scholarly journals Real-time clinical decision supports the prediction and management of in-hospital acute kidney injury with recurrent neural networks (Preprint)

2020 ◽  
Author(s):  
Kipyo Kim ◽  
Hyeonsik Yang ◽  
Suryeong Go ◽  
Hyung-Eun Son ◽  
Ji-Young Ryu ◽  
...  
Keyword(s):  
Neural Network ◽  
Real Time ◽  
Prediction Models ◽  
External Validation ◽  
Kidney Injury ◽  
Clinical Decision ◽  
The Future ◽  
National University ◽  
Seoul National University ◽  
Clinical Decision Supports

BACKGROUND Acute kidney injury(AKI) is commonly encountered in clinical practice and associated with poor patient outcomes and increased healthcare costs. AKI poses significant challenges for clinicians but effective measures for the prediction and prevention of AKI are lacking. Previously published AKI prediction models mostly have simple design without external validation. Furthermore, little is known about how to link the model output and clinical decision supports due to the blackbox nature of the neural network models. OBJECTIVE We aimed to present an externally validated recurrent neural network (RNN)-based prediction model for in-hospital AKI, and to show the explainability of the model in relation to clinical decision support. METHODS Study populations were all patients aged ≥ 18 years and hospitalized more than a week from 2013 to 2017 in two tertiary hospitals in Korea (Seoul National University Bundang Hospital and Seoul National University Hospital). All demographics, laboratory values, vital signs, and clinical conditions were obtained from the EHR of each hospital. A total of 102 variables included in the model. Each variable falls into two categories: static and dynamic variables. We developed two-stage hierarchical prediction models (model 1 and model 2) using RNN algorithms. The outcome variable for Model 1 was the occurrence of AKI within 7 days from the present. Model 2 predicts the future trajectory of Cr values up to 72 hours. Internal validation was performed by 5-fold cross validation using the training set, and then external validation was done using independent test set. RESULTS Of a total of 118,893 patients initially screened, after excluding cases with missing data and estimated glomerular filtration rate <15 ml/min/1.73m2 or end-stage kidney disease, 40,552 patients in training cohort and 4,084 in external validation cohort (test cohort) were used for model development. Model 1 with the observation window of 3 days predicts AKI development with the area under the curve of 0.80 (sensitivity 0.72, specificity 0.89) in external validation. The model 2 predicted the future creatinine values within 3 days with the mean square errors of 0.04-0.06 for patients with higher risks of AKI and 0.05-0.12 for those with lower risks. On the basis of the developed models, we showed the probability of AKI according to the feature values in total patients and each individual with partial dependence plots and individual conditional expectation plots. In addition, we estimated the effects of feature modifications such as nephrotoxic drug discontinuation on the future creatinine levels. CONCLUSIONS We developed and externally validated a real-time AKI prediction model using RNN algorithms. Our model could provide real-time assessment of future AKI occurrences and individualized risk factors for AKI in general inpatient cohorts. These suggest approaches to support clinical decisions based on the prediction models for in-hospital AKI.

P0616REAL-TIME CLINICAL DECISION SUPPORT TO PREDICT AND MANAGE IN-HOSPITAL ACUTE KIDNEY INJURY UTILIZING RECURRENT NEURAL NETWORK

2020 ◽  
Vol 35 (Supplement_3) ◽  
Author(s):  
Hea Eun Kim ◽  
Hyeonsik Yang ◽  
Sejoong Kim ◽  
Kipyo Kim
Keyword(s):  
Prediction Model ◽  
Validation Cohort ◽  
External Validation ◽  
Vital Signs ◽  
Clinical Decision ◽  
Internal Validation ◽  
Training Cohort ◽  
National University ◽  
Seoul National University ◽  
Dynamic Variables

Abstract Background and Aims Rapidly Increasing electronic health record (EHR) data and recent development of machine learning methods offers the possibilities of improvement in quality of care in clinical practice. Machine learning can incorporate huge amount of features into the model, and enable non-linear algorithms with great performance. Previously published AKI prediction models have simple design without real-time assessment. Major risk factors in in-hospital AKI include use of various nephrotoxins, repeatedly measured laboratory findings, and vital signs, which are dynamic variables rather than static. Given that recurrent neural network (RNN) is a powerful tool to handle the sequential data, using RNN method in the prediction model is a promising approach. Therefore, in the present study, we proposed a RNN-based prediction model with external validation for in-hospital AKI and aimed to provide a framework to link the developed model with clinical decision supports. Method Study populations were all patients aged ≥ 18 years and hospitalized more than a week at Seoul National University Bundang Hospital (SNUBH) from 2013 to 2017 (training cohort) and at Seoul National University Hospital (SNUH) in 2017 (validation cohort). All demographics, laboratory values, vital signs, and clinical conditions were obtained from the EHR of each hospital. A total of 102 variables included in the model. Each variable falls into two categories: static and dynamic variable; static variable was time-invariant values during hospitalization, and dynamic variables were daily-updated values. Baseline creatinine was determined by searching the minimum serum Cr level within 2 weeks before admission. We developed two different models (model 1 and model 2) using RNN algorithms. The outcome for model 1 was the occurrence of AKI within 7 days from the present. In model 2, we constructed the prediction model of the trajectory of Cr values after 24 hours, 48 hours, and 72 hours, using available Cr values from 7 days ago to the present. Internal validation was performed by 5-fold cross validation using the training set (SNUBH), and then external validation was done using test set (SNUH). Results A total of 40,552 patients in training cohort and 4,000 patients in external validation cohort (test cohort) were included in the study. The mean age of participants was 62.2 years in training cohort and 58.7 years in test cohort. Baseline eGFR was 93.8 ± 40.4 ml/min/1.73m2 in training cohort and 88.4 ± 23.2 ml/min/1.73m2 in test cohort. In model 1 for the prediction of AKI occurrence within 7 days, the area under the curve was 0.93 (sensitivity 0.90, specificity 0.96) in internal validation, and 0.83 (sensitivity 0.83, specificity 0.82) in external validation. The model 2 predicted the creatinine trajectory within 3 days accurately; root mean square error was 0.1 in training cohort and 0.3 in test cohort. To support the clinical decision for AKI manage, we estimated the predicted trajectories of future creatinine levels after renal insult removal, such as nephrotoxic drugs, based on the established model 2. Conclusion We developed and validated a real-time AKI prediction model using RNN algorithms. This model showed high performance and can accurately visualize future creatinine trajectories. In addition, the model can provide the information about modifiable factors in patients with high risk of AKI.

scholarly journals Early Identification of Patients with Acute Gastrointestinal Bleeding using Electronic Health Record Phenotyping

2020 ◽  
Author(s):  
Dennis Shung ◽  
Cynthia Tsay ◽  
Loren Laine ◽  
Prem Thomas ◽  
Caitlin Partridge ◽  
...  
Keyword(s):  
Risk Stratification ◽  
Gastrointestinal Bleeding ◽  
Real Time ◽  
Decision Rule ◽  
Language Processing ◽  
Clinical Decision Making ◽  
External Validation ◽  
Clinical Decision ◽  
Internal Validation ◽  
Electronic Health

Background and AimGuidelines recommend risk stratification scores in patients presenting with gastrointestinal bleeding (GIB), but such scores are uncommonly employed in practice. Automation and deployment of risk stratification scores in real time within electronic health records (EHRs) would overcome a major impediment. This requires an automated mechanism to accurately identify (“phenotype”) patients with GIB at the time of presentation. The goal is to identify patients with acute GIB by developing and evaluating EHR-based phenotyping algorithms for emergency department (ED) patients.MethodsWe specified criteria using structured data elements to create rules for identifying patients, and also developed a natural-language-processing (NLP)-based algorithm for automated phenotyping of patients, tested them with tenfold cross-validation (n=7144) and external validation (n=2988), and compared them with the standard method for encoding patient conditions in the EHR, Systematized Nomenclature of Medicine (SNOMED). The gold standard for GIB diagnosis was independent dual manual review of medical records. The primary outcome was positive predictive value (PPV).ResultsA decision rule using GIB-specific terms from ED triage and from ED review-of-systems assessment performed better than SNOMED on internal validation (PPV=91% [90%-93%] vs. 74% [71%-76%], P<0.001) and external validation (PPV=85% [84%-87%] vs. 69% [67%-71%], P<0.001). The NLP algorithm (external validation PPV=80% [79-82%]) was not superior to the structured-datafields decision rule.ConclusionsAn automated decision rule employing GIB-specific triage and review-of-systems terms can be used to trigger EHR-based deployment of risk stratification models to guide clinical decision-making in real time for patients with acute GIB presenting to the ED.

Technology Focus: Offshore Drilling and Completion (October 2021)

2021 ◽  
Vol 73 (10) ◽  
pp. 45-45
Author(s):  
Martin Rylance
Keyword(s):  
Machine Learning ◽  
Real Time ◽  
Oil And Gas ◽  
Hybrid Approach ◽  
Data Communication ◽  
Oil And Gas Industry ◽  
Oil Field ◽  
Real Time Processing ◽  
National University ◽  
Seoul National University

Communication and prediction are symmetrical. Communication, in effect, is prediction about what has happened. And prediction is communication about what is going to happen. Few industries contain as many phases, steps, and levels of interface between the start and end product as the oil and gas industry—field, office, offshore, plant, subsea, downhole, not to mention the disciplinary, functional, managerial, logistics handovers, and boundaries that exist. It therefore is hardly surprising that communication, in all its varied forms, is at the very heart of our business. The papers selected this month demonstrate how improved communication can deliver the prediction required for a variety of reasons, including safety, efficiency, and informational purposes. The application of new and exciting ways of working, partially accelerated by recent events, is leading to breakthrough improvements on all levels. Real-time processing, improved visualization, and predictive and machine-learning methods, as well as improvements in all forms of data communication, are all contributing to incremental enhancements across the board. This month, I encourage the reader to review the selected articles and determine where and how the communication and prediction are occurring and what they are delivering. Then perhaps consider performing an exercise wherein your own day-to-day roles—your own areas of communication, interfacing, and cooperation—are reviewed to see what enhancements you can make as an individual. You may be pleasantly surprised that some simple tweaks to your communication style, frequency, and format can deliver quick wins. In an era of remote working for many individuals, it is an exercise that has some value. Recommended additional reading at OnePetro: www.onepetro.org. OTC 30184 - Augmented Machine-Learning Approach of Rate-of-Penetration Prediction for North Sea Oil Field by Youngjun Hong, Seoul National University, et al. OTC 31278 - A Digital Twin for Real-Time Drilling Hydraulics Simulation Using a Hybrid Approach of Physics and Machine Learning by Prasanna Amur Varadarajan, Schlumberger, et al. OTC 31092 - Integrated Underreamer Technology With Real-Time Communication Helped Eliminate Rathole in Exploratory Operation Offshore Nigeria by Raphael Chidiogo Ozioko, Baker Hughes, et al.

scholarly journals Real-time utilisation of administrative data in the ED to identify older patients at risk: development and validation of the Dynamic Silver Code

BMJ Open ◽  
2019 ◽  
Vol 9 (12) ◽  
pp. e033374 ◽  
Author(s):  
Daniela Balzi ◽  
Giulia Carreras ◽  
Francesco Tonarelli ◽  
Luca Degli Esposti ◽  
Paola Michelozzi ◽  
...  
Keyword(s):  
At Risk ◽  
Cohort Study ◽  
Real Time ◽  
Administrative Data ◽  
Older Patients ◽  
Clinical Decision Making ◽  
External Validation ◽  
Clinical Decision ◽  
Patients At Risk ◽  
Ed Visits

ObjectiveIdentification of older patients at risk, among those accessing the emergency department (ED), may support clinical decision-making. To this purpose, we developed and validated the Dynamic Silver Code (DSC), a score based on real-time linkage of administrative data.Design and settingThe ‘Silver Code National Project (SCNP)’, a non-concurrent cohort study, was used for retrospective development and internal validation of the DSC. External validation was obtained in the ‘Anziani in DEA (AIDEA)’ concurrent cohort study, where the DSC was generated by the software routinely used in the ED.ParticipantsThe SCNP contained 281 321 records of 180 079 residents aged 75+ years from Tuscany and Lazio, Italy, admitted via the ED to Internal Medicine or Geriatrics units. The AIDEA study enrolled 4425 subjects aged 75+ years (5217 records) accessing two EDs in the area of Florence, Italy.InterventionsNone.Outcome measuresPrimary outcome: 1-year mortality. Secondary outcomes: 7 and 30-day mortality and 1-year recurrent ED visits.ResultsAdvancing age, male gender, previous hospital admission, discharge diagnosis, time from discharge and polypharmacy predicted 1-year mortality and contributed to the DSC in the development subsample of the SCNP cohort. Based on score quartiles, participants were classified into low, medium, high and very high-risk classes. In the SCNP validation sample, mortality increased progressively from 144 to 367 per 1000 person-years, across DSC classes, with HR (95% CI) of 1.92 (1.85 to 1.99), 2.71 (2.61 to 2.81) and 5.40 (5.21 to 5.59) in class II, III and IV, respectively versus class I (p<0.001). Findings were similar in AIDEA, where the DSC predicted also recurrent ED visits in 1 year. In both databases, the DSC predicted 7 and 30-day mortality.ConclusionsThe DSC, based on administrative data available in real time, predicts prognosis of older patients and might improve their management in the ED.

STEP-OP: Short-term Event Prediction in the Operating Room using Hybrid Deep Learning to Forecast Five-Minute Intraoperative Hypotension (Preprint)

2021 ◽  
Author(s):  
Sooho Choe ◽  
Eunjeong Park ◽  
Wooseok Shin ◽  
Bonah Koo ◽  
Dongjin Shin ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Deep Learning ◽  
Weighted Average ◽  
University Hospital ◽  
Short Term ◽  
Event Prediction ◽  
Intraoperative Hypotension ◽  
National University ◽  
Seoul National University

BACKGROUND Intraoperative hypotension has an adverse impact on postoperative outcomes, However, it is difficult to predict and treat intraoperative hypotension with individual clinical parameter in advance. OBJECTIVE To develop a prediction model to forecast five-minute intraoperative hypotension based on the weighted average ensemble of individual neural networks, which utilize the biosignals recorded during non-cardiac surgery. METHODS In this retrospective observational study, arterial wave form was recorded during non-cardiac operation held between August 2016 and December 2019, at Seoul National University Hospital, Seoul, South Korea. We analyzed the arterial waveforms from the big data in VitalDB repository of electronic health records. We defined 2 s hypotension as the moving average of arterial pressure under 65 mm Hg for 2 s, and intraoperative hypotensive events as the case in which 2 s hypotension lasts for at least 60 s. We developed an artificial intelligence-enabled process called short-term event prediction in the operating room (STEP-OP) for predicting short-term intraoperative hypotension. RESULTS The study was performed on 18,813 subjects undergoing non-cardiac surgeries. Deep-learning algorithms (convolutional neural network [CNN] and recurrent neural network [RNN]) using raw waveforms as input showed a greater area under the precision-recall curve (AUPRC) scores than the logistic regression algorithm (0.698 [95% confidence interval {CI}, 0.690–0.705], 0.706 [95% CI, 0.698–0.715]), compared with 0.673 (95% CI, 0.665–0.682), respectively. STEP-OP performed better and had greater AUPRC values than RNN and CNN algorithms (0.716 [95% CI, 0.708–0.723]). CONCLUSIONS We developed STEP-OP, the weighted average of deep-learning models. It predicted intraoperative hypotension more accurately than the CNN, RNN, and logistic regression models. CLINICALTRIAL The study was approved by the institutional review board of Seoul National University Hospital (H-2008-175-1152). (Trial Registration: ClinicalTrials.gov NCT02914444). Arterial Pressure; artificial intelligence; biosignals; deep learning; hypotension; machine learning

Development of a real-time control strategy with artificial neural network for automatic control of a continuous-flow sequencing batch reactor

2001 ◽  
Vol 44 (1) ◽  
pp. 95-104 ◽  
Author(s):  
B. C. Cho ◽  
S.-L. Liaw ◽  
C.-N. Chang ◽  
R.-F. Yu ◽  
S.-J. Yang ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Real Time ◽  
Control Strategy ◽  
Continuous Flow ◽  
Prediction Models ◽  
Process Models ◽  
Real Time Control ◽  
Time Control ◽  
Artificial Neural

The purpose of this study is to develop a reliable and effective real-time control strategy by integrating artificial neural network (ANN) process models to perform automatic operation of a dynamic continuous-flow SBR system. The ANN process models, including ORP/pH simulation models and water quality ([NH4+-N] and [NOx--N]) prediction models, can assist in real-time searching the ORP and pH control points and evaluating the operation performances of aerobic nitrification and anoxic denitrification operation phases. Since the major biological nitrogen removal mechanisms were controlled at nitritification (NH4+-N→NO2--N) and denitritification (NO2--N→N2) stages, as well as the phosphorus uptake and release could be completely controlled during aerobic and anoxic operation phases, the system operation performances under this ANN real-time control system revealed that both the aeration time and overall hydraulic retention time could be shortened to about 1.9-2.5 and 4.8-6.2 hrs/cycle respectively. The removal efficiencies of COD, ammonia nitrogen, total nitrogen, and phosphate were 98%, 98%, 97%, and 84% respectively, which were more effective and efficient than under conventional fixed-time control approach.

scholarly journals Derivation and External Validation of Prediction Models for Advanced Chronic Kidney Disease Following Acute Kidney Injury

JAMA ◽  
2017 ◽  
Vol 318 (18) ◽  
pp. 1787 ◽  
Author(s):  
Matthew T. James ◽  
Neesh Pannu ◽  
Brenda R. Hemmelgarn ◽  
Peter C. Austin ◽  
Zhi Tan ◽  
...  
Keyword(s):  
Chronic Kidney Disease ◽  
Acute Kidney Injury ◽  
Kidney Disease ◽  
Prediction Models ◽  
External Validation ◽  
Kidney Injury ◽  
Advanced Chronic Kidney Disease

scholarly journals Light People: Professor Byoungho Lee

2022 ◽  
Vol 11 (1) ◽  
Author(s):  
Hui Wang ◽  
Cun Yu
Keyword(s):  
Mixed Reality ◽  
Active Role ◽  
Optical Components ◽  
Academic Organizations ◽  
The North ◽  
The Future ◽  
National University ◽  
Seoul National University ◽  
The University ◽  
North And South

EditorialMajor developments were made recently in both VR (virtual reality) and AR (augmented reality) technologies, which became the focus of attention. In recent years, MR (mixed reality) technology has also emerged, and optical components play an irreplaceable role in these technologies.Professor Byoungho Lee, who graduated from the University of California at Berkeley and currently works at Seoul National University in South Korea, has been committed to the development of optical components used in VR and AR technologies. As a pioneer of optical electronics in Korea, he is involved in various well-known academic organizations in the optical field, such as the Optica, SPIE, and IEEE, as well as serving as the president of the Optical Society of Korea, leading the direction of the development of optical industry in Korea. As the ambassador of China-Korea Optoelectronics Exchange, Prof. Lee has also played an active role in Chinese optical events and activities. Over the years, he and the Journal Light: Science & Applications (LIGHT) have made progress together and have both made their names in the vast field of optoelectronics.So where did the story between Prof. Lee and the LIGHT journal begin? And what kind of link does the professor have with Changchun Institute of Optics, Fine Mechanics and Physics (CIOMP)? How did he become a pioneer in optoelectronics technology? These are the questions we are eager to ask Prof. Byoungho Lee.The future cannot be predicted, but it can be invented, said Dennis Gabor who had invented holography. The pace of human technological advancements has never stopped. Who is to say that we cannot take a virtual tour of the Palace Museum or explore the north and south poles in the future? Scientists like Prof. Lee are working hard to use technology to provide mankind with an intelligent lifestyle, and lead a new technological trend. I am sure we are all looking forward to it.

Biostatistiek versus machine learning: van traditionele predictiemodellen naar geautomatiseerde medische analyse

2021 ◽  
Author(s):  
M. VALKEMA ◽  
H. LINGSMA ◽  
P. LAMBIN ◽  
J. VAN LANSCHOT
Keyword(s):  
Machine Learning ◽  
Prediction Models ◽  
External Validation ◽  
Clinical Decision Support Systems ◽  
Clinical Decision ◽  
Learning Models ◽  
Selection Of Variables ◽  
Medical Analysis ◽  
Machine Learning Models ◽  
Selection Of

Biostatistics versus machine learning: from traditional prediction models to automated medical analysis Machine learning is increasingly applied to medical data to develop clinical prediction models. This paper discusses the application of machine learning in comparison with traditional biostatistical methods. Biostatistics is well-suited for structured datasets. The selection of variables for a biostatistical prediction model is primarily knowledge-driven. A similar approach is possible with machine learning. But in addition, machine learning allows for analysis of unstructured datasets, which are e.g. derived from medical imaging and written texts in patient records. In contrast to biostatistics, the selection of variables with machine learning is mainly data-driven. Complex machine learning models are able to detect nonlinear patterns and interactions in data. However, this requires large datasets to prevent overfitting. For both machine learning and biostatistics, external validation of a developed model in a comparable setting is required to evaluate a model’s reproducibility. Machine learning models are not easily implemented in clinical practice, since they are recognized as black boxes (i.e. non-intuitive). For this purpose, research initiatives are ongoing within the field of explainable artificial intelligence. Finally, the application of machine learning for automated imaging analysis and development of clinical decision support systems is discussed.

Sign in / Sign up

Export Citation Format

Share Document