Thermodiluted relative tidal volume estimation using a thermal camera in operating room under spinal anesthesia.

BACKGROUND: Estimating relative tidal volume is an important factor when monitoring breathing status. The relationship between temperature and respiration volume has rarely been studied. In this paper, a formula was derived for calculating thermodiluted respiration volume from temperature changes in the nasal cavity. To evaluate the proposed formula, the study compared the relative tidal volume estimated by the proposed formula with that recorded by a respiration volume monitor (Exspiron1Xi, RVM). Thermal data were obtained for 8 cases at a rate of 10 measurements per second. Simultaneous recordings by the RVM are regarded as the reference. RESULTS: The mean of ICC coefficient is 0.948 ± 0.030, RMSE is 0.1026 ± 0.0284, R-squared value is 0.8962 ± 0.065 and linear regression coefficient [Formula: see text] is 0.966 ± 0.104, [Formula: see text] is 0.042 ± 0.057. Bland-Altman plot showed 96.01% of samples that the difference between the measured and estimated values exists within 2 standard deviations. CONCLUSIONS: In this paper, a model that can thermodynamically calculate the relationship between thermal energy and respiration volume is proposed. The thermodiluted model is a feasible method for estimating relative respiration tidal volumes.

Multi-slice representational learning of convolutional neural network for Alzheimer's disease classification using positron emission tomography.

BACKGROUND: Alzheimer's Disease (AD) is a degenerative brain disorder that often occurs in people over 65 years old. As advanced AD is difficult to manage, accurate diagnosis of the disorder is critical. Previous studies have revealed effective deep learning methods of classification. However, deep learning methods require a large number of image datasets. Moreover, medical images are affected by various environmental factors. In the current study, we propose a deep learning-based method for diagnosis of Alzheimer's disease (AD) that is less sensitive to different datasets for external validation, based upon F-18 fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT). RESULTS: The accuracy, sensitivity, and specificity of our proposed network were 86.09%, 80.00%, and 92.96% (respectively) using our dataset, and 91.02%, 87.93%, and 93.57% (respectively) using the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. We observed that our model classified AD and normal cognitive (NC) cases based on the posterior cingulate cortex (PCC), where pathological changes occur in AD. The performance of the GAP layer was considered statistically significant compared to the fully connected layer in both datasets for accuracy, sensitivity, and specificity (p < 0.01). In addition, performance comparison between the ADNI dataset and our dataset showed no statistically significant differences in accuracy, sensitivity, and specificity (p > 0.05). CONCLUSIONS: The proposed model demonstrated the effectiveness of AD classification using the GAP layer. Our model learned the AD features from PCC in both the ADNI and Severance datasets, which can be seen in the heatmap. Furthermore, we showed that there were no significant differences in performance using statistical analysis.

Recognition of Negative Emotion using Long Short-Term Memory with Bio-Signal Feature Compression.

Negative emotion is one reason why stress causes negative feedback. Therefore, many studies are being done to recognize negative emotions. However, emotion is difficult to classify because it is subjective and difficult to quantify. Moreover, emotion changes over time and is affected by mood. Therefore, we measured electrocardiogram (ECG), skin temperature (ST), and galvanic skin response (GSR) to detect objective indicators. We also compressed the features associated with emotion using a stacked auto-encoder (SAE). Finally, the compressed features and time information were used in training through long short-term memory (LSTM). As a result, the proposed LSTM used with the feature compression model showed the highest accuracy (99.4%) for recognizing negative emotions. The results of the suggested model were 11.3% higher than with a neural network (NN) and 5.6% higher than with SAE.

Radar-Based Detection of Respiration Rate with Adaptive Harmonic Quefrency Selection.

Continuous respiration monitoring is important for predicting a potential disease. Due to respiration measurements using contact sensors, it is difficult to achieve continuous measurement because the sensors are inconvenient to attach. In this study, a radar sensor was used for non-contact respiration measurements. The radar sensor had a high precision and could even be used in the dark. It could also be used continuously regardless of time and place. The radar sensor relied on the periodicity of respiration to detect the respiration rate. A respiration adaptive interval was set and the respiration rate was detected through harmonic quefrency selection. As a result, it was confirmed that there was no difference between the respiratory rate measured using a respiration belt and the respiratory rate detected using a radar sensor. Furthermore, case studies on changes in the radar position and about measurement for long periods confirmed that the radar sensor could detect respiration rate continuously regardless of the position and measurement duration.

Semantic Segmentation of White Matter in FDG-PET Using Generative Adversarial Network.

In the diagnosis of neurodegenerative disorders, F-18 fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) is used for its ability to detect functional changes at early stages of disease process. However, anatomical information from another modality (CT or MRI) is still needed to properly interpret and localize the radiotracer uptake due to its low spatial resolution. Lack of structural information limits segmentation and accurate quantification of the 18F-FDG PET/CT. The correct segmentation of the brain compartment in 18F-FDG PET/CT will enable the quantitative analysis of the 18F-FDG PET/CT scan alone. In this paper, we propose a method to segment white matter in 18F-FDG PET/CT images using generative adversarial network (GAN). The segmentation result of GAN model was evaluated using evaluation parameters such as dice, AUC-PR, precision, and recall. It was also compared with other deep learning methods. As a result, the proposed method achieves superior segmentation accuracy and reliability compared with other deep learning methods.

Reuniting overnutrition and undernutrition, macronutrients, and micronutrients.

Over-nutrition and its late consequences are a dominant theme in medicine today. In addition to the health hazards brought on by over-nutrition, the medical community has recently accumulated a roster of health benefits with obesity, grouped under "obesity paradox." Throughout the world and throughout history until the 20th century, under-nutrition was a dominant evolutionary force. Under-nutrition brings with it a mix of benefits and detriments that are opposite to and continuous with those of over-nutrition. This continuum yields J-shaped or U-shaped curves relating body mass index to mortality. The overweight have an elevated risk of dying in middle age of degenerative diseases while the underweight are at increased risk of premature death from infectious conditions. Micronutrient deficiencies, major concerns of nutritional science in the 20th century, are being neglected. This "hidden hunger" is now surprisingly prevalent in all weight groups, even among the overweight. Because micronutrient replacement is safe, inexpensive, and predictably effective, it is now an exceptionally attractive target for therapy across the spectrum of weight and age. Nutrition-related conditions worthy of special attention from caregivers include excess vitamin A, excess vitamin D, and deficiency of magnesium.

A Deep Neural Network-Based Pain Classifier Using a Photoplethysmography Signal.

Side effects occur when excessive or low doses of analgesics are administered compared to the required amount to mediate the pain induced during surgery. It is important to accurately assess the pain level of the patient during surgery. We proposed a pain classifier based on a deep belief network (DBN) using photoplethysmography (PPG). Our DBN learned about a complex nonlinear relationship between extracted PPG features and pain status based on the numeric rating scale (NRS). A bagging ensemble model was used to improve classification performance. The DBN classifier showed better classification results than multilayer perceptron neural network (MLPNN) and support vector machine (SVM) models. In addition, the classification performance was improved when the selective bagging model was applied compared with the use of each single model classifier. The pain classifier based on DBN using a selective bagging model can be helpful in developing a pain classification system.

Design of User-Customized Negative Emotion Classifier Based on Feature Selection Using Physiological Signal Sensors.

First, the Likert scale and self-assessment manikin are used to provide emotion analogies, but they have limits for reflecting subjective factors. To solve this problem, we use physiological signals that show objective responses from cognitive status. The physiological signals used are electrocardiogram, skin temperature, and electrodermal activity (EDA). Second, the degree of emotion felt, and the related physiological signals, vary according to the individual. KLD calculates the difference in probability distribution shape patterns between two classes. Therefore, it is possible to analyze the relationship between physiological signals and emotion. As the result, features from EDA are important for distinguishing negative emotion in all subjects. In addition, the proposed feature selection algorithm showed an average accuracy of 92.5% and made it possible to improve the accuracy of negative emotion recognition.

Exploring for the optimal structural design for the 3D-printing technology for cranial reconstruction: a biomechanical and histological study comparison of solid vs. porous structure.

PURPOSE: Cranioplasty for recovering skull defects carries the risk for a number of complications. Various materials are used, including autologous bone graft, metallic materials, and non-metallic materials, each of which has advantages and disadvantages. If the use of autologous bone is not feasible, those artificial materials also have constraints in the case of complex anatomy and/or irregular defects. MATERIAL AND METHODS: This study used metal 3D-printing technology to overcome these existing drawbacks and analyze the clinical and mechanical performance requirements. To find an optimal structure that satisfied the structural and mechanical stability requirements, we evaluated biomechanical stability using finite element analysis (FEA) and mechanical testing. To ensure clinical applicability, the model was subjected to histological evaluation. Each specimen was implanted in the femur of a rabbit and was evaluated using histological measurements and push-out test. RESULTS AND CONCLUSION: We believe that our data will provide the basis for future applications of a variety of unit structures and further clinical trials and research, as well as the direction for the study of other patient-specific implants.

Breathing-Associated Facial Region Segmentation for Thermal Camera-Based Indirect Breathing Monitoring.

Breathing can be measured in a non-contact method using a thermal camera. The objective of this study investigates non-contact breathing measurements using thermal cameras, which have previously been limited to measuring the nostril only from the front where it is clearly visible. The previous method is challenging to use for other angles and frontal views, where the nostril is not well-represented. In this paper, we defined a new region called the breathing-associated-facial-region (BAFR) that reflects the physiological characteristics of breathing, and extract breathing signals from views of 45 and 90 degrees, including the frontal view where the nostril is not clearly visible. Experiments were conducted on fifteen healthy subjects in different views, including frontal with and without nostril, 45-degree, and 90-degree views. A thermal camera (A655sc model, FLIR systems) was used for non-contact measurement, and biopac (MP150, Biopac-systems-Inc) was used as a chest breathing reference. The results showed that the proposed algorithm could extract stable breathing signals at various angles and views, achieving an average breathing cycle accuracy of 90.9% when applied compared to 65.6% without proposed algorithm. The average correlation value increases from 0.587 to 0.885. The proposed algorithm can be monitored in a variety of environments and extract the BAFR at diverse angles and views.

A mobile tele-radiology imaging system with JPEG2000 for an emergency care.

The aim of this study was to design a tele-radiology imaging system for rapid emergency care via mobile networks and to assess the diagnostic feasibility of the Joint Photographic Experts Group 2000 (JPEG2000) radiological imaging using portable devices. Rapid patient information and image exchange is helpful to make clinical decisions. We assessed the usefulness of the mobile tele-radiology system by measuring both a quantitative method, PNSR calculation, for image qualities, and its transmission time via mobile networks in different mobile networks, respectively; code division multiple access evolution-data optimized, wireless broadband, and high-speed downlink packet access; and the feasibility of the JPEG2000 computed tomography (CT) images by qualitatively assessing with the Alberta stroke program early CT score method with 12 CT image cases (seven normal and five abnormal cases). We found that the quality of the JPEG2000 radiological images was satisfied quantitatively and was judged as acceptable qualitatively at 5:1 and 10:1 compression levels for the mobile tele-radiology imaging system. The JPEG2000-format radiological images achieved a fast transmission while maintaining a diagnosis quality on a portable device via mobile networks. Unfortunately, a PDA device, having a limited screen resolution, posed difficulties in reviewing the JPEG2000 images regardless of the compression levels. An ultra mobile PC was preferable to study the medical image. The mobile tele-radiology imaging systems supporting JPEG2000 image transmission can be applied to actual emergency care services under mobile computing environments.

Respiration Rate Estimation Based on Independent Component Analysis of Accelerometer Data: Pilot Single-Arm Intervention Study.

BACKGROUND: As the mobile environment has developed recently, there have been studies on continuous respiration monitoring. However, it is not easy for general users to access the sensors typically used to measure respiration. There is also random noise caused by various environmental variables when respiration is measured using noncontact methods in a mobile environment. OBJECTIVE: In this study, we aimed to estimate the respiration rate using an accelerometer sensor in a smartphone. METHODS: First, data were acquired from an accelerometer sensor by a smartphone, which can easily be accessed by the general public. Second, an independent component was extracted to calibrate the three-axis accelerometer. Lastly, the respiration rate was estimated using quefrency selection reflecting the harmonic component because respiration has regular patterns. RESULTS: From April 2018, we enrolled 30 male participants. When the independent component and quefrency selection were used to estimate the respiration rate, the correlation with respiration acquired from a chest belt was 0.7. The statistical results of the Wilcoxon signed-rank test were used to determine whether the differences in the respiration counts acquired from the chest belt and from the accelerometer sensor were significant. The P value of the difference in the respiration counts acquired from the two sensors was .27, which was not significant. This indicates that the number of respiration counts measured using the accelerometer sensor was not different from that measured using the chest belt. The Bland-Altman results indicated that the mean difference was 0.43, with less than one breath per minute, and that the respiration rate was at the 95% limits of agreement. CONCLUSIONS: There was no relevant difference in the respiration rate measured using a chest belt and that measured using an accelerometer sensor. The accelerometer sensor approach could solve the problems related to the inconvenience of chest belt attachment and the settings. It could be used to detect sleep apnea through constant respiration rate estimation in an internet-of-things environment.

Frequency decomposition and phase synchronization of the visual evoked potential using the empirical mode decomposition.

Objectives The phase characteristics of the representative frequency components of the Electroencephalogram (EEG) can be a means of understanding the brain functions of human senses and perception. In this paper, we found out that visual evoked potential (VEP) is composed of the dominant multi-band component signals of the EEG through the experiment. Methods We analyzed the characteristics of VEP based on the theory that brain evoked potentials can be decomposed into phase synchronized signals. In order to decompose the EEG signal into across each frequency component signals, we extracted the signals in the time-frequency domain with high resolution using the empirical mode decomposition method. We applied the Hilbert transform (HT) to extract the signal and synthesized it into a frequency band signal representing VEP components. VEP could be decomposed into phase synchronized δ, θ, α, and ß frequency signals. We investigated the features of visual brain function by analyzing the amplitude and latency of the decomposed signals in phase synchronized with the VEP and the phase-locking value (PLV) between brain regions. Results In response to visual stimulation, PLV values were higher in the posterior lobe region than in the anterior lobe. In the occipital region, the PLV value of theta band was observed high. Conclusions The VEP signals decomposed into constituent frequency components through phase analysis can be used as a method of analyzing the relationship between activated signals and brain function related to visual stimuli.

Synthesis of Electrocardiogram V-Lead Signals From Limb-Lead Measurement Using R-Peak Aligned Generative Adversarial Network.

Recently, portable electrocardiogram (ECG) hardware devices have been developed using limb-lead measurements. However, portable ECGs provide insufficient ECG information because of limitations in the number of leads and measurement positions. Therefore, in this study, V-lead ECG signals were synthesized from limb leads using an R-peak aligned generative adversarial network (GAN). The data used the Physikalisch-Technische Bundesanstalt (PTB) dataset provided by PhysioNet. First, R-peak alignment was performed to maintain the physiological information of the ECG. Second, time domain ECG was converted to bi-dimensional space by ordered time-sequence embedding. Finally, the GAN was learned through the pairs between the modified limb II (MLII) lead and each chest (V) lead. The result showed that the mean structural similarity index (SSIM) was 0.92, and the mean error rate of the percent mean square difference (PRD) of the chest leads was 7.21%.

Level-Set Segmentation-Based Respiratory Volume Estimation Using a Depth Camera.

In this paper, a method is proposed to measure human respiratory volume using a depth camera. The level-set segmentation method, combined with spatial and temporal information, was used to measure respiratory volume accurately. The shape of the human chest wall was used as spatial information. As temporal information, the segmentation result from the previous frame in the time-aligned depth image was used. The results of the proposed method were verified using a ventilator. The proposed method was also compared with other level-set methods. The result showed that the mean tidal volume error of the proposed method was 8.41% compared to the actual tidal volume. This was calculated to have less error than with two other methods: the level-set method with spatial information (14.34%) and the level-set method with temporal information (10.93%). The difference between these methods of tidal volume error was statistically significant [Formula: see text]. The intra-class correlation coefficient (ICC) of the respiratory volume waveform measured by a ventilator and by the proposed method was 0.893 on an average, while the ICC between the ventilator and the other methods were 0.837 and 0.879 on an average.

The effectiveness of cardiopulmonary resuscitation instruction: animation versus dispatcher through a cellular phone.

INTRODUCTION: We developed a cardiopulmonary resuscitation (CPR) instruction programme using motion capture animation integrated into cellular phones. We compared the effectiveness of animation-assisted CPR instruction with dispatcher-assisted instruction in participants with no previous CPR training. METHODS: This study was a single blind cluster randomized trial. Participants were allocated to either animation-assisted CPR (AA-CPR; 8 clusters, 44 participants) group or dispatcher-assisted CPR (DA-CPR; 8 clusters, 41 participants). The overall performance and time of each step of CPR cycle were recorded on a checklist by 3 assessors. The objective performances were evaluated using the Resusci Anne SkillReporter Manikin. Differences between the groups were compared using an independent t-test adjusted for the effect of clustering. RESULTS: The AA-CPR group had a significantly better checklist score (p<0.001) and time to completion of 1 CPR cycle (p<0.001) than the DA-CPR group. In an objective assessment of psychomotor skill, the AA-CPR group demonstrated more accurate hand positioning (68.8+/-3.6%, p=0.033) and compression rate (72.4+/-3.7%, p=0.015) than DA-CPR group. However, the accuracy of compression depth (p=0.400), ventilation volume (p=0.977) and flow rate (p=0.627) were below 30% in both groups. CONCLUSION: Audiovisual animated CPR instruction through a cellular phone resulted in better scores in checklist assessment and time interval compliance in participants without CPR skill compared to those who received CPR instructions from a dispatcher; however, the accuracy of important psychomotor skill measures was unsatisfactory in both groups.

Delay-bounded semi-reliable vital sign transmission protocol for mobile telemedicine over a CDMA 1x EV-DO network.

The reliable and instant transmission of vital signs is important for remote time-critical patient care through a telemedicine system. However, sometimes the reliability and instantaneity conditions cannot be satisfied simultaneously under a high-noise mobile network, because they are reciprocal to each other. In this paper, the vital sign transmission protocol (VSTP) running over a CDMA 1x EVDO (Code Division Multiple Access 1x Evolution Data Only) mobile network is proposed to comply with both the reliability and instantaneity requirements. The switching buffer management scheme is combined with a hybrid error control scheme, consisting of forward error correction (FEC) and automatic repeat request (ARQ). The CDMA 1x EVDO mobile network is modeled by two states using the Markov wireless channel model to test transmission performance under diverse network conditions. Throughout the noisy environment simulation, the performance of the VSTP is compared with the Transmission Control Protocol (TCP) and User Datagram Protocol (UDP) to demonstrate its efficacy over error-prone mobile network.

Web-based secure access from multiple patient repositories.

BACKGROUND: Internet-based health-record management requires not only the provision of strong data protection to prevent privacy intrusion and unauthorized access, but also the introduction of a common healthcare-record format to allow cooperation using heterogeneous repositories held at various hospitals. METHODS: A secure multi-agent architecture is proposed for accessing healthcare information through the Internet from multiple heterogeneous repositories. The proposed system is organized into a four-tier architecture that consists of client applications, a central access-control system, local access-control systems, and hospital information systems. The eXtensible Markup Language (XML) and the role-based access-control (RBAC) system are combined for efficient repository management by providing methods for access-control, information exchange, user authentication, data integrity, and selective encryption. RESULT: A multi-agent architecture using XML and RBAC can interconnect heterogeneous repositories with different formats and different hospital policies, and allow them to communicate securely. The authorized client, having confirmed access privileges, can retrieve the requested healthcare data in an XML-based common data format with embedded confidentiality. CONCLUSION: The proposed method for Internet-based exchange of patient data is particularly useful for cooperative healthcare and the creation of lifetime healthcare records.

Implementation of a large-scale hospital information infrastructure for multi-unit health-care services.

With the increase in demand for high quality medical services, the need for an innovative hospital information system has become essential. An improved system has been implemented in all hospital units of the Yonsei University Health System. Interoperability between multi-units required appropriate hardware infrastructure and software architecture. This large-scale hospital information system encompassed PACS (Picture Archiving and Communications Systems), EMR (Electronic Medical Records) and ERP (Enterprise Resource Planning). It involved two tertiary hospitals and 50 community hospitals. The monthly data production rate by the integrated hospital information system is about 1.8 TByte and the total quantity of data produced so far is about 60 TByte. Large scale information exchange and sharing will be particularly useful for telemedicine applications.

Motion artifact reduction in photoplethysmography using independent component analysis.

Removing the motion artifacts from measured photoplethysmography (PPG) signals is one of the important issues to be tackled for the accurate measurement of arterial oxygen saturation during movement. In this paper, the motion artifacts were reduced by exploiting the quasi-periodicity of the PPG signal and the independence between the PPG and the motion artifact signals. The combination of independent component analysis and block interleaving with low-pass filtering can reduce the motion artifacts under the condition of general dual-wavelength measurement. Experiments with synthetic and real data were performed to demonstrate the efficacy of the proposed algorithm.