METHODS: C0 were retrieved from a large neonatal vancomycin dataset. Individual estimates of AUC0-24 were obtained from Bayesian post hoc estimation. Various ML algorithms were used for model building to C0 and AUC0-24. An external dataset was used for predictive performance evaluation.
RESULTS: Before starting treatment, C0 can be predicted a priori using the Catboost-based C0-ML model combined with dosing regimen and nine covariates. External validation results showed a 42.5% improvement in prediction accuracy by using the ML model compared with the population pharmacokinetic model. The virtual trial showed that using the ML optimized dose; 80.3% of the virtual neonates achieved the pharmacodynamic target (C0 in the range of 10-20 mg/L), much higher than the international standard dose (37.7-61.5%). Once therapeutic drug monitoring (TDM) measurements (C0) in patients have been obtained, AUC0-24 can be further predicted using the Catboost-based AUC-ML model combined with C0 and nine covariates. External validation results showed that the AUC-ML model can achieve an prediction accuracy of 80.3%.
CONCLUSION: C0-based and AUC0-24-based ML models were developed accurately and precisely. These can be used for individual dose recommendations of vancomycin in neonates before treatment and dose revision after the first TDM result is obtained, respectively.
METHODS: EEG single trials are decomposed with discrete wavelet transform (DWT) up to the [Formula: see text] level of decomposition using a biorthogonal B-spline wavelet. The coefficients of DWT in each trial are thresholded to discard sparse wavelet coefficients, while the quality of the signal is well maintained. The remaining optimum coefficients in each trial are encoded into bitstreams using Huffman coding, and the codewords are represented as a feature of the ERP signal. The performance of this method is tested with real visual ERPs of sixty-eight subjects.
RESULTS: The proposed method significantly discards the spontaneous EEG activity, extracts the single-trial visual ERPs, represents the ERP waveform into a compact bitstream as a feature, and achieves promising results in classifying the visual objects with classification performance metrics: accuracies 93.60[Formula: see text], sensitivities 93.55[Formula: see text], specificities 94.85[Formula: see text], precisions 92.50[Formula: see text], and area under the curve (AUC) 0.93[Formula: see text] using SVM and k-NN machine learning classifiers.
CONCLUSION: The proposed method suggests that the joint use of discrete wavelet transform (DWT) with Huffman coding has the potential to efficiently extract ERPs from background EEG for studying evoked responses in single-trial ERPs and classifying visual stimuli. The proposed approach has O(N) time complexity and could be implemented in real-time systems, such as the brain-computer interface (BCI), where fast detection of mental events is desired to smoothly operate a machine with minds.
METHODS: Drug formulations were administered to the experimental animals via oral, intravenous and intraperitoneal routes. Blood samples were collected at different pre-determined time intervals to determine the pharmacokinetic parameters. To understand the biodistribution profile of HCZ, tissue samples were isolated from different groups of Sprague-Dawley rats at different time points. The pharmacokinetic parameters of HZC were evaluated after administration through oral (100 mg/kg), intraperitoneal (100 mg/kg) and intravenous (10 mg/kg) routes.
RESULTS: Significantly (p
MATERIALS AND METHODS: In this trial, a total of 56 eligible subjects were randomly assigned to the fasting group and the postprandial group. The two groups were given 250 mg of the test and reference preparation, respectively. Liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) was applied to determine the plasma concentration of cefalexin. PhoenixWinNonlin software (V7.0) was used to calculate the pharmacokinetic parameters of cefalexin using the non-compartmental model (NCA), and the bioequivalence and safety results were calculated by SAS (V9.4) software.
RESULTS: The main pharmacokinetic parameters of the test and reference preparations were as follows, the fasting group: Cmax 12.59 ± 2.65 μg/mL, 12.72 ± 2.28 μg/mL; AUC0-8h 20.43 ± 3.47 h×μg/mL, 20.66 ± 3.38 h×μg/mL; AUC0-∞ 20.77 ± 3.53 h×μg/mL, 21.02 ± 3.45 h×μg/mL; the postprandial group: Cmax 5.25 ± 0.94 μg/mL, 5.23 ± 0.80 μg/mL; AUC0-10h 16.92 ± 2.03 h×μg/mL, 17.09 ± 2.31 h×μg/mL; AUC0-∞ 17.33 ± 2.09 h×μg/mL, 17.67 ± 2.45 h×μg/mL.
CONCLUSION: The 90% confidence intervals of geometric mean ratios of test preparation and reference preparation were calculated, and the 90% confidence intervals of geometric mean ratios of Cmax, AUC0-10h, and AUC0-∞ were within the 80.00% ~ 125.00% range in both groups. Both Cmax and AUC met the pre-determined criteria for assuming bioequivalence. The test and reference products were bioequivalent after administration under fasting as well as under fed conditions in healthy Chinese subjects. This study may suggest that successful generic versions of cefalexin not only guarantee the market supply of such drugs but can also improve the safety and effectiveness and quality controllability of cefalexin through a new process and a new drug composition ratio.
METHODS: Subjects were recruited among those responding to a social media announcement or patients attending the SEGi Oral Health Care Centre between May and December 2019, and among some staff at the centre. Five ml of unstimulated whole saliva was collected and salivary LDH enzyme activity levels were measured with a LDH colorimetric assay kit. Salivary LDH activity level was determined for each group and compared statistically.
RESULTS: Eighty-eight subjects were categorized into three groups (control n=30, smokers n=29, and vapers n=29). The mean ± standard deviation (SD) values for salivary LDH activity levels for vapers, smokers, and control groups were 35.15 ± 24.34 mU/ml, 30.82 ± 20.73 mU/ml, and 21.45 ± 15.30 mU/ml, respectively. The salivary LDH activity levels of smoker and vaper groups were significantly higher than in the control group (p = 0.031; 0.017). There was no significant difference of salivary LDH activity level in vapers when compared with smokers (p= 0.234).
CONCLUSION: Our findings showed higher LDH levels in the saliva of vapers when compared with controls, confirming cytotoxic and harmful effects of e-cigarettes on the oral mucosa.
METHODS: We did an individual patient data meta-analysis, in which we searched PubMed and Web of Science for studies published from database inception until April 30, 2019. Studies reporting original biopsy-controlled data of CAP for non-invasive grading of steatosis were eligible. Probe recommendation was based on automated selection, manual assessment of skin-to-liver-capsule distance, and a body-mass index (BMI) criterion. Receiver operating characteristic methods and mixed models were used to assess diagnostic properties and covariates. Patients with non-alcoholic fatty liver disease (NAFLD) were analysed separately because they are the predominant patient group when using the XL probe. This study is registered with PROSPERO, CRD42018099284.
FINDINGS: 16 studies reported histology-controlled CAP including the XL probe, and individual data from 13 papers and 2346 patients were included. Patients with a mean age of 46·5 years (SD 14·5) were recruited from 20 centres in nine countries. 2283 patients had data for BMI; 673 (29%) were normal weight (BMI <25 kg/m2), 530 (23%) were overweight (BMI ≥25 to <30 kg/m2), and 1080 (47%) were obese (BMI ≥30 kg/m2). 1277 (54%) patients had NAFLD, 474 (20%) had viral hepatitis, 285 (12%) had alcohol-associated liver disease, and 310 (13%) had other liver disease aetiologies. The XL probe was recommended in 1050 patients, 930 (89%) of whom had NAFLD; among the patients with NAFLD, the areas under the curve were 0·819 (95% CI 0·769-0·869) for S0 versus S1 to S3 and 0·754 (0·720-0·787) for S0 to S1 versus S2 to S3. CAP values were independently affected by aetiology, diabetes, BMI, aspartate aminotransferase, and sex. Optimal cutoffs differed substantially across aetiologies. Risk of bias according to QUADAS-2 was low.
INTERPRETATION: CAP cutoffs varied according to cause, and can effectively recognise significant steatosis in patients with viral hepatitis. CAP cannot grade steatosis in patients with NAFLD adequately, but its value in a NAFLD screening setting needs to be studied, ideally with methods beyond the traditional histological reference standard.
FUNDING: The German Federal Ministry of Education and Research and Echosens.
Methods: We evaluated commonly used surrogate and imputed baseline creatinine values against a "reference" creatinine measured during follow-up in an adult clinical trial cohort. Known AKI incidence (Kidney Disease: Improving Global Outcomes [KDIGO] criteria) was compared with AKI incidence classified by (1) back-calculation using the Modification of Diet in Renal Disease (MDRD) equation with and without a Chinese ethnicity correction coefficient; (2) back-calculation using the Chronic Kidney Disease-Epidemiology Collaboration (CKD-EPI) equation; (3) assigning glomerular filtration rate (GFR) from age and sex-standardized reference tables; and (4) lowest measured creatinine during admission. Back-calculated distributions were performed using GFRs of 75 and 100 ml/min.
Results: All equations using an assumed GFR of 75 ml/min underestimated AKI incidence by more than 50%. Back-calculation with CKD-EPI and GFR of 100 ml/min most accurately predicted AKI but misclassified all AKI stages and had low levels of agreement with true AKI diagnoses. Back-calculation using MDRD and assumed GFR of 100 ml/min, age and sex-reference GFR values adjusted for good health, and lowest creatinine during admission performed similarly, best predicting AKI incidence (area under the receiver operating characteristic curves [AUC ROCs] of 0.85, 0.87, and 0.85, respectively). MDRD back-calculation using a cohort mean GFR showed low total error (22%) and an AUC ROC of 0.85.
Conclusion: Current methods for estimating baseline creatinine are large sources of potential error in acute infection studies. Preferred alternatives include MDRD equation back-calculation with a population mean GFR, age- and sex-specific GFR values corrected for "good health," or lowest measured creatinine. Studies using surrogate baseline creatinine values should report specific methodology.
Methods: A cross-sectional study on 50 patients of age 50 and above with contrast-enhanced CT (CECT) and dual-energy X-ray absorptiometry (DXA) was conducted from November 2018 to November 2019. Single region of interest (ROI) was placed at the anterior trabecular part of L1 vertebra on CECT to obtain HU value. Correlation of CT HU value of L1 vertebra and DXA T-score, interrater reliability agreement between HU value of L1 vertebra and T-score in determining groups of with and without osteoporosis, ROC curve analysis for diagnostic accuracy and cut-off value of CT for detection of osteoporosis were identified.
Results: Significant correlation between HU value of L1 vertebra and L1 T-score (r = 0.683)/lowest skeletal T-score (r = 0.703) (P < 0.001). Substantial agreement between HU value of L1 vertebra and DXA in determining the groups with and without osteoporosis (k = 0.8; P < 0.001). The area under the receiver operating characteristic (AUROC) curve was 0.93 (95% CI: 0.86, 1.00) using HU value (P < 0.001). Cut-off value for osteoporosis was 149 HU.
Conclusion: HU value of lumbar vertebra is an effective alternative for the detection of osteoporosis with high diagnostic accuracy in hospitals without DXA facility.
MATERIALS AND METHODS: Contrast enhanced computed tomography (CT) images of 194 multi-racial NSCLC patients (79 EGFR mutants and 115 wildtypes) were collected from three different countries using 5 manufacturers' scanners with a variety of scanning parameters. Ninety-nine cases obtained from the University of Malaya Medical Centre (UMMC) in Malaysia were used for training and validation procedures. Forty-one cases collected from the Kyushu University Hospital (KUH) in Japan and fifty-four cases obtained from The Cancer Imaging Archive (TCIA) in America were used for a test procedure. Radiomic features were obtained from BN maps, which represent topologically invariant heterogeneous characteristics of lung cancer on CT images, by applying histogram- and texture-based feature computations. A BN-based signature was determined using support vector machine (SVM) models with the best combination of features that maximized a robustness index (RI) which defined a higher total area under receiver operating characteristics curves (AUCs) and lower difference of AUCs between the training and the validation. The SVM model was built using the signature and optimized in a five-fold cross validation. The BN-based model was compared to conventional original image (OI)- and wavelet-decomposition (WD)-based models with respect to the RI between the validation and the test.
RESULTS: The BN-based model showed a higher RI of 1.51 compared with the models based on the OI (RI: 1.33) and the WD (RI: 1.29).
CONCLUSION: The proposed model showed higher robustness than the conventional models in the identification of EGFR mutations among NSCLC patients. The results suggested the robustness of the BN-based approach against variations in image scanner/scanning parameters.
OBJECTIVE: Apply machine learning for the prediction and identification of factors associated with short and long-term mortality in Asian STEMI patients and compare with a conventional risk score.
METHODS: The National Cardiovascular Disease Database for Malaysia registry, of a multi-ethnic, heterogeneous Asian population was used for in-hospital (6299 patients), 30-days (3130 patients), and 1-year (2939 patients) model development. 50 variables were considered. Mortality prediction was analysed using feature selection methods with machine learning algorithms and compared to Thrombolysis in Myocardial Infarction (TIMI) score. Invasive management of varying degrees was selected as important variables that improved mortality prediction.
RESULTS: Model performance using a complete and reduced variable produced an area under the receiver operating characteristic curve (AUC) from 0.73 to 0.90. The best machine learning model for in-hospital, 30 days, and 1-year outperformed TIMI risk score (AUC = 0.88, 95% CI: 0.846-0.910; vs AUC = 0.81, 95% CI:0.772-0.845, AUC = 0.90, 95% CI: 0.870-0.935; vs AUC = 0.80, 95% CI: 0.746-0.838, AUC = 0.84, 95% CI: 0.798-0.872; vs AUC = 0.76, 95% CI: 0.715-0.802, p < 0.0001 for all). TIMI score underestimates patients' risk of mortality. 90% of non-survival patients are classified as high risk (>50%) by machine learning algorithm compared to 10-30% non-survival patients by TIMI. Common predictors identified for short- and long-term mortality were age, heart rate, Killip class, fasting blood glucose, prior primary PCI or pharmaco-invasive therapy and diuretics. The final algorithm was converted into an online tool with a database for continuous data archiving for algorithm validation.
CONCLUSIONS: In a multi-ethnic population, patients with STEMI were better classified using the machine learning method compared to TIMI scoring. Machine learning allows for the identification of distinct factors in individual Asian populations for better mortality prediction. Ongoing continuous testing and validation will allow for better risk stratification and potentially alter management and outcomes in the future.
Methods: We enrolled and reviewed 122 biopsy-proven NAFLD patients. Advanced fibrosis was defined as fibrosis stages 3-4. Noninvasive assessments included aspartate aminotransferase/alanine aminotransferase (AST/ALT) ratio, AST-to-platelet ratio index (APRI), AST/ALT ratio, diabetes (BARD) score, fibrosis-4 (FIB-4) score, and NAFLD fibrosis score.
Results: FIB-4 score had the highest area under the receiver operating characteristic curve (AUROC) and negative predictive value (NPV) of 0.86 and 94.3%, respectively, for the diagnosis of advanced fibrosis. FIB-4 score
Materials and Methods: Sixty-three diabetic foot patients admitted from June 15, 2019 to February 15, 2020. Methods included one-on-one interview for clinico-demographic data, physical examination to determine the classification. Patients were followed-up and outcomes were determined. Pearson Chi-square or Fisher's Exact determined association between clinico-demographic data, the classifications, and outcomes. The receiver operating characteristic (ROC) curve determined predictive abilities of classification systems and paired analysis compared the curves. Area Under the Receiver Operating Characteristic Curve (AUC) values used to compare the prediction accuracy. Analysis was set at 95% CI.
Results: Results showed hypertension, duration of diabetes, and ambulation status were significantly associated with major amputation. WIFi showed the highest AUC of 0.899 (p = 0.000). However, paired analysis showed AUC differences between WIFi, Wagner, and University of Texas classifications by grade were not significantly different from each other.
Conclusion: The WIFi, Wagner, and University of Texas classification systems are good predictors of major amputation with WIFi as the most predictive.
METHODS: We used relevant key words to search electronic databases to identify studies published up to 2019 that used receiver operating characteristic (ROC) curves for assessing the cut-off points of anthropometric indices. We used a random-effects model to pool study results and assessed between-study heterogeneity by using the I2 statistic and Cochran's Q test.
RESULTS: This meta-analysis included 38 cross-sectional and 2 cohort studies with 105 to 137,256 participants aged 18 or older. The pooled area under the ROC curve (AUC) value for BMI was 0.66 (95% CI, 0.63-0.69) in both men and women. The pooled AUC values for WC were 0.69 (95% CI, 0.67-0.70) in men and 0.69 (95% CI, 0.64-0.74) in women, and the pooled AUC values for WHR were 0.69 (95% CI, 0.66-0.73) in men and 0.71 (95% CI, 0.68-0.73) in women.
CONCLUSION: Our findings indicated a slight difference between AUC values of these anthropometric indices. However, indices of abdominal obesity, especially WHR, can better predict CVD occurrence.