Predicting driver's takeover time based on individual characteristics, external environment, and situation awareness.

The driver's takeover time is crucial to ensure a safe takeover transition in conditional automated driving. The study aimed to construct a prediction model of driver's takeover time based on individual characteristics, external environment, and situation awareness variables. A total of 18 takeover events were designed with scenarios, non-driving-related tasks, takeover request time, and traffic flow as variables. High-fidelity driving simulation experiments were carried out, through which the driver's takeover data was obtained. Fifteen basic factors and three dynamic factors were extracted from individual characteristics, external environment, and situation awareness. In this experiment, these 18 factors were selected as input variables, and XGBoost and Shapely were used as prediction methods. A takeover time prediction model (BM + SA model) was then constructed. Moreover, we analyzed the main effect of input variables on takeover time, and the interactive contribution made by the variables. And in this experiment, the 15 basic factors were selected as input variables, and the basic takeover time prediction model (BM model) was constructed. In addition, this study compared the performance of the two models and analyzed the contribution of input variables to takeover time. The results showed that the goodness of fit of the BM + SA model (Adjusted_R2) was 0.7746. The XGBoost model performs better than other models (support vector machine, random forest, CatBoost, and LightBoost models). The relative importance degree of situation awareness variables, individual characteristic variables, and external environment variables to takeover time gradually reduced. Takeover time increased with the scan and gaze durations and decreased with pupil area and self-reported situation awareness scores. There was also an interaction effect between the variables to affect takeover time. Overall, the performance of the BM + SA model was better than that of the BM model. This study can provide support for predicting driver's takeover time and analyzing the mechanism of influence on takeover time. This study can provide support for the development of real-time driver's takeover ability prediction systems and optimization of human-machine interaction design in automated vehicles, as well as for the management department to evaluate and improve the driver's takeover performance in a targeted manner.

Prognostic value of serum ammonia in critical patients with non-hepatic disease: A prospective, observational, multicenter study.

Background and Objectives: Non-hepatic hyperammonemia can damage the central nervous system (CNS), and possible prognostic factors are lacking. This study aimed to investigate the prognostic and risk factors for patients admitted to the intensive care unit (ICU). Materials and Methods: This prospective, observational, multicenter study was conducted between November and December 2019 at 11 ICUs in the Chinese Heilongjiang province. Changes in blood ammonia level during and after ICU admission were continuously monitored and expressed as the high level (H-), mean level (M-), and initial level (I-) of ammonia. The risk factors of poor prognosis were investigated by conducting univariate and multivariate logistic regression analyses. Receiver operating characteristic (ROC) curve analysis was conducted to compare the predictive ability of Acute Physiologic Assessment and Chronic Health Evaluation II (APACHE-II) score, lactic acid, total bilirubin (TBil), and M-ammonia. Results: A total of 1060 patients were included in this study, of which 707 (67%) had a favorable prognosis and 353 (33%) had a poor prognosis. As shown by univariate models, a poor prognosis was associated with elevated serum levels of lactic acid, TBil, and ammonia (P < 0.05) and pathologic scores from three assessments: APACHE-II, Glasgow Coma Scale (GCS), and Sequential Organ Failure Assessment (SOFA). Multivariate analysis revealed that circulating mean ammonia levels in ICU patients were independently associated with a poor prognosis (odds ratio [OR] = 1.73, 95% confidence interval [CI]: 1.07-2.80, P = 0.02). However, the APACHE-II score (area under the curve [AUC]: 0.714, sensitivity: 0.86, specificity: 0.68, P < 0.001) remained the most predictive factor for patient prognosis by ROC analysis. Conclusion: Elevated serum levels of ammonia in the blood were independently prognostic for ICU patients without liver disease.

[Natural history of small hepatocellular carcinoma: studied by MRI].

OBJECTIVE: To study natural history of small hepatocellular carcinoma (sHCC) and value of MRI in this research. METHODS: From 1999 to 2002, serial MR imagings were performed in 192 patients with cirrhosis no less than twice in order to detect sHCC focus. Characteristic MRI findings of focus were analysed. Diameters of focus (D) were measured. Tumorigenesis detected time (T) was counted; T and D were analysed by statistics and tumor volume doubling time (T(d)) was calculated. RESULTS: Fifty-eight new tumor foci in 33 patients were found, including 31 foci in 23 patients developed on the basis of cirrhosis, with a 12.0% cumulative tumorigenic rate of three years. The tumorigenesis detected time (T) was 386.9 +/- 256.4 days and the maximum diameter (D) was 2.258 +/- 1.074 cm. T was statistical correlated with D(3) (P < 0.01) and regressive formulas were established: D(3) = -2.69 + 0.058T (P = 0.0007), D(3) = -72.13 + 16.04lnT (P = 0.0064). Tumor volume doubling time was 20 approximately 279 days (mean: 104 days). CONCLUSION: The percentage of tumorigenesis is high in patients with cirrhosis. Regular MRI examinations are necessary to detect sHCC early. The optimal interval time is 3 approximately 6 months.

[Clinical value of MR cholangiopancreatography combining with dynamic enhanced MRI in the detection of cholangiocarcinoma].

OBJECTIVES: To evaluate the clinical value of breath-hold magnetic resonance cholangiopancreatography (MRCP) combining with dynamic enhanced MRI in the diagnosis of cholangiocarcinoma. METHODS: MRCP findings of 88 cholangiocarcinoma patients proved surgically and pathologically were analyzed retrospectively. RESULTS: MRCP examination succeeded in all the 88 patients and the pancreaticobiliary ducts were shown satisfactorily. The accuracy of MRCP in the location of both hilar and extrahepatic cholangiocarcinoma was 100%, and the accuracy of detecting hilar and extrahepatic cholangiocarcinoma were 100% and 52.2%, respectively. Combining with dynamic enhanced MRI, the detecting accuracy of extrahepatic cholangiocarcinoma improved to 91.3%. CONCLUSIONS: MRCP examination has a high successful rate and can accurately determine the location of hilar and extrahepatic cholangiocarcinoma, and the accuracy of qualitative diagnosis for the former two is high. Combining with dynamic enhanced MRI, the specificity of determining extrahepatic cholangiocarcinoma is also high.