Publications by authors named "Koji Morikawa"
Background: Early prediction of oxygen therapy in patients with coronavirus disease 2019 (COVID-19) is vital for triage. Several machine-learning prognostic models for COVID-19 are currently available. However, external validation of these models has rarely been performed.
View Article and Find Full Text PDFArticle Synopsis
- * Researchers analyzed data from over 10,000 patients undergoing tracheal intubation at 13 facilities, developing seven different machine learning models to assess difficult airways and first-pass success rates.
- * The findings indicate that machine learning models significantly outperform the modified LEMON criteria in predicting difficult airways and show enhanced predictive ability for first-pass success, suggesting these models could lead to better patient outcomes in emergency settings.
This study presents a new blood pressure (BP) estimation algorithm utilizing machine learning (ML). A cuffless device that can measure BP without calibration would be precious for portability, continuous measurement, and comfortability, but unfortunately, it does not currently exist. Conventional BP measurement with a cuff is standard, but this method has various problems like inaccurate BP measurement, poor portability, and painful cuff pressure.
View Article and Find Full Text PDFEstimation of respiratory signal from thoracic impedance cardiography in low electrical current.
Annu Int Conf IEEE Eng Med Biol Soc
July 2015
Article Synopsis
- A new method for estimating respiratory signals from thoracic impedance is introduced, aiming to improve battery-powered, wearable monitoring devices.
- Traditional methods struggle under low current conditions due to the larger cardiac signal overshadowing the respiratory signal.
- The proposed approach uses an envelope curve derived from the cardiac T waves, resulting in higher accuracy compared to conventional techniques.
Estimating uncomfortable loudness levels using evoked potentials to auditory stimuli for hearing aid fitting.
Annu Int Conf IEEE Eng Med Biol Soc
August 2013
Article Synopsis
- The study aims to develop a method for estimating the uncomfortable loudness level (UCL) for individuals with hearing loss using auditory evoked potentials (AEPs).
- It involved 18 adults with normal hearing, who listened to a series of tone bursts at decreasing sound intensities while their brain activity was recorded via an electroencephalogram (EEG).
- The results indicated that UCLs could be effectively estimated, with a mean error of only 4.9 dB, suggesting this method may help in understanding loudness tolerance in those with hearing impairments.
This study investigated the relationship between event-related brain potentials (ERPs) to abridged content information in the media and the subsequent decisions to view the full content. Student volunteers participated in a task that simulated information selection on the basis of the content information. Screenshots of television clips and headlines of news articles on the Web were used as content information for the image condition and the headline condition, respectively.
View Article and Find Full Text PDFIdentification of event-related potentials elicited by conceptual mismatch between expectations and self-chosen TV images.
Appl Psychophysiol Biofeedback
September 2011
Article Synopsis
- The study investigates how unexpected outcomes from voluntary actions, such as choosing TV programs, affect brain responses, specifically the late positive potential (LPP).
- Twelve student volunteers participated in trials that involved selecting from different types of cues, including image and text categories, to simulate expectations.
- Results indicated that when the outcome did not match expectations, participants displayed a larger LPP with image and text cues, and a distinct P3 response with images, highlighting how we process mismatches between our expectations and outcomes.
Charles Taylor and John Marshall explain the utility of mathematical modeling for evaluating the effectiveness of population replacement strategy. Insight is given into how computational models can provide information on the population dynamics of mosquitoes and the spread of transposable elements through A. gambiae subspecies.
View Article and Find Full Text PDFEvent-related potentials elicited by unexpected visual stimuli after voluntary actions.
Int J Psychophysiol
December 2007
Event-related potentials from visual stimuli that were presented after voluntary actions were recorded to examine how people expect their action effects. Participants pressed a button in response to a cue stimulus (L or R) either in the fixed condition where participants always pressed a center button or in the choice condition where they selectively pressed the corresponding left or right button. Immediately after the button press, a second stimulus (left or right) was presented visually to inform that their action was registered.
View Article and Find Full Text PDF