Polysomnography (PSG) is treated as the gold standard for diagnosing obstructive sleep apnea (OSA). However, it is labor-intensive, time-consuming, and expensive. This study evaluates validity of overnight pulse oximetry as a diagnostic tool for moderate to severe OSA patients. A total of 699 patients with possible OSA were recruited for overnight oximetry and PSG examination at the Sleep Center of a University Hospital from Jan. 2004 to Dec. 2005. By excluding 23 patients with poor oximetry recording, poor EEG signals, or respiratory artifacts resulting in a total recording time less than 3 hours; 12 patients with total sleeping time (TST) less than 1 hour, possibly because of insomnia; and 48 patients whose ages less than 20 or more than 85 years old, data of 616 patients were used for further study. By further considering 76 patients with TST < 4 h, a group of 540 patients with TST ≥ 4 h was used to study the effect of insufficient sleeping time. Alice 4 PSG recorder (Respironics Inc., USA) was used to monitor patients with suspected OSA and to record their PSG data. After statistical analysis and feature selection, models built based on support vector machine (SVM) were then used to diagnose moderate and moderate to severe OSA patients with a threshold of AHI = 30 and AHI = 15, respectively. The SVM models designed based on the oxyhemoglobin desaturation index (ODI) derived from oximetry measurements provided an accuracy of 90.42-90.55%, a sensitivity of 89.36-89.87%, a specificity of 91.08-93.05%, and an area under ROC curve (AUC) of 0.953-0.957 for the diagnosis of severe OSA patients; as well as achieved an accuracy of 87.33-87.77%, a sensitivity of 87.71-88.53%, a specificity of 86.38-86.56%, and an AUC of 0.921-0.924 for the diagnosis of moderate to severe OSA patients. The predictive outcome of ODI to diagnose severe OSA patients is better than to diagnose moderate to severe OSA patients. Overnight pulse oximetry provides satisfactory diagnostic performance in detecting severe OSA patients. Home-styled oximetry may be a tool for severe OSA diagnosis.

In recent years, the number of yoga practitioners has been drastically increased and there are more men and older people practice yoga than ever before. IoT-based yoga training system is needed for those who want to practice yoga at home. Some studies have proposed RGB/Kinect camera-based or wearable device-based yoga posture recognition methods with a high accuracy; however, the former has a privacy issue and the latter is impractical in the long-term application. Thus, this study proposes an IoT-based privacy-preserving yoga posture recognition system employing a deep convolutional neural network (DCNN) and a low-resolution infrared sensor-based wireless sensor network (WSN). The WSN has three nodes (x, y, and z-axes) where each integrates 8×8 pixels’ thermal sensor module and a Wi-Fi module for connecting the deep learning server. We invited 18 volunteers to perform 26 yoga postures for two sessions each lasted for 20 seconds. First, recorded sessions are saved as.csv files, then preprocessed and converted to grayscale posture images. Totally, 93200 posture images are employed for the validation of the proposed DCNN models. The 10-fold cross-validation results revealed that F1-scores of the models trained with xyz (all 3-axes) and y (only y-axis) posture images were 0.9989 and 0.9854, respectively. An average latency for a single posture image classification on the server was 107 ms. Thus, we conclude that the proposed IoT-based yoga posture recognition system has a great potential in the privacy-preserving yoga training system.

In this era of ubiquitous information, patient record exchange among hospitals still has technological and individual barriers including resistance to information sharing. Most research on user attitudes has been limited to one type of user or aspect. Because few analyses of attitudes toward electronic patient records (EPRs) have been conducted, understanding the attitudes among different users in multiple aspects is crucial to user acceptance. This proof-of-concept study investigated the attitudes of users toward the inter-hospital EPR exchange system implemented nationwide and focused on discrepant behavioral intentions among three user groups. The system was designed by combining a Health Level 7-based protocol, object-relational mapping, and other medical informatics techniques to ensure interoperability in realizing patient-centered practices. After implementation, three user-specific questionnaires for physicians, medical record staff, and patients were administered, with a 70 % response rate. The instrument showed favorable convergent construct validity and internal consistency reliability. Two dependent variables were applied: the attitudes toward privacy and support. Independent variables comprised personal characteristics, work characteristics, human aspects, and technology aspects. Major statistical methods included exploratory factor analysis and general linear model. The results from 379 respondents indicated that the patients highly agreed with privacy protection by their consent and support for EPRs, whereas the physicians remained conservative toward both. Medical record staff was ranked in the middle among the three groups. The three user groups demonstrated discrepant intentions toward privacy protection and support. Experience of computer use, level of concerns, usefulness of functions, and specifically, reason to use electronic medical records and number of outpatient visits were significantly associated with the perceptions. Overall, four categories of independent variables were associated with the mean difference in the perceptions. Discrepant attitudes toward privacy and support among the three user groups are identified. Patients may require further education and communication regarding the system. Culturally fit e-Consent should be incorporated into the system to fully utilize the computing power of the Internet when also considering workload. The concern for misuse of EPRs might lead to low support among physicians. Highly readable EPR documents and managerial incentives for information exchange may improve system use.

Image retake of radiological examinations not only increases the risk of radiation exposure of the patients, but also wastes the medical resource and degrades the quality of services of the hospitals. This study aimed at discovering factors affecting image retake of general digital radiography for setting guidelines to reduce the image retaking rate. A total of 98,503 general X-ray images retrieved from the picture archiving and communication system database of a medical center in central Taiwan were analyzed. The results showed that the total retaking rate was 4.89% with the position error (56.05%) was the main factor causing image retakes and chest examination showed the highest frequency (1544 cases). On the other hand, skull/face exhibited the highest retaking rate (9.81%) among various types of examinations. After discovering the factors affecting the image retaking rate, suitable guidelines were proposed and introduced. The image retake rate had been significantly reduced to 4.38% and 3.57% 1 month and 6 months, respectively, after the introduction of guidelines. In conclusion, image retake analysis is a quality indicator and is effective for quality assurance of digital radiology. Regular analysis of image retake can find factors inducing image retake and is useful for designing guidelines to reduce the image retake rate.

Convolutional neural networks (CNNs) have shown great promise in human activity recognition (HAR), but long-term dependencies in time series data can be difficult to capture using standard CNNs. This study introduces a new CNN architecture that incorporates a multihead attention mechanism (CNN-MHA) to address this challenge. This mechanism is composed of several attention heads, each independently calculating attention weights for distinct segments of the input. The attention head outputs are then concatenated and processed through a fully connected layer to produce the final attention representation. A multihead attention (MHA) mechanism allows the network to focus on relevant features and maintain long-term dependencies in the input data. The proposed model is evaluated on the physical activity monitoring for aging people data set (PAMAP2) from the UCI machine learning repository, which is preprocessed by cleaning, normalization, segmentation, and reshaping before splitting into training, validation, and testing sets. The experimental results demonstrate that the CNN-MHA model outperforms existing models, achieving F1-score of 95.7%. Particularly, the MHA mechanism significantly improves the model’s ability to recognize complex activity patterns. Furthermore, our model attained an average inference latency of 0.304 s, which can be crucial in real-time applications. The findings clearly demonstrate the substantial promise of the proposed CNN-MHA architecture for optimizing HAR tasks, offering a powerful tool for advancing the state-of-the-art in this domain.