Improvement of human error prediction accuracy in single-trial analysis of electroencephalogram.

The prevention of human error is an important task that has already been researched. Previous studies have shown that EEG signals can predict the occurrence of human errors. However, high accuracy has not yet been achieved in a single-trial analysis. This study is aimed to improve the accuracy of single-trial analysis, and propose a method for anomaly detection with auto encoder(AE). In the experiment, we conducted "Press the button(Go)" or "Do nothing(No-Go)" according to the visual stimulus and analyzed the EEG signal from -1000 ms to 0 ms when the stimulus was displayed. We prepared two types of inputs, time series data and frequency spectrum, and an AE was trained to reconstruct the inputs. We then calculated the difference between the reconstructed data and input data and predicted human error by its largeness. In the prediction using Support Vector Machine (SVM) based on the frequency spectrum, some over-fitting occurred and the average accuracy was 43 %. In the prediction using anomaly detection with frequency spectrum was 53 % and could not be classified. The time series data was 63 % which improved the accuracy. A previous study has shown frequency-dependent features such as -band activity and rhythm, as precursors of human error. However, in single-trial analysis, we obtained a higher accuracy by time series data than when by using the frequency spectrum. However, there was no noticeable difference between SVM and anomaly detection methods other than over-fitting. Therefore, in this case, the improvement in accuracy by the anomaly detection method could not be confirmed. However, the result suggests that it is more effective to use the frequency spectrum than the time series data in the single-trial analysis in the future.

Identification of Vortex Cores in Cerebral Aneurysms on 4D Flow MRI.

BACKGROUND AND PURPOSE: The complexity and instability of the vortex flow in aneurysms are factors related to the rupture risk of unruptured cerebral aneurysms. We identified aneurysm vortex cores on 4D flow MR imaging and examined the relationship of these factors with the characteristics of cerebral aneurysms. MATERIALS AND METHODS: We subjected 40 aneurysms (37 unruptured, 3 ruptured) to 4D flow MR imaging. We visualized streamlines with velocities below the threshold-that is, a percentage value of the aneurysm maximum inflow velocity-and progressively decreased the threshold to identify vortex cores as thin, streamline bundles with minimum velocities. Complexity and stability were compared in aneurysms with a smooth surface and those with blebs or daughter sacs. RESULTS: The threshold for visualizing vortex cores ranged from 3% to 13% of the maximum inflow velocity. Vortex cores could be visualized in 38 aneurysms; in 2, they were not visualized through the cardiac cycle. A simple flow pattern (single vortex core) was identified in 27 aneurysms; the other 13 exhibited a complex flow pattern. The cores were stable in 32 and unstable in 8 aneurysms. Significantly more aneurysms with-than-without blebs or daughter sacs had a complex flow pattern (P = .006). Of the 3 ruptured aneurysms, 1 aneurysm had an unstable vortex core; in the other 2, the vortex core was not visualized. CONCLUSIONS: The identification of vortex cores on 4D flow MR imaging may help to stratify the rupture risk of unruptured cerebral aneurysms.

[Separate evaluation of beta-methyl fatty acid uptake and perfusion in rat myocardium].

The kinetics and distribution of I-125 beta-methyl iodophenyl pentadecanoic acid (BMIPP) in rat's heart were studied for separate evaluation of perfusion and metabolism. Tl-201 and BMIPP were simultaneously injected. The experimental groups consisted of control (C), glucose (G) and sodium lactate loaded group (L). In C, myocardial uptake at 5 minutes after BMIPP injection was 3.60% ID/g and remained constant up to 60 minutes. The myocardium/lung ratio (2.44) and the myocardium/muscle ratio (4.55) of BMIPP were almost equal to those of Tl-201. But myocardium/liver ratio was low (1.31). In G, myocardial uptake of BMIPP (1.94 +/- 0.36% ID/g) and g-BMIPP/Tl (0.31 +/- 0.03) at 15 minutes after injection were significantly decreased (p less than 0.001) than those of C (3.16 +/- 0.18% ID/g and 0.48 +/- 0.05). In L. myocardial perfusion was decreased and g-BMIPP/Tl (0.73 +/- 0.14) was significantly higher (p less than 0.01) than those of C. Coefficient of variance of the density within a myocardium, and the ratio of inner to outer layer of myocardium (I/O ratio) were calculated from autoradiogram by videodensitometry. The myocardial distribution of BMIPP was more inhomogeneous, and the I/O ratio was lower than that of Tl-201, although these were not specific for metabolic interventions. In conclusion BMIPP is suitable for SPECT imaging and dual nuclide imaging by BMIPP and Tl-201 will provide informations about myocardial fatty acid metabolism and perfusion.