METHODS: MiRNA profiling was conducted on plasma samples from 18 patients with primary aldosteronism taken during adrenal venous sampling on an Illumina MiSeq platform. Bioinformatics and machine learning identified 9 miRNAs for validation by reverse transcription real-time quantitative polymerase chain reaction. Validation was performed on a cohort consisting of 108 patients with known subdifferentiation. A 30-patient subset of the validation cohort involved both adrenal venous sampling and peripheral, the rest only peripheral samples. A neural network model was used for feature selection and comparison between adrenal venous sampling and peripheral samples, while a deep-learning model was used for classification.
RESULTS: Our model identified 10 miRNA combinations achieving >85% accuracy in distinguishing unilateral primary aldosteronism and bilateral adrenal hyperplasia on a 30-sample subset, while also confirming the suitability of peripheral samples for analysis. The best model, involving 6 miRNAs, achieved an area under curve of 87.1%. Deep learning resulted in 100% accuracy on the subset and 90.9% sensitivity and 81.8% specificity on all 108 samples, with an area under curve of 86.7%.
CONCLUSIONS: Machine learning analysis of circulating miRNAs offers a minimally invasive alternative for primary aldosteronism lateralization. Early identification of bilateral adrenal hyperplasia could expedite treatment initiation without the need for further localization, benefiting both patients and health care providers.
METHODS: A cohort of 4,240 Sepsis-3 patients was analyzed, with 783 experiencing 30-day mortality and 3,457 surviving. Fifteen biomarkers were selected using feature ranking methods, including Extreme Gradient Boosting (XGBoost), Random Forest, and Extra Tree, and the Logistic Regression (LR) model was used to assess their individual predictability with a fivefold cross-validation approach for the validation of the prediction. The dataset was balanced using the SMOTE-TOMEK LINK technique, and a stacking-based meta-classifier was used for 30-day mortality prediction. The SHapley Additive explanations analysis was performed to explain the model's prediction.
RESULTS: Using the LR classifier, the model achieved an area under the curve or AUC score of 0.99. A nomogram provided clinical insights into the biomarkers' significance. The stacked meta-learner, LR classifier exhibited the best performance with 95.52% accuracy, 95.79% precision, 95.52% recall, 93.65% specificity, and a 95.60% F1-score.
CONCLUSIONS: In conjunction with the nomogram, the proposed stacking classifier model effectively predicted 30-day mortality in Sepsis patients. This approach holds promise for early intervention and improved outcomes in treating Sepsis cases.