The accelerometers are commonly applied to measure the vibrations in the fields of motion control and precision measurement, whose sensitivities are essentially important to their applications. The vibration calibration is utilized to determine their sensitivities before they are used or after a period of time. At present, the Nyquist sampling (NS), bandpass sampling (BPS), and mixer and low-pass filter sampling (MLPFS) based heterodyne laser interferometry are widely utilized to accomplish the vibration calibration. Compared with the NS method, the latter two methods can significantly reduce the sampling rate and extend the calibration frequency range. However, the BPS method has to adopt the complex algorithm and prior information so as to get its sampling rate, and the MLPFS method is inevitably influenced by an extra phase delay. In this article, a novel heterodyne laser interferometry is investigated to simultaneously determine the sensitivity magnitude and phase of the accelerometers with high accuracy in a wide frequency range. This method significantly eliminates the phase delay by introducing an appropriate symmetric differential demodulation strategy, which can improve the sensitivity phase calibration accuracy, especially at higher frequencies. The comparison experiments with the Earth's gravitation and monocular vision methods at low frequencies and the traditional MLPFS and NS methods confirm that the investigated method can calibrate the sensitivity magnitude and phase of accelerometers with the uncertainties of about 0.5% and 0.5° in the range from 0.1 Hz to 20 kHz.

This study aimed to establish and validate a machine learning (ML) model for predicting in-hospital mortality in patients with sepsis-associated acute kidney injury (SA-AKI). This study collected data on SA-AKI patients from 2008 to 2019 using the Medical Information Mart for Intensive Care IV. After employing Lasso regression for feature selection, six ML approaches were used to build the model. The optimal model was chosen based on precision and area under curve (AUC). In addition, the best model was interpreted using SHapley Additive exPlanations (SHAP) values and Local Interpretable Model-Agnostic Explanations (LIME) algorithms. There were 8129 sepsis patients eligible for participation; the median age was 68.7 (interquartile range: 57.2–79.6) years, and 57.9% (4708/8129) were male. After selection, 24 of the 44 clinical characteristics gathered after intensive care unit admission remained linked with prognosis and were utilized developing ML models. Among the six models developed, the eXtreme Gradient Boosting (XGBoost) model had the highest AUC, at 0.794. According to the SHAP values, the sequential organ failure assessment score, respiration, simplified acute physiology score II, and age were the four most influential variables in the XGBoost model. Individualized forecasts were clarified using the LIME algorithm. We built and verified ML models that excel in early mortality risk prediction in SA-AKI and the XGBoost model performed best.