• ETRI Journal
• /
• 제46권1호
• /
• pp.82-95
• /
• 2024

As automated essay scoring (AES) has progressed from handcrafted techniques to deep learning, holistic scoring capabilities have merged. However, specific trait assessment remains a challenge because of the limited depth of earlier methods in modeling dual assessments for holistic and multi-trait tasks. To overcome this challenge, we explore providing comprehensive feedback while modeling the interconnections between holistic and trait representations. We introduce the DualBERT-Trans-CNN model, which combines transformer-based representations with a novel dual-scale bidirectional encoder representations from transformers (BERT) encoding approach at the document-level. By explicitly leveraging multi-trait representations in a multi-task learning (MTL) framework, our DualBERT-Trans-CNN emphasizes the interrelation between holistic and trait-based score predictions, aiming for improved accuracy. For validation, we conducted extensive tests on the ASAP++ and TOEFL11 datasets. Against models of the same MTL setting, ours showed a 2.0% increase in its holistic score. Additionally, compared with single-task learning (STL) models, ours demonstrated a 3.6% enhancement in average multi-trait performance on the ASAP++ dataset.

• 대한치과기공학회지
• /
• 제37권4호
• /
• pp.213-218
• /
• 2015

Purpose: The purpose of this study compared of reproducibility of prepared tooth impression scanning utilized with white and blue light scanners. Methods: To evaluate reproducibility with white and blue light scanners, the impression of premolar were rotated by $10^{\circ}{\sim}20^{\circ}$ and scanned. These data were compared with the first 3-D data (STL file), and the error sizes were measured (n=5). Independent t test was used to evaluation the reproducibility of impression of premolar with white versus blue light scanners through discrepancies of mean, RMS (${\alpha}=0.05$). Results: Discrepancies of mean with regard to reproducibility were $11.2{\mu}m$, $5.8{\mu}m$, respectively, with white and blue light scanners (p<0.047). And discrepancies of RMS with regard to reproducibility were $33.4{\mu}m$, $18.8{\mu}m$, respectively, with white and blue light scanners (p<0.045). Conclusion: Our results indicate a good reproducibility of prepared tooth impression digitized with blue light scanner more than that with white light scanner.

• 한국의류산업학회지
• /
• 제20권3호
• /
• pp.353-359
• /
• 2018

In this study, fused deposition modellig(FDM) 3D printing technology has been applied directly to polyester voil fabric to produce 3D printed lace/voil composite fabrics. A stereolithograpy(STL) file with a lace type 3D modelling under the various thickness were prepared and transformed into a g-code file using a g-code generator. The extrusion conditions for FDM 3D printing were controlled by 50mm/s of nozzle speed, $235^{\circ}C$ of nozzle temperature, $40^{\circ}C$ of heating bed temperature. 3D printed lace/voil composite fabriscs manufactured by 3D printing based on FDM using a thermoplactic polyurethane(TPU) filaments were obtained. To evaluate the mechanical properties and washability of the fabricated 3D printed lace/voil composite fabric, KES-FB system test, washing fastness test and dry cleaning resistance test were conducted. As 3D printing thickness increased, KOSHI, NUMERI, and FUKURAMI of 3D printed lace/voil composite fabric increased. From the results of the primary hand value test, 3D printed lace/voil composite fabrics were confirmed to be applicable to women's summer garments. As a result of the washability and dry cleaning resistance test of the 3D printed lace/voil composite fabrics, all samples were graded 4-5.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• 제38권
• /
• pp.13.1-13.12
• /
• 2016

Background: Mandibular motion tracking system (ManMoS) has been developed for orthognathic surgery. This article aimed to introduce the ManMoS and to examine the accuracy of this system. Methods: Skeletal and dental models are reconstructed in a virtual space from the DICOM data of three-dimensional computed tomography (3D-CT) recording and the STL data of 3D scanning, respectively. The ManMoS uniquely integrates the virtual dento-skeletal model with the real motion of the dental cast mounted on the simulator, using the reference splint. Positional change of the dental cast is tracked by using the 3D motion tracking equipment and reflects on the jaw position of the virtual model in real time, generating the mixed-reality surgical simulation. ManMoS was applied for two clinical cases having a facial asymmetry. In order to assess the accuracy of the ManMoS, the positional change of the lower dental arch was compared between the virtual and real models. Results: With the measurement data of the real lower dental cast as a reference, measurement error for the whole simulation system was less than 0.32 mm. In ManMoS, the skeletal and dental asymmetries were adequately diagnosed in three dimensions. Jaw repositioning was simulated with priority given to the skeletal correction rather than the occlusal correction. In two cases, facial asymmetry was successfully improved while a normal occlusal relationship was reconstructed. Positional change measured in the virtual model did not differ significantly from that in the real model. Conclusions: It was suggested that the accuracy of the ManMoS was good enough for a clinical use. This surgical simulation system appears to meet clinical demands well and is an important facilitator of communication between orthodontists and surgeons.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• 제25권4호
• /
• pp.295-303
• /
• 1999

Presented in this paper are the experimental results that measure rapid prototyping (RP) errors in 3D medical models. We identified various factors that can cause dimensional errors when producing RP models, specifically in maxillofacial areas. For the experiment, we used a human dry skull. A number of linear measurements based on landmarks were first obtained on the skull. This was followed by CT scanning, 3D model reconstruction, and RP model fabrication. The landmarks were measured again on both the reconstructed models and the physical RP models, and these were compared with those on dry skull. We focused on major sources of errors, such as CT scanning, conversion from CT data to STL models, and RP model fabrication. The results show that the overall error from skull to RP is $0.64{\times}0.36mm(0.71{\times}0.66%)$ in absolute value. This indicates that the RP technology can be acceptable in the real clinical applications. A clinical case that has applied RP models successfully for treatment planning and surgical rehearsal is presented. Although the use of RP models is rare in the medical area yet, we believe RP is promising in that it has a great potential in developing new tools which can aid diagnosis, treatment planning, surgical rehearsal, education, and so on.

• 한국방사선학회논문지
• /
• 제13권7호
• /
• pp.947-953
• /
• 2019

임상에서 사용하는 진단 검사 장치인 전산화 단층촬영기와 자동화된 설계 소프트웨어(MediACE 3D Prog ram), 3D 프린터로 손목 보조기를 제작하고자 하였다. 전산화단층촬영기로 상지의 Dicom 파일을 획득한 후 MediACE 3D Program을 통해 손목 보조기를 디자인하여 "STL(stereolithography)"파일을 만들었고, 디자인된 손목 보조기는 3D 프린터를 이용하여 인쇄하였다. 3D 프린팅 기술로 제작된 손목보조기의 효용성 검증을 위해 뼈와 피부에 가해지는 압력 및 보조기의 스트레스 분포를 유한요소해석으로 나타내었다. 손목 보조기를 제작할 때 유한요소해석의 결과를 가지고 뼈와 피부가 압력에 의한 손상과 보조기의 파손이 자주 일어나는 부위를 보강하여 손목 보조기를 제작할 수 있을 것이라고 기대된다.

• 대한방사선기술학회지:방사선기술과학
• /
• 제43권1호
• /
• pp.9-14
• /
• 2020

Commercial plate bolus is generally used for treatment of surface tumor and required surface dose. We fabricated 3D-printed bolus by using 3D printing technology and usability of 3D-printed bolus was evaluated. RT-structure of contoured plate bolus in the TPS was exported to DICOM files and converted to STL file by using converting program. The 3D-printed bolus was manufactured with rubber-like translucent materials using a 3D printer. The dose distribution calculated in the TPS and compared the characteristics of the plate bolus and the 3D printed bolus. The absolute dose was measured inserting an ion chamber to the depth of 5 cm and 10 cm from the surface of the blue water phantom. HU and ED were measured to compare the material characteristics. 100% dose was distributed at Dmax of 1.5 cm below the surface when was applied without bolus. When the plate bolus and 3D-plate bolus were applied, dose distributed at 0.9 cm and 0.8 cm below the surface of the bolus. After the comparative analysis of the radiation dose at the reference depth, differences in radiation dose of 0.1 ~ 0.3% were found, but there was no difference dose. The usability of the 3D-printed bolus was thus confirmed and it is considered that the 3D-printed bolus can be applied in radiation therapy.

• Journal of Acupuncture Research
• /
• 제24권1호
• /
• pp.29-38
• /
• 2007

Objectives : The objective of this study was to research delayed growth with acupuncture and moxibustion treatment. Methods : We search the oriental medical literature related to delayed growth, especially loose skull, pigeon chest(龜胸), turtle back(龜背), five kinds of flaccidity(五軟), five kinds of retardations(五遲) and infantile malnutrition(疳證). Results : 1. Loose skull is treated with moxibustion therapy of CV8(神厥) and two points(1.5cm upper and under of CV8) 2. Pigeon chest(龜胸) is treated with moxibustion therapy of GB38(外丘), ST18(乳根) and 6 points around of STl7(乳中). 3. Turtle back(龜背) is treated with moxibustion therapy of BLl3(肺兪), BL15(心兪) and BLl7(膈兪) 4. Acupuncture therapy of five kinds of flaccidity(五軟) was rare. but there are one case to stimulate Hwatahyeopcheok point using plum-blossom needle. 5. Five kinds of retardations(五遲) is treated with moxibustion therapy of BLl5(心兪) and two point of medial malleolus 6. Infantile malnutrition(疳證) is treated with acupuncture therapy of the spleen channel and stomach. channel, therapy using three-edged needle, cutting therapy(LUlO(魚際) and Sabong), moxibustion therapy(LRl3(章門) and BL2l(胃兪), and Ch'una therapy. Conclusion: We expect that acupuncture and moxibustion treatment of delayed growth will be applied practically in clinical medicine due to further study on delayed growth.

• The Journal of Advanced Prosthodontics
• /
• 제10권4호
• /
• pp.321-327
• /
• 2018

PURPOSE. The purpose of this study was to compare the cutting method and the lamination method to investigate whether the CAD data of the proposed inlay shape are machined correctly. MATERIALS AND METHODS. The Mesial-Occlusal shape of the inlay was modeled by changing the stereolithography (STL). Each group used SLS (metal powder) or SLA (photocurable resin) in the additive method, and wax or zirconia in the subtractive method (n=10 per group, total n=40). Three-dimensional (3D) analysis program (Geomagic Control X inspection software; 3D systems) was used for the alignment and analysis. The root mean square (RMS) in the 2D plane state was measured within $50{\mu}m$ radius of eight comparison measuring points (CMP). Differences were analyzed using one-way analysis of variance and post-hoc Tukey's test were used (${\alpha}=.05$). RESULTS. There was a significant difference in RMS only in SLA and SLS of 2D section (P<.05). In CMP mean, CMP 4 ($-5.3{\pm}46.7{\mu}m$) had a value closest to 0, while CMP 6 ($20.1{\pm}42.4{\mu}m$) and CMP 1 ($-89.2{\pm}61.4{\mu}m$) had the greatest positive value and the greatest negative value, respectively. CONCLUSION. Since the errors obtained from the study do not exceed the clinically acceptable values, the lamination method and the cutting method can be used clinically.

• 비파괴검사학회지
• /
• 제29권1호
• /
• pp.36-42
• /
• 2009

역공학의 한 방법으로 엑스선 전산 단층촬영 장치에서 획득한 삼차원 데이터를 CAD(computeraided design) 데이터로 변환하여 쾌속조형 모델을 제작할 수 있고, 또한 삼차원 가시화를 통해 피검사체의 오차를 확인하고 분석할 수 있는 알고리즘들에 대한 연구를 수행하였다. 연구 결과를 바탕으로 GUI(graphical user interface) 기반의 소프트웨어를 개발하였다. 개발한 소프트웨어의 검증을 위하여 산업 및 의료용 샘플들에 대한 전산 단층촬영 스캐닝을 수행하고 CAD 데이터 변환 및 컴퓨터 원용 결함진단 시뮬레이션을 수행하였다. 이를 통해 컴퓨터 원용 결함진단의 산업 및 의료분야 적용 가능성을 확인하였다.