Naoki Sugiyama

Background/Objectives: Posture is a significant indicator of health status in older adults. This study aimed to develop an automatic posture assessment tool based on sagittal photographs by validating recognition models using convolutional neural networks. Methods: A total of 9140 images were collected with data augmentation, and each image was labeled as either Ideal or Non-Ideal posture by physical therapists. The hidden and output layers of the models remained unchanged, while the loss function and optimizer were varied to construct four different model configurations: mean squared error and Adam (MSE & Adam), mean squared error and stochastic gradient descent (MSE & SGD), binary cross-entropy and Adam (BCE & Adam), and binary cross-entropy and stochastic gradient descent (BCE & SGD). Results: All four models demonstrated an improved accuracy in both the training and validation phases. However, the two BCE models exhibited divergence in validation loss, suggesting overfitting. Conversely, the two MSE models showed stability during learning. Therefore, we focused on the MSE models and evaluated their reliability using sensitivity, specificity, and Prevalence-Adjusted Bias-Adjusted Kappa (PABAK) based on the model’s output and correct label. Sensitivity and specificity were 85% and 84% for MSE & Adam and 67% and 77% for MSE & SGD, respectively. Moreover, PABAK values for agreement with the correct label were 0.69 and 0.43 for MSE & Adam and MSE & SGD, respectively. Conclusions: Our findings indicate that the MSE & Adam model, in particular, can serve as a useful tool for screening inspections.

Japan has the highest aging rate worldwide, underscoring the importance of maintaining daily health for older adults. Postural assessment serves as a valuable indicator of health status. The purpose of this study is to construct an automatic posture recognition model using photographs. As a preliminary investigation, pre-processing methods suitable for machine learning datasets was examined. A total of 278 older adults from sagittal were captured using Kinect v2. the photographs were cropped to exclude non-relevant areas and transformed into grayscale. Subsequently, the cropped images underwent background removed, four edge-detection methods (Prewitt, Sobel, Laplacian 4-neighbors, and Laplacian 8-neighbors), and silhouette extraction, respectively, along with the original images, resulting in seven distinct datasets. A posture the images were classified into Ideal and Non-ideal categories according to physical therapists. The recognition model employed a Support Vector Machine (SVM), with feature extraction methods utilizing Scale-Invariant Feature Transform (SIFT) and Speeded-Up Robust Features (SURF). The dataset was divided into training (70%) and test (30%) subsets, with 15 cross-validation sets generated for robustness. Results showed that the Prewitt edge detection method achieved the highest average of F1 score (0.45 ± 0.07) with SIFT, while silhouette extraction yielded the best performance (0.48 ± 0.08) with SURF. The overall accuracy was relatively low; however, when compared to the cropping images, all methods demonstrated higher values, and the order of accuracy was clearly established. These results suggest that further improvements in accuracy could be achieved through tuning the recognition model, highlighting the potential applicability to deep learning frameworks.

Postural assessment is one of the indicators of health status in older adults. Since the number of older adults is on the rise, it is essential to assess simpler methods and automated ones in the future. Therefore, we focused on a visual method (imaging method). The purpose of this study is to determine the degree of agreement between the imaging method and the palpation and visual methods (clinical method). In addition, the influence of differences in the information content of the sagittal plane images on the assessment was also investigated. In this experiment, 28 sagittal photographs of older adults whose posture had already been assessed using the clinical method were used. Furthermore, based on these photographs, 28 gray and silhouette images (G and S images) were generated, respectively. The G and S images were assessed by 28 physical therapists (PTs) using the imaging method. The assessment was based on the Kendall classification, with one of four categories selected for each image: ideal, kyphosis lordosis, sway back, and flat back. Cross-tabulation matrices of the assessments using the clinical method and imaging method were created. In this table, four categories and two categories of ideal and non-ideal (KL, SB, and FB) were created. The agreement was evaluated using the prevalence-adjusted bias-adjusted kappa (PABAK). In addition, sensitivity and specificity were calculated to confirm the reliability. When comparing the clinical and imaging methods in the four posture categories, the PABAK values were −0.14 and −0.29 for the S and G images, respectively. In the case of the two categories, the PABAK values were 0.57 and 0.5 for the S and G images, respectively. The sensitivity and specificity were 86% and 57% for the S images and 76% and 71% for the G images, respectively. The four categories show that the imaging method is difficult to assess regardless of the image processing. However, in the case of the two categories, the same assessment of the clinical method applied to the imaging method for both the S and G images. Therefore, no differences in image processing were observed, suggesting that PTs can identify posture using the visual method.

This study compares the piloting practices and drone flight trajectories of skilled and novice drone pilots. Markers for 3D movement analysis were attached to the fingers that move the control stick. Similarly, drones were also marked and the flight movement of the drones analyzed. These two sets of data were cross-checked to examine the characteristics of the subjects. As a result, the following results were obtained.・The expert pilot did not adjust the position of the object directly in front of the object to be photographed, but at a distance of about 90 mm in the lateral direction.・The expert moved the drone in both the first axis and the second axis directions

The use of drones provides a variety of images and video footage that we have not seen before. In non-destructive inspection, on the other hand, it is necessary to obtain an accurate image of the inspection area. The quality of the image is important because the image is used for inspection.In this study, expert and beginners drone pilots operated a drone to photograph the three subjects in the designated areas. Three subjects were photographed by different conditions. The quality of the photographs obtained was compared. The results showed that expert pilot were more likely than beginners to ensure that the subject was in the centre of the picture taken. In addition, distances between drone and subjects were set in almost the same position by expert.