In this paper, we propose a pre-processing method that converts to Spider Chart image data for classification of gripping movement using EMG (electromyography) sensors and Convolution Neural Networks (CNN) deep learning. First, raw data for six hand gestures are extracted from five test subjects using an 8-channel armband and converted into Spider Chart data of octagonal shapes, which are divided into several sliding windows and are learned. In classifying six hand gestures, the classification performance is compared with the proposed pre-processing method and the existing methods. Deep learning was performed on the dataset by dividing 70% of the total into training, 15% as testing, and 15% as validation. For system performance evaluation, five cross-validations were applied by dividing 80% of the entire dataset by training and 20% by testing. The proposed method generates 97% and 94.54% in cross-validation and general tests, respectively, using the Spider Chart preprocessing, which was better results than the conventional methods.

The purpose of this study is to evaluate the biomechanical properties of fractured adjacent soft tissue during closed reduction after forearm fracture using the finite element method. To accomplish this, a finite element (FE) model of the forearm including soft tissue was constructed, and the material properties reported in previous studies were implemented. Based on this, nine finite element models with different fracture types and fracture positions, which are the main parameters, were subjected to finite element analysis under the same load and boundary conditions. The load condition simulated the traction of increasing the fracture site spacing from 0.4 mm to 1.6 mm at intervals of 0.4 mm at the distal end of the radioulnar bone. Through the finite element analysis, the fracture type, fracture location, and displacement were compared and analyzed for the fracture site spacing of the fractured portion and the maximum equivalent stress of the soft tissues adjacent to the fracture(interosseous membrane, muscle, fat, and skin). The results of this study are as follows. The effect of the major parameters on the fracture site spacing of the fractured part is negligible. Also, from the displacement of 1.2 mm, the maximum equivalent stress of the interosseous membrane and muscle adjacent to the fractured bone exceeds the ultimate tensile strength of the material. In addition, it was confirmed that the maximum equivalent stresses of soft tissues(fat, skin) were different in size but similar in trend. As a result, this study was able to numerically confirm the damage to the adjacent soft tissue due to the fracture site spacing during closed reduction of forearm fracture.

In this study, the optimal driving frequency was derived through finite element analysis (FEA) to optimize the developed piezoelectric ultrasonic medical devices(PUMD) for bone surgery. The core of the PUMD is the piezoelectric ceramic (PZT), which is a vibrator that generates vibration energy. The piezoelectric ceramic shows the maximum current value with respect to the input voltage at the resonance frequency, which generates the maximum mechanical vibration. In the past, various studies have been conducted related to the analysis of PUMD, but most of the research so far has been limited to free vibration analysis. However, in order to derive the accurate resonant frequency, the initial stress generated by bolt tightening in the bolt-clamped Langevin type transducer (BLT) must be considered. In this study, after designing a PUMD, the driving performance according to the bolt tightening value was analyzed through FEA, and this was experimentally verified. First, the resonance mode and frequency response were confirmed through modal and harmonic analysis at 20-40 kHz, which is known as the optimal driving frequency band of PUMD for bone surgery. In addition, the design of the PUMD was confirmed by checking the mechanical behavior of the tip and the piezoelectric ceramic at the resonant frequency. Consequentially, the characteristic evaluation was performed, and it was confirmed that the resonant frequency result derived through the FEA was reasonable. Through this study, we presented a more rational FEA method than before for BLT transducers. We expect that this will shorten the time and cost of developing a PUMD, and will enable the development of more stable and high-quality products.

Early afterdepolarization (EAD), a significant cause of fatal ventricular arrhythmias including Torsade de Pointes (TdP) in long QT syndromes, is a depolarizing afterpotential at the plateau or repolarization phase in action potential (AP) profile early before completing one pace. AP duration prolongation is related to EAD but is not necessarily accounted for EAD. Several computational studies suggested EAD can form from an abnormality in the late plateau and/or repolarization phase of AP shape. In this sense, we hypothesized the slope during repolarization has the characteristics to predict TdP risk, mainly focusing on the maximum slope during repolarization (dVm/dt_{max_repol}). This study aimed to predict the sensitivity of dVm/dt_{max_repol} to ion channel conductances as a TdP risk metric through a population simulation considering multiple effects of simultaneous reduction in six ion channel conductances of g_NaL, g_Kr, g_Ks, g_to, g_K1, and g_CaL. Additionally, we verified the availability of dVm/dt_{max_repol} for TdP risk prediction through the correlation analysis with qNet, the representative TdP metric. We performed the population simulations based on the methodology of Gemmel et al. using the human ventricular myocyte model of Dutta et al. Among the sixion channel conductances, dVm/dt_{max_repol} and qNet responded most sensitively to the change in g_Kr, followed by g_NaL. Furthermore, dVm/dt_{max_repol} showed a statistically significant high negative correlation with qNet. The dVm/dt_{max_repol} values were significantly different according to three TdP risk levels of high, intermediate, and low by qNet (p<0.001). In conclusion, we suggested dVm/dt_{max_repol} as a new biomarker metric for TdP risk assessment.

Medical image segmentation is the most important task in radiation therapy. Especially, when segmenting medical images, the liver is one of the most difficult organs to segment because it has various shapes and is close to other organs. Therefore, automatic segmentation of the liver in computed tomography (CT) images is a difficult task. Since tumors also have low contrast in surrounding tissues, and the shape, location, size, and number of tumors vary from patient to patient, accurate tumor segmentation takes a long time. In this study, we propose a method algorithm for automatically segmenting the liver and tumor for this purpose. As an advantage of setting the boundaries of the tumor, the liver and tumor were automatically segmented from the CT image using the 2D CoordConv DeepLab V3+ model using the CoordConv layer. For tumors, only cropped liver images were used to improve accuracy. Additionally, to increase the segmentation accuracy, augmentation, preprocess, loss function, and hyperparameter were used to find optimal values. We compared the CoordConv DeepLab v3+ model using the CoordConv layer and the DeepLab V3+ model without the CoordConv layer to determine whether they affected the segmentation accuracy. The data sets used included 131 hepatic tumor segmentation (LiTS) challenge data sets (100 train sets, 16 validation sets, and 15 test sets). Additional learned data were tested using 15 clinical data from Seoul St. Mary's Hospital. The evaluation was compared with the study results learned with a two-dimensional deep learning-based model. Dice values without the CoordConv layer achieved 0.965 ± 0.01 for liver segmentation and 0.925 ± 0.04 for tumor segmentation using the LiTS data set. Results from the clinical data set achieved 0.927 ± 0.02 for liver division and 0.903 ± 0.05 for tumor division. The dice values using the CoordConv layer achieved 0.989 ± 0.02 for liver segmentation and 0.937 ± 0.07 for tumor segmentation using the LiTS data set. Results from the clinical data set achieved 0.944 ± 0.02 for liver division and 0.916 ± 0.18 for tumor division. The use of CoordConv layers improves the segmentation accuracy. The highest of the most recently published values were 0.960 and 0.749 for liver and tumor division, respectively. However, better performance was achieved with 0.989 and 0.937 results for liver and tumor, which would have been used with the algorithm proposed in this study. The algorithm proposed in this study can play a useful role in treatment planning by improving contouring accuracy and reducing time when segmentation evaluation of liver and tumor is performed. And accurate identification of liver anatomy in medical imaging applications, such as surgical planning, as well as radiotherapy, which can leverage the findings of this study, can help clinical evaluation of the risks and benefits of liver intervention.

Laryngeal disease harms quality of life, and laryngoscopy is critical in identifying causative lesions. This study extracts and analyzes using radiomics quantitative features from the lesion in laryngoscopy images and will fit and validate a classifier for finding meaningful features. Searching the region of interest for lesions not classified by the YOLOv5 model, features are extracted with radionics. Selected the extracted features are through a combination of three feature selectors, and three estimator models. Through the selected features, trained and verified two classification models, Random Forest and Gradient Boosting, and found meaningful features. The combination of SFS, LASSO, and RF shows the highest performance with an accuracy of 0.90 and AUROC 0.96. Model using features to select by SFM, or RIDGE was low lower performance than other things. Classification of larynx lesions through radiomics looks effective. But it should use various feature selection methods and minimize data loss as losing color data.

The purpose of this study was to develop a hand rehabilitation training system for hemiplegic patients. We also tried to find out five hand postures (WF: Wrist Flexion, WE: Wrist Extension, BG: Ball Grip, HG: Hook Grip, RE: Rest) in real-time using multi-channel EMG-based deep learning. We performed a pre-processing method that converts to Spider Chart image data for the classification of hand movement from five test subjects (total 1,500 data sets) using Convolution Neural Networks (CNN) deep learning with an 8-channel armband. As a result of this study, the recognition accuracy was 92% for WF, 94% for WE, 76% for BG, 82% for HG, and 88% for RE. Also, ten physical therapists participated for the usability evaluation. The questionnaire consisted of 7 items of acceptance, interest, and satisfaction, and the mean and standard deviation were calculated by dividing each into a 5-point scale. As a result, high scores were obtained in immersion and interest in game (4.6±0.43), convenience of the device (4.9±0.30), and satisfaction after treatment (4.1±0.48). On the other hand, Conformity of intention for treatment (3.90±0.49) was relatively low. This is thought to be because the game play may be difficult depending on the degree of spasticity of the hemiplegic patient, and compensation may occur in patient with weakened target muscles. Therefore, it is necessary to develop a rehabilitation program suitable for the degree of disability of the patient.