Search

BackgroundElectrocardiogram (ECG) monitoring in daily life is essential for effective management of cardiovascular disease, a leading cause of death. Wearable ECG measurement systems in the form of clothing have been proposed to replace Holter monitors used for clinical ECG monitoring; however, they have limitations in daily use because they compress the upper body and, in doing so, cause discomfort during wear. ObjectiveThe purpose of this study was to develop a wireless wearable ECG monitoring system that includes a textile ECG electrode that can be applied to the lining of pants and can be used in the same way that existing lower clothing is worn, without compression to the upper body. MethodsA textile electrode with stretchable characteristics was fabricated by knitting a conductive yarn together with polyester-polyurethane fiber, which was then coated with silver compound; an ECG electrode was developed by placing it on an elastic band in a modified limb lead configuration. In addition, a system with analog-to-digital conversion, wireless communication, and a smartphone app was developed, allowing users to be able to check and store their own ECGs in real time. A signal processing algorithm was also developed to remove noise from the obtained signal and to calculate the heart rate. To evaluate the ECG and heart rate measurement performance of the developed module, a comparative evaluation with a commercial device was performed. ECGs were measured for 5 minutes each in standing, sitting, and lying positions; the mean absolute percentage errors of heart rates measured with both systems were then compared. ResultsThe system was developed in the form of a belt buckle with a size of 53 × 45 × 12 mm (width × height × depth) and a weight of 23 g. In a qualitative evaluation, it was confirmed that the P-QRS-T waveform was clearly observed in ECGs obtained with the wearable system. From the results of the heart rate estimation, the developed system could track changes in heart rate as calculated by a commercial ECG measuring device; in addition, the mean absolute percentage errors of heart rates were 1.80%, 2.84%, and 2.48% in the standing, sitting, and lying positions, respectively. ConclusionsThe developed system was able to effectively measure ECG and calculate heart rate simply through being worn as existing clothing without upper body pressure. It is anticipated that general usability can be secured through further evaluation under more diverse conditions.

BackgroundAlthough commercially available analgesic indices based on biosignal processing have been used to quantify nociception during general anesthesia, their performance is low in conscious patients. Therefore, there is a need to develop a new analgesic index with improved performance to quantify postoperative pain in conscious patients. ObjectiveThis study aimed to develop a new analgesic index using photoplethysmogram (PPG) spectrograms and a convolutional neural network (CNN) to objectively assess pain in conscious patients. MethodsPPGs were obtained from a group of surgical patients for 6 minutes both in the absence (preoperatively) and in the presence (postoperatively) of pain. Then, the PPG data of the latter 5 minutes were used for analysis. Based on the PPGs and a CNN, we developed a spectrogram–CNN index for pain assessment. The area under the curve (AUC) of the receiver-operating characteristic curve was measured to evaluate the performance of the 2 indices. ResultsPPGs from 100 patients were used to develop the spectrogram–CNN index. When there was pain, the mean (95% CI) spectrogram–CNN index value increased significantly—baseline: 28.5 (24.2-30.7) versus recovery area: 65.7 (60.5-68.3); P

Misidentification errors in tumor marker tests can lead to serious diagnostic and treatment errors. This study aims to develop a method for detecting these errors using a machine learning (ML)-based delta check approach, overcoming limitations of conventional methods. We analyzed five tumor marker test results: alpha-fetoprotein (AFP), cancer antigen 19-9 (CA19-9), cancer antigen 125 (CA125), carcinoembryonic antigen (CEA), and prostate-specific antigen (PSA). A total of 246,261 records were used in the analysis. Of these, 179,929 records were used for model training and 66,332 records for performance evaluation. We developed a misidentification error detection model based on the random forest (RF) and deep neural network (DNN) methods. We performed an in silico simulation with 1 % random sample shuffling. The performance of the developed models was evaluated and compared to conventional delta check methods such as delta percent change (DPC), absolute DPC (absDPC), and reference change values (RCV). The DNN model outperformed the RF, DPC, absDPC, and RCV methods in detecting sample misidentification errors. It achieved balanced accuracies of 0.828, 0.842, 0.792, 0.818, and 0.833 for AFP, CA19-9, CA125, CEA, and PSA, respectively. Although the RF method performed better than DPC and absDPC, it showed similar or lower performance compared to RCV. Our research results demonstrate that an ML-based delta check method can more effectively detect sample misidentification errors compared to conventional delta check methods. In particular, the DNN model demonstrated superior and stable detection performance compared to the RF, DPC, absDPC, and RCV methods. [ABSTRACT FROM AUTHOR]

In this study, we conducted research on a deep learning‐based blood pressure (BP) estimation model suitable for wearable environments. To measure BP while wearing a wearable watch, it needs to be considered that computing power for signal processing is limited and the input signals are subject to noise interference. Therefore, we employed a convolutional neural network (CNN) as the BP estimation model and utilized time‐series electrocardiogram (ECG) and photoplethysmogram (PPG) signals, which are quantifiable in a wearable context. We generated periodic input signals and used differential and thresholding methods to decrease noise in the preprocessing step. We then applied a max‐pooling technique with filter sizes of 2 × 1 and 5 × 1 within a 3‐layer convolutional neural network to estimate BP. Our method was trained, validated, and tested using 2.4 million data samples from 49 patients in the intensive care unit. These samples, totaling 3.1 GB were obtained from the publicly accessible MIMIC database. As a result of a test with 480,000 data samples, the average root mean square error in BP estimation was 3.41, 5.80, and 2.78 mm Hg in the prediction of pulse pressure, systolic BP (SBP), and diastolic BP (DBP), respectively. The cumulative error percentage less than 5 mm Hg was 68% and 93% for SBP and DBP, respectively. In addition, the cumulative error percentage less than 15 mm Hg was 98% and 99% for SBP and DBP. Subsequently, we evaluated the impact of changes in input signal length (1 cycle vs. 30 s) and the introduction of noise on BP estimation results. The experimental results revealed that the length of the input signal did not significantly affect the performance of CNN‐based analysis. When estimating BP using noise‐added ECG signals, the mean absolute error (MAE) for SBP and DBP was 9.72 and 6.67 mm Hg, respectively. Meanwhile, when using noise‐added PPG signals, the MAE for SBP and DBP was 26.85 and 14.00 mm Hg, respectively. Therefore, this study confirmed that using ECG signals rather than PPG signals is advantageous for noise reduction in a wearable environment. Besides, short sampling frames without calibration can be effective as input signals. Furthermore, it demonstrated that using a model suitable for information extraction rather than a specialized deep learning model for sequential data can yield satisfactory results in BP estimation. We have developed a deep learning methodology to predict blood pressure, designed specifically for use in noisy wearable environments, and using input signals from electrocardiograms and photoplethysmograms. Through comprehensive experimentation, we determined which signals within electrocardiograms and photoplethysmograms are resilient to noise interference. Furthermore, we identified the optimal size, data type, and format of the input for an effective deep learning analysis. Based on these insights, we were able to construct a lightweight deep learning model that is well‐suited for these analytical tasks. [ABSTRACT FROM AUTHOR]

Few studies have reported on delta checks for tumour markers, even though these markers are often evaluated serially. Therefore, this study aimed to establish a practical delta check limit in different clinical settings for five tumour markers: alpha-fetoprotein, cancer antigen 19-9, cancer antigen 125, carcinoembryonic antigen, and prostate-specific antigen. Pairs of patients' results (current and previous) for five tumour markers between 2020 and 2021 were retrospectively collected from three university hospitals. The data were classified into three subgroups, namely: health check-up recipient (subgroup H), outpatient (subgroup O), and inpatient (subgroup I) clinics. The check limits of delta percent change (DPC), absolute DPC (absDPC), and reference change value (RCV) for each test were determined using the development set (the first 18 months, n=179,929) and then validated and simulated by applying the validation set (the last 6 months, n=66,332). The check limits of DPC and absDPC for most tests varied significantly among the subgroups. Likewise, the proportions of samples requiring further evaluation, calculated by excluding samples with both current and previous results within the reference intervals, were 0.2–2.9% (lower limit of DPC), 0.2–2.7% (upper limit of DPC), 0.3–5.6% (absDPC), and 0.8–35.3% (RCV99.9%). Furthermore, high negative predictive values >0.99 were observed in all subgroups in the in silico simulation. Using real-world data, we found that DPC was the most appropriate delta-check method for tumour markers. Moreover, Delta-check limits for tumour markers should be applied based on clinical settings. [ABSTRACT FROM AUTHOR]

Unlike traditional pulse transit time (PTT), continuous PTT (CPTT) can be used to calculate PTT from all samples within the cardiac cycle. It has the potential to be utilized for continuous blood pressure (BP) estimation. This study evaluated the feasibility of CPTT as a non‐invasive consecutive blood pressure estimation method in 20 volunteers. The CPTT was calculated with a time delay in all discrete samples of photoplethysmograms measured at two different body sites. BP was then calculated with a regression equation. For comparative evaluation, BP based on PTT was also estimated. Continuous blood pressure was measured using a non‐invasive volume clamp BP monitoring device. Four types of BP measurement, systolic BP (SBP), diastolic BP (DBP), mean arterial pressure (MAP), and pulse pressure (PP), were estimated using PTT and CPTT. Correlation coefficients and root‐mean‐squared‐error (RMSE) were used for evaluating BP estimation performance. For estimating SBP, DBP, PP, and MAP, PTT‐based BP estimation showed correlations of.407,.373,.410, and.286, respectively, and CPTT‐based BP estimation showed correlations of.436,.446,.506, and.097, respectively. With PTT‐based estimation, the RMSE between the estimated BP and the baseline BP was 5.44 ± 1.56 mmHg for SBP, 3.14 ± 0.46 mmHg for DBP, 3.66 ± 0.70 mmHg for MAP, and 3.73 ± 1.31 mmHg for PP. The estimated BP using CPTT showed RMSE of 5.36 ± 1.39 mmHg for SBP, 3.02 ± 0.49 mmHg for SBP, 3.44 ± 0.63 mmHg for MAP, and 3.91 ± 1.41 mmHg for PP. We present an innovative application of a non‐invasive blood pressure (BP) estimation method based on continuous pulse transit time (CPTT). Conventional pulse transit time (PTT)‐based BP estimation uses a PTT value per pulse that does not reflect the continuous nature of blood flow. We propose an alternative method to estimate BP using PTT measured continuously and we validated its feasibility. CPTT‐based BP estimation performance was equal to or higher than that of conventional PTT. [ABSTRACT FROM AUTHOR]