Search for a command to run...
Gravity-induced loss of consciousness (G-LOC) is a transient cerebral ischemic episode experienced by fighter pilots upon sudden exposure to +Gz (head-to-foot) hyper-gravity. This condition renders pilots incapable of controlling the aircraft and is therefore likely to precipitate aviation accidents. Accordingly, the establishment of objective methods to detect G-LOCs is urgently required. Focusing on head tilt during the G-LOC, we aimed to develop a depth camera-based detection method. We collected RGB (red-green-blue)/depth video and +Gz levels from student pilots undergoing hyper-gravity training (cohort 1: n = 102; 22.5 ± 2.0 yr). RGB video analysis showed forward, rightward, or leftward head tilts in approximately 87% of the G-LOC events. For this reason, we devised a method to calculate the head-region area within the field of view of a depth camera under both normal and G-LOC conditions. We determined the optimal area thresholds for G-LOC detection through receiver operating characteristic analysis. By integrating these thresholds with a minimum +Gz level of 4.0 G that induces G-LOC, we prototyped a G-LOC detection criterion. Applying the prototype criterion to an independent test dataset (cohort 2: n = 102; 23.3 ± 1.9 yr) for performance evaluation, this criterion achieved a sensitivity of 89.1%, a specificity of 74.3%, and an accuracy of 78.3%. These findings indicate the efficacy of the depth sensor approach, combined with a +Gz level, as a novel method for detecting G-LOC.