Article Contents
INTRODUCTION: Gravity-induced loss of consciousness (G-LOC) is a major threat to fighter pilots and may result in fatal accidents. The brain has a period of 5–6 s from the onset of high +Gz exposure, called the functional buffer period, during which transient ischemia is tolerated without loss of consciousness. We tried to establish a method for predicting G-LOC within the functional buffer period by using machine learning. We used a support vector machine (SVM), which is a popular classification algorithm in machine learning.
METHODS: The subjects were 124 flight course students. We used a linear soft-margin SVM, a nonlinear SVM Gaussian kernel function (GSVM), and a polynomial kernel function, for each of which 10 classifiers were built every 0.5 s from the onset of high +Gz exposure (Classifiers 0.5–5.0) to predict G-LOC. Explanatory variables used for each SVM were age, height, weight, with/without anti-G suit, +Gz level, cerebral oxyhemoglobin concentration, and deoxyhemoglobin concentration.
RESULTS: The performance of GSVM was better than that of other SVMs. The accuracy of each classifier of GSVM was as follows: Classifier 0.5, 58.1%; 1.0, 54.8%; 1.5, 57.3%; 2.0, 58.1%; 2.5, 64.5%; 3.0, 63.7%; 3.5, 65.3%; 4.0, 64.5%; 4.5, 64.5%; and 5.0, 64.5%.
CONCLUSION: We could predict G-LOC with an accuracy rate of approximately 65% from 2.5 s after the onset of high +Gz exposure by using GSVM. Analysis of a larger number of cases and factors to enhance accuracy may be needed to apply those classifiers in centrifuge training and actual flight.
Ohrui N, Iino Y, Kuramoto K, Kikukawa A, Okano K, Takada K, Tsujimoto T. G-induced loss of consciousness prediction using a support vector machine. Aerosp Med Hum Perform. 2024; 95(1):29–36.
Profiles of centrifuge training in the Aeromedical Laboratory, Japan Air Self-Defense Force (advanced training 2 and 6). In this study, the following centrifuge training conditions were analyzed: A) during +6 Gz in Profile 2 (+6 Gz, onset rate of 1 Gz · s−1, without an anti-G suit); B) during +7 Gz in Profile 2 (+7 Gz, onset rate of +1Gz · s−1, without an anti-G suit); and C) during +8 Gz in Profile 6 (+8 Gz, onset rate of +6 Gz · s−1, with an anti-G suit) (circled in dashed lines).
Test data in the feature space and the decision boundary. The margin is the distance between the decision boundary and the closest data points.
Principle of a nonlinear SVM. A) Test data sets in the input space are not linearly separable. B) Feature data sets are nonlinearly transformed into a higher dimension. In this space, the separating hyperplane may be definable. C) Decision boundary in the input space.
Leave-one-out cross validation. This study was conducted on 124 data points: 123 data points were used as training data and the remaining 1 data point was used as test data.
Contributor Notes