• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.921-923
• /
• 2021

최근 반려동물 시장이 커짐으로 인해, 반려동물들의 헬스케어를 위한 제품들이 증가하고 있다. 이에 따라 펫 웨어러블 디바이스를 통한 연구가 활발히 진행되고 있지만, 웨어러블 디바이스를 통해 수집되는 센싱 데이터는 변칙적인 반려동물의 특징 때문에 연구의 한계를 갖는다. 이를 위해 본 논문에서는 1-Dimensional CNN과 LSTM 하이브리드 모델을 기반으로 한 반려동물 행동 분류를 제안한다. 웨어러블 디바이스를 이용해 자이로와 가속도 센서를 수집하여 걸음수를 측정하고, 이후 수집된 센싱 데이터로 반려동물의 행동을 4가지로 분류한다. 행동 분류는 걷기, 뛰기, 앉기, 서기로 분류한다.

• Korean Journal of Remote Sensing
• /
• v.39 no.3
• /
• pp.297-309
• /
• 2023

The degradation of coastal ecosystems and fishery environments is accelerating due to the recent phenomenon of invertebrate grazers. To effectively monitor and implement preventive measures for this phenomenon, the adoption of remote sensing-based monitoring technology for extensive maritime areas is imperative. In this study, we compared and analyzed the robustness of deep learning-based object detection modelsfor detecting and monitoring invertebrate grazersfrom underwater videos. We constructed an image dataset targeting seven representative species of invertebrate grazers in the coastal waters of South Korea and trained deep learning-based object detection models, You Only Look Once (YOLO)v7 and YOLOv8, using this dataset. We evaluated the detection performance and speed of a total of six YOLO models (YOLOv7, YOLOv7x, YOLOv8s, YOLOv8m, YOLOv8l, YOLOv8x) and conducted robustness evaluations considering various image distortions that may occur during underwater filming. The evaluation results showed that the YOLOv8 models demonstrated higher detection speed (approximately 71 to 141 FPS [frame per second]) compared to the number of parameters. In terms of detection performance, the YOLOv8 models (mean average precision [mAP] 0.848 to 0.882) exhibited better performance than the YOLOv7 models (mAP 0.847 to 0.850). Regarding model robustness, it was observed that the YOLOv7 models were more robust to shape distortions, while the YOLOv8 models were relatively more robust to color distortions. Therefore, considering that shape distortions occur less frequently in underwater video recordings while color distortions are more frequent in coastal areas, it can be concluded that utilizing YOLOv8 models is a valid choice for invertebrate grazer detection and monitoring in coastal waters.

• Proceedings of the Korean Radioactive Waste Society Conference
• /
• 2008.11a
• /
• pp.387-389
• /
• 2008

• Proceedings of the Korean Nutrition Society Conference
• /
• 2004.05a
• /
• pp.113-113
• /
• 2004

• 한국생활과학회:학술대회논문집
• /
• 2009.05a
• /
• pp.165-166
• /
• 2009

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2011.06a
• /
• pp.1409-1410
• /
• 2011

• Proceedings of the Materials Research Society of Korea Conference
• /
• 2009.05a
• /
• pp.44.1-44.1
• /
• 2009

최근 경조직 재생 (hard tissue regeneration) 에 대한 연구가 활발히 진행되고 있다. 그러나 이와같은 연구는 결손도입의 어려움 및 이차적인 골절의 위험성 때문에 대형동물을 중심으로 진행되고 있으며, 그 결과 동물실험에 있어서 시간적 경제적으로 큰 리스크를 수반한다. 그러나 유전자 변형동물의 대부분은 마우스이며, 분자생물학적 관점에서 골재생의 과정을 이해하기 위해서는 마우스를 이용한 골재생 모델의 확립이 필요하다. 따라서 본 연구에서는 마우스를 통해 경조직 재생모델의 제작방법을 검토함과 동시에, 골재생부위에 대한 골질 (bone quality) 및 골양 (bone quantity) 평가의 방법을 수립하는 것을 목적으로 하였다. 골결손은 생후 8주의 마우스에 시술하였다. 치과용 드릴을 이용하여 경골 (tibia) 길이의 30 % 부근의 내측(medial) 면에서 골수강 (marrow cavity) 방향으로 $500\;{\mu}m\varphi$의 원주형 결손을 도입하였다. 시술 후의 골재생과정을 관찰하기 위해 ${\mu}CT$ (SMX-100CT: Simazu) 를 이용하여 주기적으로 촬영하였으며, 골양 (BV/TV) 의 회복과정은3D-bon (RATOC) 을 이용하여 정량적인 해석을 수행하였다. 그리고 재생부의 골질 (아파타이트 배행성; BAp orientation) 평가는 투과형micro-beam XRD (R-AXIS BQ: Rigaku)를 이용하여 수행하였다.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.43 no.5
• /
• pp.631-640
• /
• 2014

The inhibitory effects of Boswellia serrata (BW) extracts on degenerative osteoarthritis were investigated in primary-cultured rat cartilage cells and a monosodium-iodoacetate (MIA)-induced osteoarthritis rat model. To identify the protective effects of BW extract against $H_2O_2$ ($800{\mu}M$, 2 hr) in vitro, cell survival was measured by MTT assay. Cell survival after $H_2O_2$ treatment was elevated by BW extract at a concentration of $20{\mu}g/mL$. In addition, BW extract treatment significantly reduced and normalized the productions of pro-inflammatory factors, nuclear transcription factor ${\kappa}B$, cyclooxygenase-2, tumor necrosis factor-${\alpha}$, and interleukin-6 at a concentration of $20{\mu}g/mL$. Treatment of chondrocytes with BW extract significantly reduced 5-lipoxygenase activity and production of prostaglandin E2, especially at a concentration of $10{\sim}20{\mu}g/mL$. For the in vivo animal study, osteoarthritis was induced by intra-articular injection of MIA into knee joints of rats. Consumption of a diet containing BW extract (100 and 200 mg/kg) for 35 days significantly inhibited the development and severity of osteoarthritis in rats. To determine the genetic expression of arthritic factors in articular cartilage, real-time PCR was applied to measure matrix metalloproteinases (MMP-3, MMP-9, and MMP-13), collagen type I, collagen type II, and aggrecan, and BW extract had protective effects at a concentration of 200 mg/kg. In conclusion, BW extract was able to inhibit articular cartilage degeneration by preventing extracellular matrix degradation and chondrocyte injury. One can consider that BW extract may be a potential therapeutic treatment for degenerative osteoarthritis.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.11
• /
• pp.201-208
• /
• 2020

This study examined the anti-obesity effect of krill oil (KO) by regulating adipokines in a high-fat diet (HFD)-induced obese mouse model. The mice were fed a 60 kcal% HFD for 16 weeks, and KO was then administered at an oral dose of 100, 200, and 500 mg/kg/day for four weeks before the end of the experiment. The administration of KO at concentrations of 200 and 500 mg/kg/day decreased body weight gain significantly compared with the HFD-fed control group. In addition, the HFD-fed control group showed the abnormal release of adipokines by an increase in leptin and decrease in adiponectin, compared to the normal diet-fed normal group. On the other hand, KO (500 mg/kg/day)-administered group attenuated the abnormal release of adipokines by the down-regulation of leptin and the up-regulation of adiponectin. Therefore, KO could be a promising therapeutic agent for obesity by the regulation of adipokines.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.1218-1219
• /
• 2023

본 논문은 반려동물의 행동 분석을 개선하기 위해 IMU 센서 데이터와 딥러닝 모델을 결합하는 방법을 제안한다. 이를 위해 IMU 웨어러블 디바이스를 통해 행동 데이터를 수집한다. 수집된 데이터는 총 6개의 클래스로 앉다. 서다. 엎드리다, 먹다, 킁킁대다, 걷다로 분류된다. 분류된 데이터는 클래스별로 데이터 증강 및 전처리 단계를 거친다. 행동 분류를 위해 ResNet과 LSTM을 결합한 하이브리드 모델을 사용하여 학습을 진행했다. ResNet-LSTM은 Accuracy 97%, F1-score 96%로 높은 성능을 보여주었다.