Seok-Jo Yang

Background: A dataset[1,2] of electroencephalography(EEG) of frontotemporal dementia(FTD), alzheimer`s disease(AD) patients and healthy control(HC) were classified using convolutional neural network(CNN) and evaluated its performancesMethod: EEG dataset containing 88 subjects is downsampled and sliced into 10 seconds. Total dataset was divided into train, test and validation sets for each subject. CNN structure was used to extract temporal and spatial features from EEG and classify patients into 3 groups(FTD/AD/HC). Cross entropy loss was calculated to train the model using adam optimizer. Classification performances were measured to validate model such as accuracy, confusion matrix, AUC(area under curve), F1 Score.Result: After training, validation accuracy was 95.35% and macro‐averaged AUC was 0.9962. F1 score was 0.9469 and the minimum validation accuracy measured among subjects was 57.14% in FTD group. For AD group average accuracy was 92.79%, and for FTD group average accuracy was 97.52%. Average accuracy of healthy control group was 95.26%.Conclusion: We implemented CNN model that can classify EEG of dementia patients with very high performances without expert‐designed feature engineering process. These results imply that deep learning based approach is most promising for EEG‐based patient classification.

Background: Femoral varus deformities complicating the realignment of the quadriceps muscles are frequently associated with medial patellar luxation (MPL) in dogs. Therefore, distal femoral osteotomy (DFO) is recommended in dogs affected with severe MPL and a distal femoral varus deformity. The presence of an anatomic lateral distal femoral angle (aLDFA) of ≥ 102° has been anecdotally recommended as an indication for performing corrective DFO in large-breed dogs. However, the effect of a femoral varus deformity on MPL has not been scientifically evaluated. We aimed to evaluate the influence of a femoral varus deformity on MPL using a finite element method based computer model. Three-dimensionally reconstructed computed tomographic images of a normal femur from a Beagle dog were deformed using meshing software to create distal varus deformities. A total of thirteen aLDFAs, including 95°, 98° and 100°–110°, were simulated. The patellar positions and reaction force between the patella and trochlear grooves were calculated for all finite element models under constant rectus femoris muscle activation. Results: The patella was displaced medially from the trochlear groove at an aLDFA of ≥103°. With an aLDFA of 103° to 110°, the reaction force was equal to zero and then decreased to negative values during the simulation, while other models with aLDFAs of 95°, 98°, and 100°-102° had positive reaction force values. The patella began to luxate at 24.90 seconds (sec) with an aLDFA of 103°, 19.80 sec with an aLDFA of 104°, 21.40 sec with an aLDFA of 105°, 20.10 sec with an aLDFA of 106°, 18.60 sec with an aLDFA of 107°, 15.30 sec with an aLDFA of 108°, 16.60 sec with an aLDFA of 109°, and 11.90 sec with an aLDFA of 110°. Conclusion: Severe distal femoral varus with an aLDFA of ≥103° caused MPL when other anatomical factors were controlled. Thissimplified computer model provides complementary information to anecdotal cutoffs for DFO, hence it should be applied to clinical patients with caution.

Magnesium (Mg) ions are the most abundant intracellular divalent cations and play a pivotal role in numerous cellular processes. Biodegradable Mg-containin

Normally, Maglev (magnetic levitation) vehicles run on elevated guideways. The elevated guideway must satisfy various load conditions of the vehicle, and has to be designed to ensure ride quality, while ensuring that the levitation stability of the vehicle is not affected by the deflection of the guideway. However, because the elevated guideways of Maglev vehicles in South Korea and other countries fabricated so far have been based on over-conservative design criteria, the size of the structures has increased. Further, from the cost perspective, they are unfavourable when compared with other light rail transits such as monorail, rubber wheel, and steel wheel automatic guided transit. Therefore, a slender guideway that does have an adverse effect on the levitation stability of the vehicle is required through optimisation of design criteria. In this study, to predict the effect of various design parameters of the guideway on the dynamic behaviour of the vehicle, simulations were carried out using a dynamics model similar to the actual vehicle and guideway, and a limiting value of deflection ratio of the slender guideway to ensure levitation control is proposed. A guideway that meets the requirement as per the proposed limit for deflection ratio was designed and fabricated, and through a driving test of the vehicle, the validity of the slender guideway was verified. From the results, it was confirmed that although some increase in airgap and cabin acceleration was observed with the proposed slender guideway when compared with the conventional guideway, there was no notable adverse effect on the levitation stability and ride quality of the vehicle. Therefore, it can be inferred that the results of this study will become the basis for establishing design criteria for slender guideways of Maglev vehicles in future.

Significance: In the past decade, countless studies have been performed to control the mechanical and corrosion property of magnesium-based alloy, which degrades in the physiological environment, to overcome the flaws of the inert implant materials and shift the paradigm of conventional bone fixation devices. Controlled degradation of Mg-5wt%Ca-1wt%Zn alloy results in the formation of biomimicking calcification matrix at the degrading interface to initiate the bone formation process. This process facilitates early bone healing and allows the complete replacement of biodegradable Mg implant by the new bone within 1 y of implantation, as demonstrated in 53 cases of successful long-term clinical study.

자기부상철도에서 전체 투자비의 약 70%를 차지하는 가이드웨이의 건설비를 줄이고 미관을 향상시키기 위하여 가이드웨이의 경량화가 요구된다. 그에 따라서 가이드웨이의 변형이 증가하며 그 변형에 의한 부상안정성과 안내성능 영향을 분석할 필요가 있다. 본 논문에서는 가이드웨이의 구조물의 횡변형과 레일의 수평불규칙도에 의한 차량의 안내 특성에의 영향성 분석에 목적을 두고 있다. 이를 위하여 가이드웨이 횡변형을 포함하는 3 차원 모델의 이용이 제안된다. 제안된 가이드웨이와 차량이 통합된 동적 해석 모델을 이용하여 가이드웨이 횡변형에 의한 차량의 안내특성 분석을 수행하였다. 해석 결과 가이드웨이 거더의 횡변형에 의한 효과는 미비하지만 가이드웨이 레일의 횡방향 불규칙도에 대해서는 속도에 따라 영향을 받는 것을 알 수 있었다. 따라서 타당한 레일의 횡 불규칙도 설계 공차의 제한이 필요함을 알 수 있다.

The scaffold plays an important role in bone regeneration, a research field highly involved with artificial substitutes, as it acts as substrate for cell adhesion, proliferation and regeneration of new tissue. Notably, choosing appropriate materials to design a well-functioned composite scaffold that is biocompatible with human tissue is highly critical. Considering that magnesium is one of the most excellent biomaterials for orthopedic applications, as well as having easy process-ability while combining with polycaprolactone (PCL) to achieve composite scaffold, a biodegradable MgCl₂/PCL (1:5) composite scaffold was controllably fabricated through solid freeform fabrication (SFF) technique for bone regeneration applications, in which PCL serves as base matrix and micro-scaled particles of MgCl₂ as filler. Due to the addition of MgCl₂, composite scaffold not only exhibits improved hydrophilicity, but also renders itself more compressive in a mechanical manner with enhanced compressive strength and modulus compared to pure PCL scaffold. Most importantly, the in-vitro cell culture study showed that cells in the synthesized scaffolds proliferated and differentiated faster than those in the PCL scaffolds. Consequently, it is believed that the MgCl₂/PCL (1:5) composite scaffold is a more effective and suitable substrate for cell culture.