ML) to predict materials characteristics, which dramatically accelerates the discovery of novel materials. For instance, combinatorial megalibraries, where millions of distinct nanoparticles are created on a single chip, have spurred the need for automated characterization tools. This paper presents an ML model specifically developed to perform real-time binary classification of grayscale high-angle annular dark-field images of nanoparticles sourced from these megalibraries. Given the high costs associated with downstream processing errors, a primary requirement for our model was to minimize false positives while maintaining efficacy on unseen images. We elaborate on the computational challenges and our solutions, including managing memory constraints, optimizing training time, and utilizing Neural Architecture Search tools. The final model outperformed our expectations, achieving over 95% precision and a weighted F-score of more than 90% on our test data set. This paper discusses the development, challenges, and successful outcomes of this significant advancement in the application of AI and ML to materials discovery.

Prediction of DNA methylation-based tumor types from histopathology in central nervous system tumors with deep learning.

Precision in the diagnosis of diverse central nervous system (CNS) tumor types is crucial for optimal treatment. DNA methylation profiles, which capture the methylation status of thousands of individual CpG sites, are state-of-the-art data-driven means to enhance diagnostic accuracy but are also time consuming and not widely available. Here, to address these limitations, we developed Deep lEarning from histoPathoLOgy and methYlation (DEPLOY), a deep learning model that classifies CNS tumors to ten major categories from histopathology. DEPLOY integrates three distinct components: the first classifies CNS tumors directly from slide images ('direct model'), the second initially generates predictions for DNA methylation beta values, which are subsequently used for tumor classification ('indirect model'), and the third classifies tumor types directly from routinely available patient demographics. First, we find that DEPLOY accurately predicts beta values from histopathology images. Second, using a ten-class model trained on an internal dataset of 1,796 patients, we predict the tumor categories in three independent external test datasets including 2,156 patients, achieving an overall accuracy of 95% and balanced accuracy of 91% on samples that are predicted with high confidence. These results showcase the potential future use of DEPLOY to assist pathologists in diagnosing CNS tumors within a clinically relevant short time frame.

Semi-quantitative group testing for efficient and accurate qPCR screening of pathogens with a wide range of loads.

Pathogenic infections pose a significant threat to global health, affecting millions of people every year and presenting substantial challenges to healthcare systems worldwide. Efficient and timely testing plays a critical role in disease control and transmission prevention. Group testing is a well-established method for reducing the number of tests needed to screen large populations when the disease prevalence is low. However, it does not fully utilize the quantitative information provided by qPCR methods, nor is it able to accommodate a wide range of pathogen loads. To address these issues, we introduce a novel adaptive semi-quantitative group testing (SQGT) scheme to efficiently screen populations via two-stage qPCR testing. The SQGT method quantizes cycle threshold (Ct) values into multiple bins, leveraging the information from the first stage of screening to improve the detection sensitivity. Dynamic Ct threshold adjustments mitigate dilution effects and enhance test accuracy. Comparisons with traditional binary outcome GT methods show that SQGT reduces the number of tests by 24% on the only complete real-world qPCR group testing dataset from Israel, while maintaining a negligible false negative rate. In conclusion, our adaptive SQGT approach, utilizing qPCR data and dynamic threshold adjustments, offers a promising solution for efficient population screening. With a reduction in the number of tests and minimal false negatives, SQGT holds potential to enhance disease control and testing strategies on a global scale.

Tensile testing data of additive manufactured ASTM D638 standard specimens with embedded internal geometrical features.

Additive manufacturing (AM) is now widely used for research and industrial production. The benchmark data for mechanical properties of additively manufactured specimens is very useful for many communities. This data article presents a tensile testing dataset of ASTM D638 size specimens without and with embedded internal geometrical features printed using polylactic acid (PLA) in a Fused Filament Fabrication (FFF) additive manufacturing process. The added features can mimic defects of various shapes and sizes. This work is a supplement to the published research article Assisted defect detection by in-process monitoring of additive manufacturing using optical imaging and infrared thermography (Additive Manufacturing, 2023, 103483). The printed specimens were tensile tested. Stress-strain graphs were developed and used to calculate the mechanical properties such as ultimate tensile strength (UTS) and strain at UTS. The mechanical properties, the correlations between mechanical properties and size, shape and location of geometrical features (defects), and the trends in mechanical properties can be useful in benchmarking the results of other researchers.

An end-to-end deep convolutional neural network-based dose engine for parotid gland cancer seed implant brachytherapy.

Seed implant brachytherapy (SIBT) is a promising treatment modality for parotid gland cancers (PGCs). However, the current clinical standard dose calculation method based on the American Association of Physicists in Medicine (AAPM) Task Group 43 (TG-43) Report oversimplifies patient anatomy as a homogeneous water phantom medium, leading to significant dose calculation errors due to heterogeneity surrounding the parotid gland. Monte Carlo Simulation (MCS) can yield accurate dose distributions but the long computation time hinders its wide application in clinical practice. This paper aims to develop an end-to-end deep convolutional neural network-based dose engine (DCNN-DE) to achieve fast and accurate dose calculation for PGC SIBT. A DCNN model was trained using the patient's CT images and TG-43-based dose maps as inputs, with the corresponding MCS-based dose maps as the ground truth. The DCNN model was enhanced based on our previously proposed model by incorporating attention gates (AGs) and large kernel convolutions. Training and evaluation of the model were performed using a dataset comprising 188 PGC I-125 SIBT patient cases, and its transferability was tested on an additional 16 non-PGC head and neck cancers (HNCs) I-125 SIBT patient cases. Comparison studies were conducted to validate the superiority of the enhanced model over the original one and compare their overall performance. On the PGC testing dataset, the DCNN-DE demonstrated the ability to generate accurate dose maps, with percentage absolute errors (PAEs) of 0.67%±0.47% for clinical target volume (CTV) D90 and 1.04%±1.33% for skin D0.1cc. The comparison studies revealed that incorporating AGs and large kernel convolutions resulted in 8.2% (p<0.001) and 3.1% (p<0.001) accuracy improvement, respectively, as measured by dose mean absolute error. On the non-PGC HNC dataset, the DCNN-DE exhibited good transferability, achieving a CTV D90 PAE of 1.88%±1.73%. The DCNN-DE can generate a dose map in less than 10 ms. We have developed and validated an end-to-end DCNN-DE for PGC SIBT. The proposed DCNN-DE enables fast and accurate dose calculation, making it suitable for application in the plan optimization and evaluation process of PGC SIBT.

Development of interpretable machine learning models to predict in-hospital prognosis of acute heart failure patients.

In recent years, there has been remarkable development in machine learning (ML) models, showing a trend towards high prediction performance. ML models with high prediction performance often become structurally complex and are frequently perceived as black boxes, hindering intuitive interpretation of the prediction results. We aimed to develop ML models with high prediction performance, interpretability, and superior risk stratification to predict in-hospital mortality and worsening heart failure (WHF) in patients with acute heart failure (AHF). Based on the Kyoto Congestive Heart Failure registry, which enrolled 4056 patients with AHF, we developed prediction models for in-hospital mortality and WHF using information obtained on the first day of admission (demographics, physical examination, blood test results, etc.). After excluding 16 patients who died on the first or second day of admission, the original dataset (n=4040) was split 4:1 into training (n=3232) and test datasets (n=808). Based on the training dataset, we developed three types of prediction models: (i) the classification and regression trees (CART) model; (ii) the random forest (RF) model; and (iii) the extreme gradient boosting (XGBoost) model. The performance of each model was evaluated using the test dataset, based on metrics including sensitivity, specificity, area under the receiver operating characteristic curve (AUC), Brier score, and calibration slope. For the complex structure of the XGBoost model, we performed SHapley Additive exPlanations (SHAP) analysis, classifying patients into interpretable clusters. In the original dataset, the proportion of females was 44.8% (1809/4040), and the average age was 77.9±12.0. The in-hospital mortality rate was 6.3% (255/4040) and the WHF rate was 22.3% (900/4040) in the total study population. In the in-hospital mortality prediction, the AUC for the XGBoost model was 0.816 [95% confidence interval (CI): 0.815-0.818], surpassing the AUC values for the CART model (0.683, 95% CI: 0.680-0.685) and the RF model (0.755, 95% CI: 0.753-0.757). Similarly, in the WHF prediction, the AUC for the XGBoost model was 0.766 (95% CI: 0.765-0.768), outperforming the AUC values for the CART model (0.688, 95% CI: 0.686-0.689) and the RF model (0.713, 95% CI: 0.711-0.714). In the XGBoost model, interpretable clusters were formed, and the rates of in-hospital mortality and WHF were similar among each cluster in both the training and test datasets. The XGBoost models with SHAP analysis provide high prediction performance, interpretability, and reproducible risk stratification for in-hospital mortality and WHF for patients with AHF.

Signal enhancement ratio of multi-phase contrast-enhanced MRI: an imaging biomarker for survival in pancreatic adenocarcinoma.

To evaluate signal enhancement ratio (SER) for tissue characterization and prognosis stratification in pancreatic adenocarcinoma (PDAC), with quantitative histopathological analysis (QHA) as the reference standard. This retrospective study included 277 PDAC patients who underwent multi-phase contrast-enhanced (CE) MRI and whole-slide imaging (WSI) from three centers (2015-2021). SER is defined as (SIlt - SIpre)/(SIea - SIpre), where SIpre, SIea, and SIlt represent the signal intensity of the tumor in pre-contrast, early-, and late post-contrast images, respectively. Deep-learning algorithms were implemented to quantify the stroma, epithelium, and lumen of PDAC on WSIs. Correlation, regression, and Bland-Altman analyses were utilized to investigate the associations between SER and QHA. The prognostic significance of SER on overall survival (OS) was evaluated using Cox regression analysis and Kaplan-Meier curves. The internal dataset comprised 159 patients, which was further divided into training, validation, and internal test datasets (n = 60, 41, and 58, respectively). Sixty-five and 53 patients were included in two external test datasets. Excluding lumen, SER demonstrated significant correlations with stroma (r = 0.29-0.74, all p < 0.001) and epithelium (r = -0.23 to -0.71, all p < 0.001) across a wide post-injection time window (range, 25-300 s). Bland-Altman analysis revealed a small bias between SER and QHA for quantifying stroma/epithelium in individual training, validation (all within ± 2%), and three test datasets (all within ± 4%). Moreover, SER-predicted low stromal proportion was independently associated with worse OS (HR = 1.84 (1.17-2.91), p = 0.009) in training and validation datasets, which remained significant across three combined test datasets (HR = 1.73 (1.25-2.41), p = 0.001). SER of multi-phase CE-MRI allows for tissue characterization and prognosis stratification in PDAC. The signal enhancement ratio of multi-phase CE-MRI can serve as a novel imaging biomarker for characterizing tissue composition and holds the potential for improving patient stratification and therapy in PDAC. Imaging biomarkers are needed to better characterize tumor tissue in pancreatic adenocarcinoma. Signal enhancement ratio (SER)-predicted stromal/epithelial proportion showed good agreement with histopathology measurements across three distinct centers. Signal enhancement ratio (SER)-predicted stromal proportion was demonstrated to be an independent prognostic factor for OS in PDAC.

Atmospheric Chemistry Surrogate Modeling With Sparse Identification of Nonlinear Dynamics

AbstractModeling atmospheric chemistry is computationally expensive and limits the widespread use of chemical transport models. This computational cost arises from solving high‐dimensional systems of stiff differential equations. Previous work has demonstrated the promise of machine learning (ML) to accelerate air quality model simulations but has suffered from numerical instability during long‐term simulations. This may be because previous ML‐based efforts have relied on explicit Euler time integration—which is known to be unstable for stiff systems—and have used neural networks which are prone to overfitting. We hypothesize that parsimonious models combined with modern numerical integration techniques can overcome this limitation. Using a small‐scale mechanism to explore the potential of these methods, we have created a machine‐learned surrogate by (a) reducing dimensionality using singular value decomposition to create an interpretably‐compressed low‐dimensional latent space and (b) using Sparse Identification of Nonlinear Dynamics (SINDy) to create a differential‐equation‐based representation of the underlying dynamics in the compressed latent space with reduced stiffness. The root mean square error of ML model prediction for ozone concentration over 9 days is 37.8% of the root mean square concentration across all simulations in our testing data set. The surrogate model is 10× faster with 12× fewer integration timesteps compared to reference model and is numerically stable in all tested simulations. Overall, we find that SINDy can be used to create fast, stable, and accurate surrogates of a simple photochemical mechanism. In future work, we will explore the application of this method to more detailed mechanisms and their use in large‐scale simulations.

DMGM: deformable-mechanism based cervical cancer staging via MRI multi-sequence.

This study aims to leverage a deep learning approach, specifically a deformable convolutional layer, for staging cervical cancer using multi-sequence MRI images. This is in response to the challenges doctors face in simultaneously identifying multiple sequences, a task that computer-aided diagnosis systems can potentially improve due to their vast information storage capabilities.&#xD;&#xD;Approach: To address the challenge of limited sample sizes, we introduce a sequence enhancement strategy to diversify samples and mitigate overfitting. We propose a novel deformable ConvLSTM module that integrates a deformable mechanism with ConvLSTM, enabling the model to adapt to data with varying structures. Furthermore, we introduce the deformable multi-sequence guidance model (DMGM) as an auxiliary diagnostic tool for cervical cancer staging.&#xD;&#xD;Main Results: Through extensive testing, including comparative and ablation studies, we validate the effectiveness of the deformable ConvLSTM module and the DMGM. Our findings highlight the model's ability to adapt to the deformation mechanism and address the challenges in cervical cancer tumor staging, thereby overcoming the overfitting issue and ensuring the synchronization of asynchronous scan sequences. The research also utilized the multi-modal data from BraTS 2019 as an external test dataset to validate the effectiveness of the proposed methodology presented in this study.&#xD;&#xD;Significance: The DMGM represents the first deep learning model to analyze multiple MRI sequences for cervical cancer, demonstrating strong generalization capabilities and effective staging in small dataset scenarios. This has significant implications for both deep learning applications and medical diagnostics. The source code will be made available subsequently.

New Decomposition Models for Hourly Direct Normal Irradiance Estimations for Southern Africa

This research develops and validates new decomposition models for hourly direct Normal Irradiance (DNI) estimations for Southern African data. Localised models were developed using data collected from the Southern African Universities Radiometric Network (SAURAN). Clustered areas within Southern Africa were identified, and the developed cluster decomposition models highlighted the potential advantages of grouping data based on shared geographical and climatic attributes. This clustering approach could enhance decomposition model performance, particularly when local data are limited or when data are available from multiple nearby stations. Further, a regional Southern African decomposition model, which encompasses a wide spectrum of climatic regions and geographic locations, exhibited notable improvements over the baseline models despite occasional overestimation or underestimation. The results demonstrated improved DNI estimation accuracy compared to the baseline models across all testing and validation datasets. These outcomes suggest that utilising a localised model can significantly enhance DNI estimations for Southern Africa and potentially for developing similar models in diverse geographic regions worldwide. The overall metrics affirm the substantial advancement achieved with the regional model as an accurate decomposition model representing Southern Africa. Two stations were used as a validation study, as an application example where no localised model was available, and the cluster and regional models both outperformed the comparative decomposition models. This study focused on validating the model for hourly DNI in Southern Africa within a range of Kt-intervals from 0.175 to 0.875, and the range could be expanded and validated for future studies. Implementing accurate decomposition models in developing countries can accelerate the adoption of renewable energy sources, diminishing reliance on coal and fossil fuels.

Estimating software reliability using size-biased modelling

Software testing is an important step in software development where inputs are administered repeatedly to detect bugs present in the software. In this paper, we have considered the estimation of total number of bugs and software reliability as a size-biased sampling problem by introducing the concept of eventual bug size as a latent variable. We have developed a Bayesian generalised linear mixed model (GLMM) using software testing detection data to estimate software reliability and stopping phase. The model uses size-biased approach where the probability of detecting a bug is an increasing function of eventual size of the bug which is as an index for the potential number of inputs that may eventually pass through the bug. We have tested the sensitivity of the reliability estimates by varying the number of inputs and detection probability via a simulation study and have found that the key parameters could be accurately estimated. Further, we have applied our model to two empirical data sets – one from a commercial software and the other from ISRO launch mission software testing data set. The hierarchical modelling approach provides a unified modelling framework that may find applications in other fields (e.g. hydrocarbon explorations) apart from software management.

Hierarchical learning of gastric cancer molecular subtypes by integrating multi‐modal DNA‐level omics data and clinical stratification

Hierarchical learning of gastric cancer molecular subtypes by integrating multi‐modal DNA‐level omics data and clinical stratification

Segmentation of mediastinal lymph nodes in CT with anatomical priors.

Lymph nodes (LNs) in the chest have a tendency to enlarge due to various pathologies, such as lung cancer or pneumonia. Clinicians routinely measure nodal size to monitor disease progression, confirm metastatic cancer, and assess treatment response. However, variations in their shapes and appearances make it cumbersome to identify LNs, which reside outside of most organs. We propose to segment LNs in the mediastinum by leveraging the anatomical priors of 28 different structures (e.g., lung, trachea etc.) generated by the public TotalSegmentator tool. The CT volumes from 89 patients available in the public NIH CT Lymph Node dataset were used to train three 3D off-the-shelf nnUNet models to segment LNs. The public St. Olavs dataset containing 15 patients (out-of-training-distribution) was used to evaluate the segmentation performance. For LNs with short axis diameter 8mm, the 3D cascade nnUNet model obtained the highest Dice score of 67.9 ± 23.4 and lowest Hausdorff distance error of 22.8 ± 20.2. For LNs of all sizes, the Dice score was 58.7 ± 21.3 and this represented a 10% improvement over a recently published approach evaluated on the same test dataset. To our knowledge, we are the first to harness 28 distinct anatomical priors to segment mediastinal LNs, and our work can be extended to other nodal zones in the body. The proposed method has the potential for improved patient outcomes through the identification of enlarged nodes in initial staging CT scans.

Rapid in-situ calibration of computational micro-spectrometer with few-shot meta-learning

Computational micro-spectrometers comprised of detector arrays and encoding structure arrays, such as on-chip Fabry-Perot (FP) cavity filters, have great potential in many in-situ applications owing to their compact size and snapshot imaging ability. Given manufacturing deviation and environmental influence are inevitable, easy and effective calibration for spectrometer is necessary, especially for in-situ applications. Currently calibration strategies based on iterative algorithms or neural networks require accurate measurements of pixel-level (spectral) encoding functions through monochromator or large amounts of standard samples. These procedures are time-consuming and expensive, thereby impeding in-situ applications. Meta-learning algorithms with few-shot learning ability can address this challenge by incorporating the prior knowledge in the simulated dataset. In this work, we propose a meta-learning algorithm free of measuring encoding function or large amounts of standard samples to calibrate a micro-spectrometer with manufacturing deviation effectively. Our micro-spectrometer comprises 16 types of FP filters covering a wavelength range of 550-720 nm. The center wavelength of each filter type deviates from the design up to 6 nm. After calibration with 15 different color data, the average reconstruction error on the test dataset decreased from 7.2 × 10 − 3 to 1.2 × 10 − 3, and further decreased to 9.4 × 10 − 4 when the calibration data increased to 24. The performance is comparable to algorithms trained with measured encoding function both in reconstruction error and generalization ability. We estimated that the cost of in-situ calibration through reflectance measurements of color chart decreased to one percent of the cost through monochromator measurements. By exploiting prior deviation information in simulation data with meta-learning, the efficiency and cost of calibration are significantly improved, thereby facilitating the large-scale production and in-situ application of micro-spectrometers.

Diagnostic Identification of Acute Brain Dysfunction in Pediatric Sepsis and Septic Shock in the Electronic Health Record: A Comparison of Four Definitions in a Reference Dataset.

Acute brain dysfunction (ABD) in pediatric sepsis has a prevalence of 20%, but can be difficult to identify. Our previously validated ABD computational phenotype (CPABD) used variables obtained from the electronic health record indicative of clinician concern for acute neurologic or behavioral change. We tested whether the CPABD has better diagnostic performance to identify confirmed ABD than other definitions using the Glasgow Coma Scale or delirium scores. Diagnostic testing in a curated cohort of pediatric sepsis/septic shock patients. Quaternary freestanding children's hospital. The test dataset comprised 527 children with sepsis/septic shock managed between 2011 and 2021 with a prevalence (pretest probability) of confirmed ABD of 30% (159/527). CPABD was based on use of neuroimaging, electroencephalogram, and/or administration of new antipsychotic medication. We compared the performance of the CPABD with three GCS/delirium-based definitions of ABD-Proulx et al, International Pediatric Sepsis Consensus Conference, and Pediatric Organ Dysfunction Information Update Mandate. The posttest probability of identifying ABD was highest in CPABD (0.84) compared with other definitions. CPABD also had the highest sensitivity (83%; 95% CI, 76-89%) and specificity (93%; 95% CI, 90-96%). The false discovery rate was lowest in CPABD (1-in-6) as was the false omission rate (1-in-14). Finally, the prevalence threshold for the definitions varied, with the CPABD being the definition closest to 20%. In our curated dataset of pediatric sepsis/septic shock, CPABD had favorable characteristics to identify confirmed ABD compared with GCS/delirium-based definitions. The CPABD can be used to further study the impact of ABD in studies using large electronic health datasets.

Hybrid feature selection in a machine learning predictive model for perioperative myocardial injury in noncoronary cardiac surgery with cardiopulmonary bypass.

Perioperative myocardial injury (PMI) is associated with increased mobility and mortality after noncoronary cardiac surgery. However, limited studies have developed a predictive model for PMI. Therefore, we used hybrid feature selection (FS) methods to establish a predictive model for PMI in noncoronary cardiac surgery with cardiopulmonary bypass (CPB). This was a single-center retrospective study conducted at the Fuwai Hospital in China. Patients aged 18-70years who underwent elective noncoronary surgery with CPB at our institution from December 2018 to April 2021 were enrolled. The primary outcome was PMI, defined as the postoperative cardiac troponin I (cTnI) levels exceeding 220 times of upper reference limit (URL). Statistical analyses were conducted by Python (Python Software Foundation, version 3.9.7 and integrated development environment Jupyter Notebook 1.1.0) and SPSS software version 26.0 (IBM Corp., Armonk, New York, USA). A total of 1130 patients were eventually eligible for this study. The incidence of PMI was 20.3% (229/1130) in the overall patients, 20.6% (163/791) in the training dataset, and 19.5% (66/339) in the testing dataset. The logistic regression model performed the best AUC of 0.6893 (95 CI%: 0.6371-0.7382) by the traditional selection method, and the random forest model performed the best AUC of 0.6937 (95 CI%: 0.6416-0.7423) by the union of Wrapper and Embedded method, and the CatBoost model performed the best AUC of 0.6828 (95 CI%: 0.6304-0.7320) by the union of Embedded and forward logistic regression technique, and the Naïve Bayes model achieved the best AUC with 0.7254 (95 CI%: 0.6746-0.7723) by forwarding logistic regression method. Moreover, the decision tree, KNeighborsClassifier, and support vector machine models performed the worse AUC in all selection forms. Furthermore, the SHapley Additive exPlanations plot showed that prolonged CPB, aortic clamp time, and preoperative low platelets count were strongly related to the PMI risk. In total, four category feature selection methods were utilized, comprising five individual selection techniques and 15 combined methods. Notably, the combination of logistic regression and embedded methods demonstrated outstanding performance in predicting PMI risk. We also concluded that the machine learning model, including random forest, catboost, and Naive Bayes, were suitable candidates for establishing PMI predictive model. Nevertheless, additional investigation and validation are imperative for substantiating these finding.

MuLA-GAN: Multi-Level Attention GAN for Enhanced Underwater Visibility

The underwater environment presents unique challenges (color distortions, reduced contrast, blurriness) hindering accurate analysis. This work introduces MuLA-GAN, a novel approach leveraging Generative Adversarial Networks (GANs) and specifically adapted Multi-Level Attention for comprehensive underwater image enhancement. MuLA-GAN integrates Multi-Level Attention within the GAN architecture to prioritize learning discriminative features crucial for precise image restoration. These relevant features encompass information on local details within image regions leveraged by spatial attention and features at various scales across the entire image captured by multi-level attention. This allows MuLA-GAN to identify and enhance objects, textures, and edges obscured by underwater distortions while also reconstructing a more accurate and visually clear representation of the underwater scene by analyzing low-level information like edges and textures, as well as high-level information like object shapes and global scene information. By selectively focusing on these relevant features, MuLA-GAN excels at capturing and preserving intricate details in underwater imagery, which is essential for various marine research, exploration, and resource management applications. Extensive evaluations on diverse datasets (UIEB test, UIEB challenge, U45, UCCS) demonstrate MuLA-GAN's superior performance compared to existing methods. Additionally, a specialized bio-fouling and aquaculture dataset confirms the model's robustness in challenging environments. On the UIEB test dataset, MuLA-GAN achieves exceptional Peak Signal-to-Noise Ratio (PSNR) (25.59) and Structural Similarity Index (SSIM) (0.893) scores, surpassing Water-Net (24.36 PSNR, 0.885 SSIM). This work addresses a significant research gap in underwater image enhancement by demonstrating the effectiveness of combining GANs with specifically adapted Multi-Level Attention mechanisms. This tailored approach offers a novel and comprehensive framework for restoring underwater image quality, providing valuable insights for accurate underwater scene analysis. The source code for MuLA-GAN is publicly available on GitHub at https://github.com/AhsanBaidar/MuLA_GAN.git

Development and Validation of an Explainable Deep Learning Model to Predict In-Hospital Mortality for Patients With Acute Myocardial Infarction: Algorithm Development and Validation Study.

Acute myocardial infarction (AMI) is one of the most severe cardiovascular diseases and is associated with a high risk of in-hospital mortality. However, the current deep learning models for in-hospital mortality prediction lack interpretability. This study aims to establish an explainable deep learning model to provide individualized in-hospital mortality prediction and risk factor assessment for patients with AMI. In this retrospective multicenter study, we used data for consecutive patients hospitalized with AMI from the Chongqing University Central Hospital between July 2016 and December 2022 and the Electronic Intensive Care Unit Collaborative Research Database. These patients were randomly divided into training (7668/10,955, 70%) and internal test (3287/10,955, 30%) data sets. In addition, data of patients with AMI from the Medical Information Mart for Intensive Care database were used for external validation. Deep learning models were used to predict in-hospital mortality in patients with AMI, and they were compared with linear and tree-based models. The Shapley Additive Explanations method was used to explain the model with the highest area under the receiver operating characteristic curve in both the internal test and external validation data sets to quantify and visualize the features that drive predictions. A total of 10,955 patients with AMI who were admitted to Chongqing University Central Hospital or included in the Electronic Intensive Care Unit Collaborative Research Database were randomly divided into a training data set of 7668 (70%) patients and an internal test data set of 3287 (30%) patients. A total of 9355 patients from the Medical Information Mart for Intensive Care database were included for independent external validation. In-hospital mortality occurred in 8.74% (670/7668), 8.73% (287/3287), and 9.12% (853/9355) of the patients in the training, internal test, and external validation cohorts, respectively. The Self-Attention and Intersample Attention Transformer model performed best in both the internal test data set and the external validation data set among the 9 prediction models, with the highest area under the receiver operating characteristic curve of 0.86 (95% CI 0.84-0.88) and 0.85 (95% CI 0.84-0.87), respectively. Older age, high heart rate, and low body temperature were the 3 most important predictors of increased mortality, according to the explanations of the Self-Attention and Intersample Attention Transformer model. The explainable deep learning model that we developed could provide estimates of mortality and visual contribution of the features to the prediction for a patient with AMI. The explanations suggested that older age, unstable vital signs, and metabolic disorders may increase the risk of mortality in patients with AMI.