RESUMEN
Hypoglycemia is a common metabolic disorder that occurs in the neonatal period. Early identification of neonates at risk of developing hypoglycemia can optimize therapeutic strategies in neonatal care. This study aims to develop a machine learning model and implement a predictive application to assist clinicians in accurately predicting the risk of neonatal hypoglycemia within four hours after birth. Our retrospective study analyzed data from neonates born ≥35 weeks gestational age and admitted to the well-baby nursery between 1 January 2011 and 31 August 2021. We collected electronic medical records of 2687 neonates from a tertiary medical center in Southern Taiwan. Using 12 clinically relevant features, we evaluated nine machine learning approaches to build the predictive models. We selected the models with the highest area under the receiver operating characteristic curve (AUC) for integration into our hospital information system (HIS). The top three AUC values for the early neonatal hypoglycemia prediction models were 0.739 for Stacking, 0.732 for Random Forest and 0.732 for Voting. Random Forest is considered the best model because it has a relatively high AUC and shows no significant overfitting (accuracy of 0.658, sensitivity of 0.682, specificity of 0.649, F1 score of 0.517 and precision of 0.417). The best model was incorporated in the web-based application integrated into the hospital information system. Shapley Additive Explanation (SHAP) values indicated mode of delivery, gestational age, multiparity, respiratory distress, and birth weight < 2500 gm as the top five predictors of neonatal hypoglycemia. The implementation of our machine learning model provides an effective tool that assists clinicians in accurately identifying at-risk neonates for early neonatal hypoglycemia, thereby allowing timely interventions and treatments.
RESUMEN
In this research, we aimed to harness machine learning to predict the imminent risk of acute exacerbation in chronic obstructive pulmonary disease (AECOPD) patients. Utilizing retrospective data from electronic medical records of two Taiwanese hospitals, we identified 26 critical features. To predict 3- and 6-month AECOPD occurrences, we deployed five distinct machine learning algorithms alongside ensemble learning. The 3-month risk prediction was best realized by the XGBoost model, achieving an AUC of 0.795, whereas the XGBoost was superior for the 6-month prediction with an AUC of 0.813. We conducted an explainability analysis and found that the episode of AECOPD, mMRC score, CAT score, respiratory rate, and the use of inhaled corticosteroids were the most impactful features. Notably, our approach surpassed predictions that relied solely on CAT or mMRC scores. Accordingly, we designed an interactive prediction system that provides physicians with a practical tool to predict near-term AECOPD risk in outpatients.
RESUMEN
OBJECTIVE: Artificial intelligence (AI) prediction is increasingly used for decision making in health care, but its application for adverse outcomes in emergency department (ED) patients with acute pancreatitis (AP) is not well understood. This study aimed to clarify this aspect. METHODS: Data from 8274 ED patients with AP in three hospitals from 2009 to 2018 were analyzed. Demographic data, comorbidities, laboratory results, and adverse outcomes were included. Six algorithms were evaluated, and the one with the highest area under the curve (AUC) was implemented into the hospital information system (HIS) for real-time prediction. Predictive accuracy was compared between the AI model and Bedside Index for Severity in Acute Pancreatitis (BISAP). RESULTS: The mean ± SD age was 56.1 ± 16.7 years, with 67.7% being male. The AI model was successfully implemented in the HIS, with Light Gradient Boosting Machine (LightGBM) showing the highest AUC for sepsis (AUC 0.961) and intensive care unit (ICU) admission (AUC 0.973), and eXtreme Gradient Boosting (XGBoost) showing the highest AUC for mortality (AUC 0.975). Compared to BISAP, the AI model had superior AUC for sepsis (BISAP 0.785), ICU admission (BISAP 0.778), and mortality (BISAP 0.817). CONCLUSIONS: The first real-time AI prediction model implemented in the HIS for predicting adverse outcomes in ED patients with AP shows favorable initial results. However, further external validation is needed to ensure its reliability and accuracy.
Asunto(s)
Pancreatitis , Sepsis , Humanos , Masculino , Adulto , Persona de Mediana Edad , Anciano , Femenino , Pancreatitis/complicaciones , Pancreatitis/diagnóstico , Pancreatitis/terapia , Índice de Severidad de la Enfermedad , Inteligencia Artificial , Enfermedad Aguda , Reglas de Decisión Clínica , Reproducibilidad de los Resultados , Pronóstico , Estudios Retrospectivos , Valor Predictivo de las PruebasRESUMEN
Precocious puberty in girls is defined as the onset of pubertal changes before 8 years of age, and gonadotropin-releasing hormone (GnRH) agonist treatment is available for central precocious puberty (CPP). The gold standard for diagnosing CPP is the GnRH stimulation test. However, the GnRH stimulation test is time-consuming, costly, and requires repeated blood sampling. We aimed to develop an artificial intelligence (AI) prediction model to assist pediatric endocrinologists in decision making regarding the optimal timing to perform the GnRH stimulation test. We reviewed the medical charts of 161 girls who received the GnRH stimulation test from 1 August 2010 to 31 August 2021, and we selected 15 clinically relevant features for machine learning modeling. We chose the models with the highest area under the receiver operating characteristic curve (AUC) to integrate into our computerized physician order entry (CPOE) system. The AUC values for the CPP diagnosis prediction model (LH ≥ 5 IU/L) were 0.884 with logistic regression, 0.912 with random forest, 0.942 with LightGBM, and 0.942 with XGBoost. For the Taiwan National Health Insurance treatment coverage prediction model (LH ≥ 10 IU/L), the AUC values were 0.909, 0.941, 0.934, and 0.881, respectively. In conclusion, our AI predictive system can assist pediatric endocrinologists when they are deciding whether a girl with suspected CPP should receive a GnRH stimulation test. With proper use, this prediction model may possibly avoid unnecessary invasive blood sampling for GnRH stimulation tests.
RESUMEN
Background: For the intensivists, accurate assessment of the ideal timing for successful weaning from the mechanical ventilation (MV) in the intensive care unit (ICU) is very challenging. Purpose: Using artificial intelligence (AI) approach to build two-stage predictive models, namely, the try-weaning stage and weaning MV stage to determine the optimal timing of weaning from MV for ICU intubated patients, and implement into practice for assisting clinical decision making. Methods: AI and machine learning (ML) technologies were used to establish the predictive models in the stages. Each stage comprised 11 prediction time points with 11 prediction models. Twenty-five features were used for the first-stage models while 20 features were used for the second-stage models. The optimal models for each time point were selected for further practical implementation in a digital dashboard style. Seven machine learning algorithms including Logistic Regression (LR), Random Forest (RF), Support Vector Machines (SVM), K Nearest Neighbor (KNN), lightGBM, XGBoost, and Multilayer Perception (MLP) were used. The electronic medical records of the intubated ICU patients of Chi Mei Medical Center (CMMC) from 2016 to 2019 were included for modeling. Models with the highest area under the receiver operating characteristic curve (AUC) were regarded as optimal models and used to develop the prediction system accordingly. Results: A total of 5,873 cases were included in machine learning modeling for Stage 1 with the AUCs of optimal models ranging from 0.843 to 0.953. Further, 4,172 cases were included for Stage 2 with the AUCs of optimal models ranging from 0.889 to 0.944. A prediction system (dashboard) with the optimal models of the two stages was developed and deployed in the ICU setting. Respiratory care members expressed high recognition of the AI dashboard assisting ventilator weaning decisions. Also, the impact analysis of with- and without-AI assistance revealed that our AI models could shorten the patients' intubation time by 21 hours, besides gaining the benefit of substantial consistency between these two decision-making strategies. Conclusion: We noticed that the two-stage AI prediction models could effectively and precisely predict the optimal timing to wean intubated patients in the ICU from ventilator use. This could reduce patient discomfort, improve medical quality, and lower medical costs. This AI-assisted prediction system is beneficial for clinicians to cope with a high demand for ventilators during the COVID-19 pandemic.
RESUMEN
Successful weaning from prolonged mechanical ventilation (MV) is an important issue in respiratory care centers (RCCs). Delayed or premature extubation increases both the risk of adverse outcomes and healthcare costs. However, the accurate evaluation of the timing of successful weaning from MV is very challenging in RCCs. This study aims to utilize artificial intelligence algorithms to build predictive models for the successful timing of the weaning of patients from MV in RCCs and to implement a dashboard with the best model in RCC settings. A total of 670 intubated patients in the RCC in Chi Mei Medical Center were included in the study. Twenty-six feature variables were selected to build the predictive models with artificial intelligence (AI)/machine-learning (ML) algorithms. An interactive dashboard with the best model was developed and deployed. A preliminary impact analysis was then conducted. Our results showed that all seven predictive models had a high area under the receiver operating characteristic curve (AUC), which ranged from 0.792 to 0.868. The preliminary impact analysis revealed that the mean number of ventilator days required for the successful weaning of the patients was reduced by 0.5 after AI intervention. The development of an AI prediction dashboard is a promising method to assist in the prediction of the optimal timing of weaning from MV in RCC settings. However, a systematic prospective study of AI intervention is still needed.
RESUMEN
Objective: To investigate the impact of a multidisciplinary intervention on the clinical outcomes of patients with COPD. Methods: This study retrospectively extracted the data of patients enrolled in the national pay-for-performance (P4P) program for COPD in four hospitals. Only COPD patients who received regular follow-up for at least one year in the P4P program between September 2018 and December 2020 were included. Results: A total of 1081 patients were included in this study. Among them, 424 (39.2%), 287 (26.5%), 179 (16.6%), and 191 (17.7%) patients were classified as COPD Groups A, B, C, and D, respectively. Dual therapy with long-acting ß2-agonist (LABA)/long-acting muscarinic antagonist (LAMA) was the most used inhaled bronchodilator at baseline (n = 477, 44.1%) patients, followed by LAMA monotherapy (n = 195, 18.0%), triple therapy with inhaled corticosteroid (ICS)/LABA/LAMA (n = 184, 17.0%), and ICS/LABA combination (n = 165, 15.3%). After one year of intervention, 374 (34.6%) and 323 (29.9%) patients had their pre- and post-bronchodilator-forced expiratory volume in one second (FEV1) increase of more than 100 mL. Both the COPD Assessment Test (CAT) and modified British Medical Research Council (mMRC) scores had a mean change of -2.2 ± 5.5 and -0.3 ± 0.9, respectively. The improvement in pulmonary function and symptom score were observed across four groups. The decreased number of exacerbations was only observed in Groups C and D, and not in Groups A and B. Conclusion: This real-world study demonstrated that the intervention in the P4P program could help improve the clinical outcome of COPD patients. It also showed us a different view on the use of dual therapy, which has a lower cost in Taiwan.
Asunto(s)
Broncodilatadores , Enfermedad Pulmonar Obstructiva Crónica , Corticoesteroides/efectos adversos , Agonistas de Receptores Adrenérgicos beta 2/efectos adversos , Broncodilatadores/efectos adversos , Humanos , Antagonistas Muscarínicos/efectos adversos , Enfermedad Pulmonar Obstructiva Crónica/diagnóstico , Enfermedad Pulmonar Obstructiva Crónica/tratamiento farmacológico , Reembolso de Incentivo , Estudios Retrospectivos , TaiwánRESUMEN
Hepatocellular carcinoma (HCC), which is associated with an absence of obvious symptoms and poor prognosis, is the second leading cause of cancer death worldwide. Genome-wide molecular biology studies should provide biological insights into HCC development. Based on the importance of phosphorylation for signal transduction, several protein kinase inhibitors have been developed that improve the survival of cancer patients. However, a comprehensive database of HCC-related phosphorylated biomarkers (HCCPMs) and novel HCCPMs prediction platform has been lacking. We have thus constructed the dBMHCC databases to provide expression profiles, phosphorylation and drug information, and evidence type; gathered information on HCC-related pathways and their involved genes as candidate HCC biomarkers; and established a system for evaluating protein phosphorylation and HCC-related biomarkers to improve the reliability of biomarker prediction. The resulting dBMHCC contains 611 notable HCC-related genes, 234 HCC-related pathways, 17 phosphorylation-related motifs and their 255 corresponding protein kinases, 5955 HCC biomarkers, and 1077 predicted HCCPMs. Methionine adenosyltransferase 2B (MAT2B) and acireductone dioxygenase 1 (ADI1), which regulate HCC development and hepatitis C virus infection, respectively, were among the top 10 HCCPMs predicted by dBMHCC. Platelet-derived growth factor receptor alpha (PDGFRA), which had the highest evaluation score, was identified as the target of one HCC drug (Regorafenib), five cancer drugs, and four non-cancer drugs. dBMHCC is an open resource for HCC phosphorylated biomarkers, which supports researchers investigating the development of HCC and designing novel diagnosis methods and drug treatments. Database URL: http://predictor.nchu.edu.tw/dBMHCC.
Asunto(s)
Biomarcadores de Tumor/metabolismo , Carcinoma Hepatocelular/metabolismo , Biología Computacional/métodos , Bases de Datos Factuales , Neoplasias Hepáticas/metabolismo , Animales , Carcinoma Hepatocelular/diagnóstico , Humanos , Internet , Neoplasias Hepáticas/diagnóstico , Ratones , Fosforilación , PronósticoRESUMEN
This study investigated the prognostic factors and outcomes of unplanned extubation (UE) in patients in a medical center's 6 intensive care units (ICUs) and calculated their mortality risk. We retrospectively reviewed the medical records of all adult patients in Chi Mei Medical Center who underwent UE between 2009 and 2015. During the study period, there were 305 episodes of UE in 295 ICU patients (men: 199 [67.5%]; mean age: 65.7 years; age range: 18-94 years). The mean Acute Physiology and Chronic Health Evaluation (APACHE) II score was 16.4, mean therapeutic intervention scoring system (TISS) score was 26.5, and mean Glasgow coma scale score was 10.4. One hundred thirty-six patients (46.1%) were re-intubated within 48 h. Forty-five died (mortality rate: 15.3%). Multivariate analyses showed 5 risk factors-respiratory rate, APACHE II score, uremia, liver cirrhosis, and weaning status-were independently associated with mortality. In conclusion, five risk factors including a high respiratory rate before UE, high APACHE II score, uremia, liver cirrhosis, and not in the process of being weaned-were associated with high mortality in patients who underwent UE.
Asunto(s)
Extubación Traqueal , Cuidados Críticos/estadística & datos numéricos , APACHE , Adulto , Anciano , Anciano de 80 o más Años , Extubación Traqueal/estadística & datos numéricos , Biomarcadores , Femenino , Escala de Coma de Glasgow , Mortalidad Hospitalaria , Humanos , Unidades de Cuidados Intensivos , Masculino , Persona de Mediana Edad , Pronóstico , Estudios Retrospectivos , Adulto JovenRESUMEN
We investigated whether N-terminal proB-type natriuretic peptide (NT-proBNP) predicts the prognosis of patients with acute respiratory distress syndrome (ARDS). Between December 1, 2012, and May 31, 2015, this observational study recruited patients admitted to our tertiary medical center who met the Berlin criteria for ARDS and who had their NT-proBNP measured. The main outcome was 28-day mortality. We enrolled 61 patients who met the Berlin criteria for ARDS: 7 were classified as mild, 29 as moderate, and 25 as severe. The median APACHE II scores were 23 (interquartile range [IQR], 18-28), and SOFA scores were 11 (IQR, 8-13). The median lung injury score was 3.0 (IQR, 2.50-3.25), and the median level of NT-proBNP was 2011 pg/ml (IQR, 579-7216). Thirty-four patients died during this study, and the 28-day mortality rate was 55.7%. Patients who die were older and had significantly (all p < 0.05) higher APACHE II scores and NT-proBNP levels than did patients who survived. Multivariate analysis identified age (HR: 1.546, 95% CI: 1.174-2.035, p = 0.0019) and NT-proBNP (HR: 1.009, 95% CI: 1.004-1.013, p = 0.0001) as significant risk factors of death. NT-proBNP was associated with poor outcomes for patients with ARDS, and its level predicted mortality.
Asunto(s)
Péptido Natriurético Encefálico/metabolismo , Fragmentos de Péptidos/metabolismo , Síndrome de Dificultad Respiratoria/diagnóstico , Anciano , Demografía , Femenino , Humanos , Masculino , Persona de Mediana Edad , Pronóstico , Curva ROC , Síndrome de Dificultad Respiratoria/mortalidad , Análisis de Supervivencia , Resultado del TratamientoRESUMEN
The aim of this study was to establish predictors for successfully planned extubation, which can be followed by medical personnel. The patients who were admitted to the adult intensive care unit of a tertiary hospital and met the following criteria between January 2005 and December 2014 were collected retrospectively: intubation > 48âhours; and candidate for extubation. The patient characteristics, including disease severity, rapid shallow breath index (RSBI), maximal inspiratory pressure (MIP), maximal expiratory pressure (MEP), cuff leak test (CLT) before extubation, and outcome, were recorded. The CLT was classified as 2+ with audible flow without a stethoscope, 1+ with audible flow using a stethoscope, and negative (N) with no audible flow, even with a stethoscope. Failure to extubate was defined as reintubation within 48âhours. In total, 6583 patients were enrolled and 403 patients (6.1%) had extubation failures. Male patients dominated the patient cohort (4261 [64.7%]). The mean age was 64.5±16.3 years. The overall in-hospital mortality rate was 11.3%. The extubation failure rate for females was greater than males (7.7% vs 5.3%, Pâ<â0.001). The group of patients who failed extubation were older (66.7â±â14.4 vs 64.3â±â16.4, P = 0.002), had higher APACHE II scores (16.8â±â7.6 vs 15.9â±â7.8, P = 0.023), lower coma scales (10.3â±â3.7 vs 10.8â±â3.7, P = 0.07), a higher RSBI (69.9â±â37.3 vs 58.6â±â30.3, Pâ<â0.001), a lower MIP, and MEP (-35.6â±â15.3 vs -37.8â±â14.6, P = 0.0001 and 49.6â±â28.4 vs 58.6â±â30.2, Pâ<â0.001, respectively), and a higher mortality rate (25.6% vs 10.5%, Pâ<â0.001) compared to the successful extubation group. Based on multivariate logistic regression, a CLT of 2+ (odds ratio [OR] = 2.07, Pâ<â0.001), a MEP ≥ 55âcmH2O (OR = 1.73, Pâ<â0.001), and a RSBIâ<â68âbreath/min/ml (OR = 1.57, Pâ<â0.001) were independent predictors for successful extubation.This study identified 3 independent risk factors for successful extubation after a successful breathing trial, including a CLT of 2+, a MEP ≥ 55âcmH2O, and a RSBI < 68âbreath/min/ml. Furthermore, a nomogram integrating these 3 parameters, which represented the combined consideration of the upper airway patentency, cough strength, and respiratory capacity, was developed to better predict extubation success.
Asunto(s)
Extubación Traqueal/normas , Unidades de Cuidados Intensivos , Respiración Artificial/métodos , Insuficiencia Respiratoria/terapia , Adulto , Femenino , Mortalidad Hospitalaria/tendencias , Humanos , Masculino , Persona de Mediana Edad , Insuficiencia Respiratoria/mortalidad , Estudios Retrospectivos , Factores de Riesgo , Taiwán/epidemiologíaRESUMEN
The initial hypoxemic level of acute respiratory distress syndrome (ARDS) defined according to Berlin definition might not be the optimal predictor for prognosis. We aimed to determine the predictive validity of the stabilized ratio of partial pressure arterial oxygen and fraction of inspired oxygen (PaO2/FiO2 ratio) following standard ventilator setting in the prognosis of patients with ARDS.This prospective observational study was conducted in a single tertiary medical center in Taiwan and compared the stabilized PaO2/FiO2 ratio (Day 1) following standard ventilator settings and the PaO2/FiO2 ratio on the day patients met ARDS Berlin criteria (Day 0). Patients admitted to intensive care units and in accordance with the Berlin criteria for ARDS were collected between December 1, 2012 and May 31, 2015. Main outcome was 28-day mortality. Arterial blood gas and ventilator setting on Days 0 and 1 were obtained.A total of 238 patients met the Berlin criteria for ARDS were enrolled, and they were classified as mild (nâ=â50), moderate (nâ=â125), and severe (nâ=â63) ARDS, respectively. Twelve (5%) patients who originally were classified as ARDS did not continually meet the Berlin definition, and a total of 134 (56%) patients had the changes regarding the severity of ARDS from Day 0 to Day 1. The 28-day mortality rate was 49.1%, and multivariate analysis identified age, PaO2/FiO2 on Day 1, number of organ failures, and positive fluid balance within 5 days as significant risk factors of death. Moreover, the area under receiver-operating curve for mortality prediction using PaO2/FiO2 on Day 1 was significant higher than that on Day 0 (Pâ=â0.016).PaO2/FiO2 ratio on Day 1 after applying mechanical ventilator is a better predictor of outcomes in patients with ARDS than those on Day 0.