An explainable machine learning model to predict early and late acute kidney injury after major hepatectomy.

BACKGROUND: Risk assessment models for acute kidney injury (AKI) after major hepatectomy that differentiate between early and late AKI are lacking. This retrospective study aimed to create a model predicting AKI through machine learning and identify features that contribute to the development of early and late AKI. METHODS: Patients that underwent major hepatectomy were categorized into the No-AKI, Early-AKI (within 48 h) or Late-AKI group (between 48 h and 7 days). Modeling was done with 20 perioperative features and the performance of prediction models were measured by the area under the receiver operating characteristic curve (AUROCC). Shapley Additive Explanation (SHAP) values were utilized to explain the outcome of the prediction model. RESULTS: Of the 1383 patients included in this study, 1229, 110 and 44 patients were categorized into the No-AKI, Early-AKI and Late-AKI group, respectively. The CatBoost classifier exhibited the greatest AUROCC of 0.758 (95% CI: 0.671-0.847) and was found to differentiate well between Early and Late-AKI. We identified different perioperative features for predicting each outcome and found 1-year mortality to be greater for Early-AKI. CONCLUSIONS: Our results suggest that risk factors are different for Early and Late-AKI after major hepatectomy, and 1-year mortality is greater for Early-AKI.

Thermal evaluation of laser exposures in an in vitro retinal model by microthermal sensing.

A temperature detection system using a micropipette thermocouple sensor was developed for use within mammalian cells during laser exposure with an 8.6-µm beam at 532 nm. We have demonstrated the capability of measuring temperatures at a single-cell level in the microscale range by inserting micropipettebased thermal sensors of size ranging from 2 to 4 µm into the membrane of a live retinal pigment epithelium (RPE) cell subjected to a laser beam. We setup the treatment groups of 532-nm laser-irradiated single RPE cell and in situ temperature recordings were made over time. Thermal profiles are given for representative cells experiencing damage resulting from exposures of 0.2 to 2 s. The measured maximum temperature rise for each cell ranges from 39 to 73°C; the RPE cells showed a signature of death for all the cases reported herein. In order to check the cell viability, real-time fluorescence microscopy was used to identify the transition of pigmented RPE cells between viable and damaged states due to laser exposure.

Measurement of the thermal conductivity of a water-based single-wall carbon nanotube colloidal suspension with a modified 3- omega method.

A modified 3-omega method applied to a suspended platinum microwire was employed to measure the thermal conductivity and convective heat transfer coefficient of a water-based single-walled carbon nanotube (CNT) solution (metallic single-wall nanotubes with 1.33 nm diameter and 1.14 wt% concentration), and an expression for calculating the convective heat transfer coefficient in such a free convective fluid was introduced. The measurement technique was validated for three model systems including vacuum, air and deionized water. It is found that there is excellent agreement between these three model systems with theoretical predictions. In addition, the frequency dependence on the third harmonic response measured in deionized water reveals the existence of a very low working frequency below 60 mHz. The thermal conductivity and convective heat transfer coefficient of the nanofluid (water-based single-wall CNT solution) were determined to be 0.73 +/- 0.013 W m(-1) K(-1) and 14 900 +/- 260 W m(-2) K(-1), respectively, which correspond to an enhancement of 19.4% in thermal conductivity and 18.9% in convective heat transfer as compared to water.

Fabrication and electrical characterization of circuits based on individual tin oxide nanowires.

Two- and four-probe electrical measurements on individual tin oxide (SnO(2)) nanowires were performed to evaluate their conductivity and contact resistance. Electrical contacts between the nanowires and the microelectrodes were achieved with the help of an electron- and ion-beam-assisted direct-write nanolithography process. High contact resistance values and the nonlinear current-bias (I-V) characteristics of some of these devices observed in two-probe measurements can be explained by the existence of back-to-back Schottky barriers arising from the platinum-nanowire contacts. The nanoscale devices described herein were characterized using impedance spectroscopy, enabling the development of an equivalent circuit. The proposed methodology of nanocontacting and measurements can be easily applied to other nanowires and nanometre-sized materials.