Mortality prediction of rats in acute hemorrhagic shock using machine learning techniques

This study sought to determine a mortality prediction model that could be used for triage in the setting of acute hemorrhage from trauma. To achieve this aim, various machine learning techniques were applied using the rat model in acute hemorrhage. Thirty-six anesthetized rats were randomized into three groups according to the volume of controlled blood loss. Measurements included heart rate (HR), systolic and diastolic blood pressures (SBP and DBP), mean arterial pressure, pulse pressure, respiratory rate, temperature, blood lactate concentration (LC), peripheral perfusion (PP), shock index (SI, SI = HR/SBP), and a new hemorrhage-induced severity index (NI, NI = LC/PP). NI was suggested as one of the good candidates for mortality prediction variable in our previous study. We constructed mortality prediction models with logistic regression (LR), artificial neural networks (ANN), random forest (RF), and support vector machines (SVM) with variable selection. The SVM model showed better sensitivity (1.000) and area under curve (0.972) than the LR, ANN, and RF models for mortality prediction. The important variables selected by the SVM were NI and LC. The SVM model may be very helpful to first responders who need to make accurate triage decisions and rapidly treat hemorrhagic patients in cases of trauma.     

Measurement of Opening and Closing Angles of Aortic Valve Prostheses In Vivo Using Dual-Source Computed Tomography: Comparison with Those of Manufacturers' in 10 Different Types

Objective

The aims of this study were to compare opening and closing angles of normally functioning mechanical aortic valves measured on dual-source computed tomography (CT) with the manufacturers'' values and to compare CT-measured opening angles according to valve function.

Materials and Methods

A total of 140 patients with 10 different types of mechanical aortic valves, who underwent dual-source cardiac CT, were included. Opening and closing angles were measured on CT images. Agreement between angles in normally functioning valves and the manufacturer values was assessed using the interclass coefficient and the Bland-Altman method. CT-measured opening angles were compared between normal functioning valves and suspected dysfunctioning valves.

Results

The CT-measured opening angles of normally functioning valves and manufacturers'' values showed excellent agreement for seven valve types (intraclass coefficient [ICC], 0.977; 95% confidence interval [CI], 0.962-0.987). The mean differences in opening angles between the CT measurements and the manufacturers'' values were 1.2° in seven types of valves, 11.0° in On-X valves, and 15.5° in ATS valves. The manufacturers'' closing angles and those measured by CT showed excellent agreement for all valve types (ICC, 0.953; 95% CI, 0.920-0.972). Among valves with suspected dysfunction, those with limitation of motion (LOM) and an increased pressure gradient (PG) had smaller opening angles than those with LOM only (p < 0.05).

Conclusion

Dual-source cardiac CT accurately measures opening and closing angles in most types of mechanical aortic valves, compared with the manufacturers'' values. Opening angles on CT differ according to the type of valve dysfunction and a decreased opening angle may suggest an elevated PG.