Brain tumors or gliomas are fatal cancer species with high recurrence rates due to their strong invasiveness. Therefore, the goal of surgery is complete tumor resection. However, the surgery is difficult to distinguish the border because tumors and blood vessels have the same color tone and shape. The fluorescein sodium is used as a fluorescence contrast agent for boundary separation. When the external light source is irradiated, yellow fluorescence is expressed in the tumor, which helps distinguish between blood vessels and tumor boundaries. But, the fluorescence expression of fluorescence sodium depends on the concentration of fluorescein sodium and such analytical data is insufficient. The unclear fluorescence can obscure the boundaries between blood vessels and tumors. In addition, reduce the efficiency of fluorescence sodium use. This paper proposes a protocol of concentration range for fluorescence expression conditions. Fluorescent expression was observed using a near-infrared (NIR) color camera with corresponding dilution using normal saline in 1 ml microtube. The flunoresence emission density range is 1.00 mM to 0.15 mM. The fluorescence emission begin to 1.00 mM and the 0.15 mM discolor. The discolor is difficult to fluorescence emission condition obserbation. Thus, the maximum density range of the bright fluoresecein is 0.15 mM to 0.30 mM. When the concentration range of fluorescein sodium is analyzed based on the gradient of fluorescence expression and the power measurement, the brightest fluorescence is expected to facilitate the complete resection of the tumor. For the concentration range protocol, setting concentration ranges and analyzing fluorescence expression image according to saturation and brightness to find optimal fluorescence concentration are important. Concentration range protocols for fluorescence expression conditions can be used to find optimal concentrations of substances whose expression pattern varies with concentration ranges. This study is expected to be helpful in the boundary classification and resection of brain tumors and glioma.

In large-scale hospitals, the department of biomedical engineering should always provide quick repair service for damaged medical devices to guarantee continuous patient treatment. However, in actual circumstances, there are so many time-consuming issues that delays device repair for weeks or even months; therefore, it is required to prepare alternative ways for quick repair service. In this study, we first mentioned about the regulation issues in Korea about the 3D printing-based medical device repair, and then introduced the results of our preliminary study that evaluated the feasibility of 3D printing-based medical device repair before real-field application. Results of the study demonstrated that, in all of the 23 cases, parts for repair that were manufactured by 3D-printing were successfully fixed and connected to the main body of the original device, and showed sufficient rigidity for protecting internal parts of the device. Considering the experimental results, medical device repair by applying 3D printing technology can be a promising alternative in cases when regular repair process is not available or takes too much time.

Despite of a lot of studies about proprioception tests, there are little study results to investigate the relationship between the functional movement and proprioception level. In this study, we tried to perform quantitative analysis for the effect of ankle joint proprioception level on the one leg standing postural control ability. Nine healthy people volunteered for this study. Force and position aspects of proprioception were evaluated using the electromyography system (EMG) and mobile clinometer application, respectively. The center of pressure (COP) trajectories, measured by a pressure mat sensor, were used for quantitative analysis of balance for each subject. We computed indices and errors of force and position aspects of proprioception from the EMG and ankle angle. Mean velocity of total and anterior-posterior direction (Vm and Vm_ap), root mean squared distance in anterior-posterior direction (RDap), travel length (L), and area (A) of COP trajectories were also calculated as indices of postural control ability of subjects. Two aspects of proprioception showed the low correlation from each other as previous studies. However, the EMG error of gastrocnemius lateral activation showed a high correlation coefficient with COP variables such as Vm (ρ=0.817, p=0.007), Vm_ap (ρ=0.883, p=0.002), RDap (ρ=0.854, p=0.003), L (ρ=0.817, 0.007) and A (ρ=0.700, p=0.036). Within our knowledge, this is almost the first study that investigated the relationship between proprioception level and functional movement. These study results could support that the ankle joint proprioception facilitation exercise would have positive effects on functional balance rehabilitation interventions.

The purpose of this study was to present the models for classifying the wrist X-ray images by types and for segmenting the radius automatically in each image using deep learning and to verify the learned models. The data were a total of 904 wrist X-rays with the distal radius fracture, consisting of 472 anteroposterior (AP) and 432 lateral images. The learning model was the ResNet50 model for AP/lateral image classification, and the U-Net model for segmentation of the radius. In the model for AP/lateral image classification, 100.0% was showed in precision, recall, and F1 score and area under curve (AUC) was 1.0. The model for segmentation of the radius showed an accuracy of 99.46%, a sensitivity of 89.68%, a specificity of 99.72%, and a Dice similarity coefficient of 90.05% in AP images and an accuracy of 99.37%, a sensitivity of 88.65%, a specificity of 99.69%, and a Dice similarity coefficient of 86.05% in lateral images. The model for AP/lateral classification and the segmentation model of the radius learned through deep learning showed favorable performances to expect clinical application.

Although benign gastric ulcers do not develop into gastric cancer, they are similar to early gastric cancer and difficult to distinguish. This may lead to misconsider early gastric cancer as gastric ulcer while diagnosing. Since gastric cancer does not have any special symptoms until discovered, it is important to detect gastric ulcers by early gastroscopy to prevent the gastric cancer. Therefore, we developed a Convolution Neural Network (CNN) model that can be helpful for endoscopy. 3,015 images of gastroscopy of patients undergoing endoscopy at Gachon University Gil Hospital were used in this study. Using ResNet-50, three models were developed to classify normal and gastric ulcers, normal and gastric cancer, and gastric ulcer and gastric cancer. We applied the data augmentation technique to increase the number of training data and examined the effect on accuracy by varying the multiples. The accuracy of each model with the highest performance are as follows. The accuracy of normal and gastric ulcer classification model was 95.11% when the data were increased 15 times, the accuracy of normal and gastric cancer classification model was 98.28% when 15 times increased likewise, and 5 times increased data in gastric ulcer and gastric cancer classification model yielded 87.89%. We will collect additional specific shape of gastric ulcer and cancer data and will apply various image processing techniques for visual enhancement. Models that classify normal and lesion, which showed relatively high accuracy, will be re-learned through optimal parameter search.

The deep neural networks (DNN) that can replicate the behavior of the human expert who recognizes the characteristics of ECG waveform have been developed and studied to analyze ECG. However, although the existing DNNs can not provide the explanations for their decisions, those trials have attempted to determine whether patients have certain diseases or not and those decisions could not be accepted because of the absence of relating theoretical basis. In addition, these DNNs required a lot of training data to obtain sufficient accuracy in spite of the difficulty in the acquisition of relating clinical data. In this study, a small-sized continuous data processing DNN (C-DNN) was suggested to determine the simple characteristics of ECG wave that were not required additional explanations about its decisions and the C-DNN can be easily trained with small training data. Although it can analyze small input data that was selected in narrow region on whole ECG, it can continuously scan all ECG data and find important points such as start and end points of P, QRS and T waves within a short time. The star and end points of ECG waves determined by the C-DNNs were compared with the results performed by human experts to estimate the accuracies of the C-DNNs. The C-DNN has 150 inputs, 51 outputs, two hidden layers and one output layer. To find the start and end points, two C-DNNs were trained through deep learning technology and applied to a parameter acquisition algorithms. 12 lead ECG data measured in four patients and obtained through PhysioNet was processed to make training data by human experts. The accuracy of the C-DNNs were evaluated with extra data that were not used at deep learning by comparing the results between C-DNNs and human experts. The averages of the time differences between the C-DNNs and experts were 0.1 msec and 13.5 msec respectively and those standard deviations were 17.6 msec and 15.7 msec. The final step combining the results of C-DNN through the waveforms of 12 leads was successfully determined all 33 waves without error that the time differences of human experts decision were over 20 msec. The reliable decision of the ECG wave's start and end points benefits the acquisition of accurate ECG parameters such as the wave lengths, amplitudes and intervals of P, QRS and T waves.