Background Early detection of laryngeal masses without periodic visits to hospitals is essential for improving the possibility of full recovery and the long-term survival ratio after prompt treatment, as well as reducing the risk of clinical infection. Results We first propose a convolutional neural network model for automated laryngeal mass detection based on diagnostic images captured at hospitals. Thereafter, we propose a pilot system, composed of an embedded controller, a camera module, and an LCD display, that can be utilized for a home-based self-screening test. In terms of evaluating the model’s performance, the experimental results indicated a final validation loss of 0.9152 and a F1-score of 0.8371 before post-processing. Additionally, the F1-score of the original computer algorithm with respect to 100 randomly selected color-printed test images was 0.8534 after post-processing while that of the embedded pilot system was 0.7672. Conclusions The proposed technique is expected to increase the ratio of early detection of laryngeal masses without the risk of clinical infection spread, which could help improve convenience and ensure safety of individuals, patients, and medical staff.

Purpose This preliminary in-vitro study was designed to evaluate the risk factors of compression injury from use of a circular stapler for end-to-end anastomosis. Methods Transparent collagen plates were prepared in dry and wet conditions. Physical properties of collagen plates and porcine colon tissue were examined using a rheometer. Adjustable and fixed-type circular staplers were applied on the collagen plates and the gap distance and compressive pressure were measured during anvil approximation. Tissue injury was evaluated using a compression injury scale. Compression properties were accessed to optimal or overcompression based on gap distance. Results Unacceptable injuries were rarely observed on the dry collagens, regardless of compression device. In the adjustable compression, the compressibility ratio was similar between dry and wet collagen. Overcompression and unacceptable injury increased on the wet collagens. In the fixed compression, the compressibility ratio increased significantly and unacceptable injuries were observed in more than 50% of wet collagens. Peak pressure was significantly higher in the fixed-compression types than those of adjustable type. On bivariate correlation analysis, fixed-compression type and wet collagens were respectively associated with overcompression. On multivariate analysis, edematous collagen condition was the most important risk factor and proximal anvil side, fixed compression type, and overcompression were also independent risk factors for unacceptable compression injury. Conclusion In the edematous tissue condition, unintentional overcompression could be increased and result in tissue injury on the compression line of the circular stapler.

Intravenous (IV) medication administration processes have been considered as high-risk steps, because accidents during IV administration can lead to serious adverse effects, which can deteriorate the therapeutic effect or threaten the patient’s life. In this study, we propose a multi-modal infusion pump (IP) monitoring technique, which can detect mismatches between the IP setting and actual infusion state and between the IP setting and doctor’s prescription in real time using a thin membrane potentiometer and convolutional-neural-network-based deep learning technique. During performance evaluation, the percentage errors between the reference infusion rate (IR) and average estimated IR were in the range of 0.50–2.55%, while those between the average actual IR and average estimated IR were in the range of 0.22–2.90%. In addition, the training, validation, and test accuracies of the implemented deep learning model after training were 98.3%, 97.7%, and 98.5%, respectively. The training and validation losses were 0.33 and 0.36, respectively. According to these experimental results, the proposed technique could provide improved protection functions to IV-administration patients.

This study was aimed to evaluate the changes of impedance parameters of patients who were admitted to a long-term care hospital by measuring bioelectrical impedance. The subjects were 18 patients who had infusion therapy through peripheral intravenous (IV) catheters and had at least an infiltration. The impedance parameters were measured with a multi-channel impedance measuring instrument (Vector Impedance Meter) twice; at starting IV infusion after catheter insertion and infiltration detected. As results, the resistance (R) after infiltration significantly decreased compared to the initial resistance. At 50 kHz, the resistances were 498.2±79.3 [Ω] before infiltration and 369.4±85.6 [Ω] after infiltration. The magnitude of the reactance (XC) decreased after infiltration. At 50 kHz, the measured reactance was -31.1±8.3 [Ω] before infiltration and -24.5±5.9 [Ω] after infiltration. The data points plotted in the R-XC graph shifted from the first quadrant before infiltration to third quadrant after infiltration. Our findings suggest that bioelectrical impedance is an effective method for detection of infiltration in a noninvasive and quantitative manner.

Medication infusion pumps are the most popular device in almost all areas of a hospital; therefore, it is important to frequently inspect the accuracy of the infusion pump operation to prevent underdose/overdose accidents. However, the conventional infusion pump inspection devices are not suitable for quick and convenient on-site inspection by nurses. In this study, a new IR estimation technique for peristaltic infusion pumps that facilitates on-site pre-screening test with shorter inspection time was proposed. A thin membrane potentiometer was attached to a catheter and the actual IR was estimated based on a time interval between two successive line pushes of an identical cam follower using power function estimation. To evaluate the performance of the proposed IR estimation technique, in vitro experiments were performed using 11 infusion pumps (three for Infusion Pump SET 1 ( IPSET-1) and eight for Infusion Pump SET 2 ( IPSET-2)) with the same model. In experiments, error rate between the actual and the measured values (using conventional inspection device) were 0.04–1.17% range for IPSET-1 and 2.09–4.32% for IPSET-2, and those between the actual and the estimated values (using proposed method) were 0.02–0.62% range for IPSET-1 and 1.31–4.23% for IPSET-2. The proposed technique had almost equivalent performance with a commercial inspection device, but the time for inspection was reduced to almost one third. We expect that the proposed technique can provide a tool for simple and convenient on-site pre-screening of infusion pumps by nurses to improve patient safety.