Publications by authors named "Megumi Oya"
Article Synopsis
- Machine learning, particularly the random forest classifier, is being utilized to help manage the dry weight of patients undergoing hemodialysis, a task that involves complex decision-making based on multiple health indicators.
- A study involving 69,375 dialysis records from 314 Asian patients demonstrated that the classifier could effectively predict dry weight adjustments, showing areas under the curve of 0.70 and 0.74 for upward and downward adjustments, respectively.
- Key indicators for dry weight adjustments were identified: a decline in median blood pressure correlated with upward adjustments, while elevated C-reactive protein and hypoalbuminemia were linked to downward adjustments, suggesting potential clinical applications for improving patient care.
Background: Some attempts have been made to detect atrial fibrillation (AF) with a wearable device equipped with photoelectric volumetric pulse wave technology, and it is expected to be applied under real clinical conditions.
Objective: This study is the second part of a 2-phase study aimed at developing a method for immediate detection of paroxysmal AF, using a wearable device with built-in photoplethysmography (PPG). The objective of this study is to develop an algorithm to immediately diagnose AF by an Apple Watch equipped with a PPG sensor that is worn by patients undergoing cardiac surgery and to use machine learning on the pulse data output from the device.
Article Synopsis
- The study explores the use of 3D convolutional neural networks (3D-CNNs) for accurately segmenting clinical target volumes (CTV) in whole breast irradiation for breast cancer treatment.
- A 3D-UNet model was trained on three different patient datasets (left, right, and both sides) and achieved high segmentation accuracy, with Dice similarity coefficients averaging around 0.85-0.89.
- Gradient-weighted class activation mapping (Grad-CAM) was used to visualize and understand where the 3D-UNet was focusing in the breast tissue during segmentation, highlighting its effectiveness and any potential limitations.