Early prediction of mortality at sepsis diagnosis time in critically ill patients by using interpretable machine learning.

This study applied machine learning for the early prediction of 30-day mortality at sepsis diagnosis time in critically ill patients. Retrospective study using data collected from the Medical Information Mart for Intensive Care IV database. The data of the patient cohort was divided on the basis of the year of hospitalization, into training (2008-2013), validation (2014-2016), and testing (2017-2019) datasets. 24,377 patients with the sepsis diagnosis time < 24 h after intensive care unit (ICU) admission were included. A gradient boosting tree-based algorithm (XGBoost) was used for training the machine learning model to predict 30-day mortality at sepsis diagnosis time in critically ill patients. Model performance was measured in both discrimination and calibration aspects. The model was interpreted using the SHapley Additive exPlanations (SHAP) module. The 30-day mortality rate of the testing dataset was 17.9%, and 39 features were selected for the machine learning model. Model performance on the testing dataset achieved an area under the receiver operating characteristic curve (AUROC) of 0.853 (95% CI 0.837-0.868) and an area under the precision-recall curves of 0.581 (95% CI 0.541-0.619). The calibration plot for the model revealed a slope of 1.03 (95% CI 0.94-1.12) and intercept of 0.14 (95% CI 0.04-0.25). The SHAP revealed the top three most significant features, namely age, increased red blood cell distribution width, and respiratory rate. Our study demonstrated the feasibility of using the interpretable machine learning model to predict mortality at sepsis diagnosis time.

Early prediction of delirium upon intensive care unit admission: Model development, validation, and deployment.

STUDY OBJECTIVE: To develop, validate, and deploy models for predicting delirium in critically ill adult patients as early as upon intensive care unit (ICU) admission. DESIGN: Retrospective cohort study. SETTING: Single university teaching hospital in Taipei, Taiwan. PATIENTS: 6238 critically ill patients from August 2020 to August 2021. MEASUREMENTS: Data were extracted, pre-processed, and split into training and testing datasets based on the time period. Eligible variables included demographic characteristics, Glasgow Coma Scale, vital signs parameters, treatments, and laboratory data. The predicted outcome was delirium, defined as any positive result (a score ≥ 4) of the Intensive Care Delirium Screening Checklist that was assessed by primary care nurses in each 8-h shift within 48 h after ICU admission. We trained models to predict delirium upon ICU admission (ADM) and at 24 h (24H) after ICU admission by using logistic regression (LR), gradient boosted trees (GBT), and deep learning (DL) algorithms and compared the models' performance. MAIN RESULTS: Eight features were extracted from the eligible features to train the ADM models, including age, body mass index, medical history of dementia, postoperative intensive monitoring, elective surgery, pre-ICU hospital stays, and GCS score and initial respiratory rate upon ICU admission. In the ADM testing dataset, the incidence of ICU delirium occurred within 24 h and 48 h was 32.9% and 36.2%, respectively. The area under the receiver operating characteristic curve (AUROC) (0.858, 95% CI 0.835-0.879) and area under the precision-recall curve (AUPRC) (0.814, 95% CI 0.780-0.844) for the ADM GBT model were the highest. The Brier scores of the ADM LR, GBT, and DL models were 0.149, 0.140, and 0.145, respectively. The AUROC (0.931, 95% CI 0.911-0.949) was the highest for the 24H DL model and the AUPRC (0.842, 95% CI 0.792-0.886) was the highest for the 24H LR model. CONCLUSION: Our early prediction models based on data obtained upon ICU admission could achieve good performance in predicting delirium occurred within 48 h after ICU admission. Our 24-h models can improve delirium prediction for patients discharged >1 day after ICU admission.

Inhibitory effect of Malaysian coastal plants on banana (Musa acuminata colla "Lakatan"), ginger (Zingiber officinale Roscoe) and sweet potato (Ipomoea batatas) polyphenol oxidase.

Undesired browning reaction catalyzed by polyphenol oxidase (PPO) has reduced the nutritional quality and customer acceptance of the products. The inhibitory effects of six coastal plants including Sonneratia alba, Rhizophora apiculata, Syzygium grande, Rhizophora mucronata, Hibiscus tiliaceus and Bruguiera gymnorhiza on PPO in banana, sweet potato and ginger were studied based on oxidation of pyrocatechol. Banana exhibited the highest PPO activity (141,600 U), followed by sweet potato (43,440 U) and ginger (26,880 U). Banana PPO was strongly inhibited by R. apiculata (70.87%) and sweet potato PPO was strongly inhibited by S. alba (82.00%). In general, most banana PPO was the most susceptible to inhibition with all inhibitors having inhibition higher than 60.00% at 0.5 mg/ml and ginger PPO was the least susceptible with all inhibitors showing less than 50.00% inhibition at 0.5 mg/ml. Coastal plant extracts are potentially to be developed as natural inhibitors for preventing enzymatic browning of fruits and vegetables.

Appropriate locations of fixed bearings of continuous beams considering rail-bridge thermal interaction.

Due to the rail-bridge thermal interaction, the high additional axial force in continuously welded rails on continuous bridges may lead to rail buckling or breaking. However, there is little research on the influence of the location of the fixed bearing of continuous beam on the additional force of rail. In order to study the influence of bridge bearing arrangement on the additional longitudinal force of CWR, the thermal interaction model is established for rail, and simple and continuous beams considering nonlinear stiffness and the methods are proposed to determine the locations of fixed bearings of continuous beams corresponding to the maximum additional forces in rail reaching minimum values. Multiple continuous beams with several different lengths and simple beams with three types of bearing arrangements are taken into account to find the effect laws of the locations of the fixed bearings of continuous beams on the maximum additional forces in rail. The results show that as long as the same number of continuous beams, the ratios of the distances of adjacent two fixed bearings to the distance between the two fixed bearings of the simple beams neighbour to the first and last continuous beams respectively are approximately equal to each other. Furthermore the appropriate locations of the fixed bearings of continuous beams are recommended. The results can guide designing the location of the fixed bearing of continuous railway bridge while reducing the additional axial force in continuously welded rails due to bridge thermal effect.

The development an artificial intelligence algorithm for early sepsis diagnosis in the intensive care unit.

BACKGROUND: Severe sepsis and septic shock are still the leading causes of death in Intensive Care Units (ICUs), and timely diagnosis is crucial for treatment outcomes. The progression of electronic medical records (EMR) offers the possibility of storing a large quantity of clinical data that can facilitate the development of artificial intelligence (AI) in medicine. However, several difficulties, such as poor structure and heterogenicity of the raw EMR data, are encountered when introducing AI with ICU data. Labor-intensive work, including manual data entry, personal medical records sorting, and laboratory results interpretation may hinder the progress of AI. In this article, we introduce the developing of an AI algorithm designed for sepsis diagnosis using pre-selected features; and compare the performance of the AI algorithm with SOFA score based diagnostic method. MATERIALS AND METHODS: This is a prospective open-label cohort study. A specialized EMR, named TED_ICU, was implemented for continuous data recording. One hundred six clinical features relevant to sepsis diagnosis were selected prospectively. A labeling work to allocate SEPSIS or NON_SEPSIS status for each ICU patient was performed by the in-charge intensivist according to SEPSIS-3 criteria, along with the automatic recording of selected features every day by TED_ICU. Afterward, we use de-identified data to develop the AI algorithm. Several machine learning methods were evaluated using 5-fold cross-validation, and XGBoost, a decision-tree based algorithm was adopted for our AI algorithm development due to best performance. RESULTS: The study was conducted between August 2018 and December 2018 for the first stage of analysis. We collected 1588 instances, including 444 SEPSIS and 1144 NON-SEPSIS, from 434 patients. The 434 patients included 259 (59.6%) male patients and 175 female patients. The mean age was 67.6-year-old, and the mean APACHE II score was 13.8. The SEPSIS cohort had a higher SOFA score and increased use of organ support treatment. The AI algorithm was developed with a shuffle method using 80% of the instances for training and 20% for testing. The established AI algorithm achieved the following: accuracy = 82% ± 1%; sensitivity = 65% ± 5%; specificity = 88% ± 2%; precision = 67% ± 3%; and F1 = 0.66 ± 0.02. The area under the receiver operating characteristic curve (AUROC) was approximately 0.89. The SOFA score was used on the same 1588 instances for sepsis diagnosis, and the result was inferior to our AI algorithm (AUROC = 0.596). CONCLUSION: Using real-time data, collected by EMR, from the ICU daily practice, our AI algorithm established with pre-selected features and XGBoost can provide a timely diagnosis of sepsis with an accuracy greater than 80%. AI algorithm also outperforms the SOFA score in sepsis diagnosis and exhibits practicality as clinicians can deploy appropriate treatment earlier. The early and precise response of this AI algorithm will result in cost reduction, outcome improvement, and benefit for healthcare systems, medical staff, and patients as well.

Modulation of optical focusing by using optimized zone plate structures.

The focusing properties of the optimized zone plate structures which have upper and lower zones with different thicknesses are studied by the three-dimensional finite-difference time-domain method. Two kinds of materials are chosen, including silver representing metal and BK7 glass representing dielectric. An optimization algorithm is applied to tune the parameters of zone plate structures. Several optimized zone plate structures with smaller circular-shape focus are presented. By using the angular spectrum representation method, we found that the cases with smaller focal sizes have larger high-k components; however, the intensities of side lobes also become larger in comparison with the main beam. It is also found that the phase differences between different spatial field components can have the influences on focusing properties. A special case with two focuses is shown by changing the cost function of the same optimization algorithm. Our findings suggest that the optimized zone plate structures can reconstruct the light intensity distribution and have a great potential for the applications in imaging, lithography, and data storage.

Medulloblastoma-derived tumor stem-like cells acquired resistance to TRAIL-induced apoptosis and radiosensitivity.

OBJECTS: Medulloblastoma (MB) is the most malignant primary brain tumor in early childhood that contains cellular and functional heterogeneity. Recent evidence has demonstrated that the tumor stem cells (TSC) may explain the radiochemoresistance of brain tumors, including MB. The aim of the present study is to investigate the possible role of TNF-related apoptosis-inducing ligand (TRAIL) in viability and tumorigenicity of MB cells and MB-derived TSC. METHODS: MB-associated TSC were isolated and cultured by serum-free medium with bFGF and EGF. The parental MB cells and MB-TSC cells were treated with TRAIL in different concentrations and assessed for cell viability, invasion ability, colony forming ability, and radiotherapy effect. RESULTS: We enrich a subpopulation of MB-TSC cells using tumor spheroid formation approach. MB-TSC display enhanced self-renewal and highly expressed "stemness" genes (CD133, Sox-2, Bmi1, Nestin). Additionally, MB-TSC showed significant resistance to TRAIL-induced apoptosis and radiosensitivity compared to the parental MB cells due antiapoptotic gene (c-FLIP, Caspase 8, Bcl-2, and Bax) upregulation. CONCLUSIONS: Our data suggest that MB-TSC are resistant to TRAIL-induced apoptosis and tumorigenic properties. Understanding the molecular mechanisms by which to operate the physiological characteristics in MB-TSC cells offers attractive approach for MB treatment.

The influence of propagating and evanescent waves on the focusing properties of zone plate structures.

The field properties of Fresnel zone plates with wavelength-scale focal distances were numerically investigated using the finite-difference time-domain method. The fields in the focal planes are analyzed using the angular spectrum representation, and the components of the propagating and evanescent waves are reconstructed. It was found that, in the focal plane of silver zone plates, there were more evanescent waves and the propagating waves occurred at higher spatial frequencies relative to glass zone plates. The propagating and evanescent wave components vary with the material and the number of zones in the zone plate structures. Our findings suggest that more evanescent waves and higher spatial frequency components of propagating waves can shape the field and obtain a smaller focus.