RESUMO
OBJECTIVE: To develop machine learning models for predicting intensive care unit (ICU) readmission using ensemble learning algorithms. METHODS: A publicly accessible American ICU database, medical information mart for intensive care (MIMIC)-â ¢ as the data source was used, and the patients were selected by the inclusion and exclusion criteria. A set of variables that had the predictive ability of outcome including demographics, vital signs, laboratory tests, and comorbidities of patients were extracted from the dataset. We built the ICU readmission prediction models based on ensemble learning methods including random forest, adaptive boosting (AdaBoost), and gradient boosting decision tree (GBDT), and compared the prediction performance of the machine learning models with a conventional Logistic regression model. Five-fold cross validation was used to train and validate the prediction models. Average sensitivity, positive prediction value, negative prediction value, false positive rate, false negative rate, area under the receiver operating characteristic curve (AUROC) and Brier score were used as performance measures. After constructing the prediction models, top 10 predictive variables based on importance ranking were identified by the model with the best discrimination. RESULTS: Among these ICU readmission prediction models, GBDT (AUROC=0.858) had better performance than random forest (AUROC=0.827), and was slightly superior to AdaBoost (AUROC=0.851) in terms of AUROC. Compared with Logistic regression (AUROC=0.810), the discrimination of the three ensemble learning models was much better. The feature importance provided by GBDT showed that the top ranking variables included vital signs and laboratory tests. The patients with ICU readmission had higher mean arterial pressure, systolic blood pressure, diastolic blood pressure, and heart rate than the patients without ICU readmission. Meanwhile, the patients readmitted to ICU experienced lower urine output and higher serum creatinine. Overall, the patients having repeated admissions during their hospitalization showed worse heart function and renal function compared with the patients without ICU readmission. CONCLUSION: The ensemble learning based ICU readmission prediction models had better performance than Logistic regression model. Such ensemble learning models have the potential to aid ICU physicians in identifying those patients with high risk of ICU readmission and thus help improve overall clinical outcomes.
Assuntos
Estado Terminal , Readmissão do Paciente , Humanos , Unidades de Terapia Intensiva , Aprendizado de Máquina , Curva ROCRESUMO
OBJECTIVE: To construct length of intensive care unit (ICU) stay (LOS-ICU) prediction models for ICU patients, based on three machine learning models support vector machine (SVM), classification and regression tree (CART), and random forest (RF), and to compare the prediction perfor-mance of the three machine learning models with the customized simplified acute physiology score â ¡(SAPS-â ¡) model. METHODS: We used medical information mart for intensive care (MIMIC)-â ¢ database for model development and validation. The primary outcome was prolonged LOS-ICU(pLOS-ICU), defined as longer than the third quartile of patients' LOS-ICU in the studied dataset. The recursive feature elimination method was used to do feature selection for three machine learning models. We utilized 5-fold cross validation to evaluate model prediction performance. The Brier value, area under the receiver operation characteristic curve (AUROC), and estimated calibration index (ECI) were used as perfor-mance measures. Performances of the four models were compared, and performance differences between the models were assessed using two-sided t test. The model with the best prediction performance was employed to generate variable importance ranking, and the identified top five important predictors were pre-sented. RESULTS: The final cohort in our study consisted of 40 200 eligible ICU patients, of whom 23.7% were with pLOS-ICU. The proportion of the male patients was 57.6%, and the age of all the ICU patients was (61.9±16.5) years.Results showed that the three machine learning models outperformed the customized SAPS-â ¡ model in terms of all the performance measures with statistical significance (P < 0.01). Among the three machine learning models, the RF model achieved the best overall performance (Brier value, 0.145), discrimination (AUROC, 0.770) and calibration (ECI, 7.259). The calibration curve showed that the RF model slightly overestimated the risk of pLOS-ICU in high-risk ICU patients, but underestimated the risk of pLOS-ICU in low-risk ICU patients. Top five important predictors for pLOS-ICU identified by the RF model included age, heart rate, systolic blood pressure, body tempe-rature, and ratio of arterial oxygen tension to the fraction of inspired oxygen(PaO2/FiO2). CONCLUSION: The RF algorithm-based pLOS-ICU prediction model had a best prediction performance in this study. It lays a foundation for future application of the RF-based pLOS-ICU prediction model in ICU clinical practice.
Assuntos
Aprendizado de Máquina , Projetos de Pesquisa , Idoso , Humanos , Unidades de Terapia Intensiva , Masculino , Pessoa de Meia-IdadeRESUMO
OBJECTIVE: To construct an in-hospital mortality prediction model for patients with acute kidney injury (AKI) in intensive care unit (ICU) by using support vector machine (SVM), and compare it with the simplified acute physiology score II (SAPS-II) which is commonly used in the ICU. METHODS: We used Medical Information Mart for Intensive Care III (MIMIC-III) database as data source. The AKI patients in the MIMIC-III database were selected according to the 2012 Kidney Disease: Improving Global Outcomes (KDIGO) definition of AKI. We employed the same predictor variable set as used in SAPS-II to construct an SVM model. Meanwhile, we also developed a customized SAPS-II model using MIMIC-III database, and compared performances between the SVM model and the customized SAPS-II model. The performance of each model was evaluated via area under the receiver operation characteristic curve (AUROC), root mean squared error (RMSE), sensitivity, specificity, Youden's index and accuracy based on 5-fold cross-validation. The agreement of the results between the SVM model and the customized SAPS-II model was illustrated using Bland-Altman plots. RESULTS: A total number of 19 044 patients with AKI were included. The observed in-hospital mortality of the AKI patients was 13.58% in MIMIC-III. The results based on the 5-fold cross validation showed that the average AUROC of the SVM model and the customized SAPS-II model was 0.86 and 0.81, respectively (The difference between the two models was statistically significant with t=13.0, P<0.001). The average RMSE of the SVM model and the customized SAPS-II model was 0.29 and 0.31, respectively (The difference was statistically significant with t=-9.6, P<0.001). The SVM model also outperformed the customized SAPS-II model in terms of sensitivity and Youden's index with significant statistical differences (P=0.002 and <0.001, respectively).The Bland-Altman plot showed that the SVM model and the customized SAPS-II model had similar mortality prediction results when the mortality of a patient was certain, but the consistency between the mortality prediction results of the two models was poor when the mortality of a patient was with high uncertainty. CONCLUSION: Compared with the SAPS-II model, the SVM model has a better performance, especially when the mortality of a patient is with high uncertainty. The SVM model is more suitable for predicting the mortality of patients with AKI in ICU and early intervention in patients with AKI in ICU. The SVM model can effectively help ICU clinicians improve the quality of medical treatment, which has high clinical value.
Assuntos
Injúria Renal Aguda/mortalidade , Mortalidade Hospitalar , Máquina de Vetores de Suporte , Cuidados Críticos , Humanos , Unidades de Terapia Intensiva , Prognóstico , Curva ROC , Sensibilidade e EspecificidadeRESUMO
We examined the correlation between PNPLA7 gene polymorphisms at the rs61754920 and rs11137410 loci and menstrual disorder in women of reproductive age in the Central Plain. Genomic DNA was extracted from peripheral blood; polymerase chain reaction-ligase detection reaction and SNaPshot genotyping were used to detect polymorphisms in the rs61754920 and rs11137410 gene loci, respectively. The results for the 2 loci in individuals of different blood types were statistically analyzed. The proportion of the AA homozygote at the rs61754920 locus in the PNPLA7 gene was the lowest, while the proportion of the CC homozygote at the rs11137410 locus in the PNPLA7 gene was the highest. There were no statistical differences in the frequency distribution of genotypes and alleles at the 2 loci between control and test groups. The frequency of the TT genotype at the rs11137410 locus in women with type O blood was significantly lower in the test group than in the control group. Frequencies of the C and T alleles were significantly different between the 2 groups. There may be an association between the PNPLA7 gene and type O blood or a combined effect of the 2 genes.
Assuntos
Predisposição Genética para Doença , Lipase/genética , Distúrbios Menstruais/genética , Polimorfismo de Nucleotídeo Único , Sistema ABO de Grupos Sanguíneos/genética , Adulto , Alelos , Estudos de Casos e Controles , China , Cromossomos Humanos Par 9/genética , Feminino , Frequência do Gene , Loci Gênicos , Genótipo , Humanos , Lisofosfolipase , Ciclo Menstrual/genética , Reação em Cadeia da PolimeraseRESUMO
This study aimed to establish reference intervals for serum thyroid hormones [serum thyroid-stimulating hormone (TSH), triiodothyronine (TT3), thyroxine (TT4), free triiodothyronine (FT3), and free thyroxine (FT4)] in apparently healthy individuals living in Zhengzhou. According to the requirement for laboratory support for the diagnosis and monitoring of thyroid diseases in the National Academy of Clinical Biochemistry (NACB) laboratory medicine practice guidelines, a total of 211 apparently healthy individuals were enrolled (94 men, 117 women, 23-77 years old) from Zhengzhou for measurement of serum levels of TSH, TT3, TT4, FT3, and FT4 by using the Siemens ADVIA Centaur XP analyzer. All markers were analyzed across gender- and age-specific groups by using the t-test and ANOVA. The reference intervals of all markers were determined by P2.5-P97.5. We detected gender-associated statistical significances for TT3, TT4, FT3, and FT4 (t=3.299, 2.141, 5.868, 5.358; P<0.05), but not for TSH (t=-1.776, P>0.05). Correlation analysis showed that all markers were negatively correlated with age (P>0.05). The new reference intervals for TT3, TT4, FT3, FT4, and TSH were established: 0.76-1.38 ng/mL, 5.96-11.27 µg/dL, 3.88-5.59 pM, 11.69-18.84 pM, 0.89-5.93 µIU/mL, respectively. In conclusion, we added a new database of reference intervals of the serum thyroid hormones for the Chinese adult population.
Assuntos
Hormônios Tireóideos/sangue , Adulto , Fatores Etários , Idoso , China , Feminino , Voluntários Saudáveis , Humanos , Masculino , Pessoa de Meia-Idade , Valores de Referência , Reprodutibilidade dos Testes , Fatores Sexuais , Adulto JovemRESUMO
Objective: To assess the prevalence, risk factors and treatment of anemia in patients with chronic kidney disease (CKD). Methods: A descriptive method was used to analyze the prevalence and treatment of anemia in CKD patients based on regional health data in Yinzhou District of Ningbo during 2012-2018. The multivariate logistic regression analysis was used to identify independent influence factors of anemia in the CKD patients. Results: In 52 619 CKD patients, 15 639 suffered from by anemia (29.72%), in whom 5 461 were men (26.41%) and 10 178 were women (31.87%), and anemia prevalence was higher in women than in men, the difference was significant (P<0.001). The prevalence of anemia increased with stage of CKD (24.77% in stage 1 vs. 69.42% in stage 5, trend χ2 test P<0.001). Multivariate logistic regression analysis revealed that being women (aOR=1.57, 95%CI: 1.50-1.63), CKD stage (stage 2: aOR=1.10, 95%CI: 1.04-1.16;stage 3: aOR=2.28,95%CI: 2.12-2.44;stage 4: aOR=4.49,95%CI :3.79-5.32;stage 5: aOR=6.31,95%CI: 4.74-8.39), age (18-30 years old: aOR=2.40,95%CI: 2.24-2.57, 61-75 years old: aOR=1.35,95%CI:1.28-1.42, ≥76 years old: aOR=2.37,95%CI:2.20-2.55), BMI (<18.5 kg/m2:aOR=1.29,95%CI: 1.18-1.41;23.0-24.9 kg/m2:aOR=0.79,95%CI: 0.75-0.83;≥25.0 kg/m2:aOR=0.70,95%CI: 0.66-0.74), abdominal obesity (aOR=0.91, 95%CI: 0.86-0.96), chronic obstructive pulmonary disease (aOR=1.15, 95%CI: 1.09-1.22), cancer (aOR=3.03, 95%CI: 2.84-3.23), heart failure (aOR=1.44, 95%CI: 1.35-1.54) and myocardial infarction (aOR=1.54, 95%CI:1.16-2.04) were independent risk factors of anemia in CKD patients. Among stage 3-5 CKD patients with anemia, 12.03% received iron therapy, and 4.78% received treatment with erythropoiesis-stimulating agent (ESA) within 12 months after anemia was diagnosed. Conclusions: The prevalence of anemia in CKD patients was high in Yinzhou. However, the treatment rate of iron therapy and ESA were low. More attention should be paid to the anemia management and treatment in CKD patients.