• Smart Media Journal
• /
• v.11 no.4
• /
• pp.19-29
• /
• 2022

Recently, with the development of IoT technology, the number of farms using smart farms is increasing. Smart farms monitor the environment and optimise internal environment automatically to improve crop yield and quality. For optimized crop cultivation, researches on predict crop productivity are actively studied, by using collected agricultural digital data. However, most of the existing studies are based on statistical models based on existing statistical data, and thus there is a problem with low prediction accuracy. In this paper, we use various predition models for predicting the production and sales profits, and compare the performance results through models by using the agricultural digital data collected in the facility horticultural smart farm. The models that compared the performance are multiple linear regression, support vector machine, artificial neural network, recurrent neural network, LSTM, and ConvLSTM. As a result of performance comparison, ConvLSTM showed the best performance in R2 value and RMSE value.

• Journal of Korean Society of Forest Science
• /
• v.111 no.2
• /
• pp.263-275
• /
• 2022

We conducted this study to provide essential data for implementing restoration measures on the physical properties of the geology, topography, and soil of the landcreep areas in Yangpyeong-gun, Gyeonggi-do. The strata of the survey area comprised topsoil, weathered soil, weathered rock, and soft rock layers. The landcreep area, caused by colluvial debris, was located in a convex topography shape distributed as bedrock with shales and incorporated with sandstone. According to the measurement of the displacement meter, the surveyed area has crept from 1.1 mm to 6.5 mm during the recurrent landcreep between 1 July and 27 August, 2020. The landcreep had progressed over two directions (S65° W, E45° S, and E70° S) which were similar to the groundwater flow direction (E82.5° S and S16.8° W). The average slope of the landcreep area occurred on a gentle slope (19.3°), lower than the average slope of the mountain area (25°) in Korea. The bulk density in the groundwater areas was lower than that in other surveyed areas.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.11
• /
• pp.491-500
• /
• 2021

In this study, we propose a method for detecting fall behavior using MS Kinect v2 RGBD Camera-based Human-Skeleton Keypoints and a 2-Stacked Bi-LSTM model. In previous studies, skeletal information was extracted from RGB images using a deep learning model such as OpenPose, and then recognition was performed using a recurrent neural network model such as LSTM and GRU. The proposed method receives skeletal information directly from the camera, extracts 2 time-series features of acceleration and distance, and then recognizes the fall behavior using the 2-Stacked Bi-LSTM model. The central joint was obtained for the major skeletons such as the shoulder, spine, and pelvis, and the movement acceleration and distance from the floor were proposed as features of the central joint. The extracted features were compared with models such as Stacked LSTM and Bi-LSTM, and improved detection performance compared to existing studies such as GRU and LSTM was demonstrated through experiments.

• Journal of Genetic Medicine
• /
• v.18 no.2
• /
• pp.110-116
• /
• 2021

Microdeletions of chromosome 22q11.2 are one of the most common microdeletions occurring in humans, and is known to be associated with a wide range of highly variable features. These deletions occur within a cluster of low copy repeats (LCRs) in 22q11.2, referred to as LCR22 A-H. DiGeorge (DGS)/velocardiofacial syndrome is the most prevalent form of a 22q11.2 deletions, caused by mainly proximal deletions between LCR22 A and D. As deletions of distal portion to the DGS deleted regions has been extensively studied, the recurrent distal 22q11.2 microdeletions distinct from DGS has been suggested as several clinical entities according to the various in size and position of the deletions on LCRs. We report a case of long-term follow-up of a female diagnosed with a 22q11.2 distal deletion syndrome, identified a deletion of 1.9 Mb at 22q11.21q11.23 (chr22: 21,798,906-23,653,963) using single nucleotide polymorphism array. This region was categorized as distal deletion type of 22q11.2, involving LCR22 D-F. She was born as a preterm, low birth weight to healthy non-consanguineous Korean parents. She showed developmental delay, growth retardation, dysmorphic facial features, and mild skeletal deformities. The patient underwent a growth hormone administration due to growth impairment without catch-up growth. While a height gain was noted, she had become overweight and was subsequently diagnosed with pre-diabetes. Our case could help broaden the genetic and clinical spectrum of 22q11.2 distal deletions.

• Korean Journal of Head & Neck Oncology
• /
• v.38 no.1
• /
• pp.53-57
• /
• 2022

Reconstruction of subtotal maxillectomy defects with columella deficit is challenging. We report a unique case of facial reconstruction using a free radial forearm flap and a free preauricular flap for the maxillectomy and columella deficit. A 73-year-old woman was diagnosed with recurrent sebaceous carcinoma of the nose. We performed wide excision, including areas of the right cheek, nose, upper lip, maxilla, and columella. The resultant subtotal maxillectomy defect was reconstructed using a three-dimensional flap. First, a free radial forearm flap was transfered to resurface the nasal, oral, and external facial side. Second, a preauricular flap was positioned into the columella defect and anastomosed with the distal portion of the radial forearm flap pedicle. The two flaps survived, and the patient recovered uneventfully. We believe the radial forearm and preauricular double free flaps with the pedicle connection method were effective in reconstructing the present case of subtotal maxillectomy defect.

• Tuberculosis and Respiratory Diseases
• /
• v.85 no.1
• /
• pp.67-73
• /
• 2022

Background: Bronchiectasis patients with neutrophilic airway inflammation develop symptoms of chronic cough, sputum production, and recurrent exacerbations. Roflumilast has anti-inflammatory actions via decreased neutrophilic airway inflammation. The effectiveness of roflumilast to reduce bronchiectasis exacerbation has never been evaluated. Methods: We conducted a double-blinded, randomized, placebo-controlled trial. Our primary objective was to assess the effect of roflumilast compared with that of a placebo in reducing exacerbation rates in bronchiectasis patients. The secondary objectives were the changes in forced expiratory volume in 1 second (FEV₁) and St. George's Respiratory Questionnaire (SGRQ). Bronchiectasis patients older than 18 years who had had two exacerbations during the previous 12 months were randomly assigned to receive either 500 ㎍ of either roflumilast or a placebo once daily for 6 months in a 1:1 ratio. Results: Forty bronchiectasis patients who had experienced exacerbations were screened. Thirty patients completed the study after 6 months of treatment: roflumilast group (n=15) and placebo group (n=15). The rates of exacerbations were 0.57 and 0.59 per patient in the roflumilast and placebo groups, respectively. Prebronchodilator FEV₁ increased by 0.07 L from baseline in the roflumilast group and decreased by 0.015 L in the placebo group, but the difference was not significant. No significant differences were observed in the change of SGRQ scores between the roflumilast and placebo groups. Roflumilast had significant side effects, including loss of appetite and headache. Conclusion: Roflumilast did not significantly affect the rate of exacerbations or quality of life. However, FEV₁ tended to improve more in the roflumilast group than in the placebo group.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.177-183
• /
• 2022

Recently, interesting of home training is getting bigger due to COVID-19. Accordingly, research on applying HAR(human activity recognition) technology to home training has been conducted. However, existing paper of HAR proposed static activity instead of dynamic activity. In this paper, the deep learning model where dynamic exercise posture can be analyzed and the accuracy of the user's exercise posture can be shown is proposed. Fitness images of AI-hub are analyzed by blaze pose. The experiment is compared with three types of deep learning model: RNN(recurrent neural network), LSTM(long short-term memory), CNN(convolution neural network). In simulation results, it was shown that the f1-score of RNN, LSTM and CNN is 0.49, 0.87 and 0.98, respectively. It was confirmed that CNN is more suitable for human activity recognition than other models from simulation results. More exercise postures can be analyzed using a variety learning data.

• Journal of Korean Neurosurgical Society
• /
• v.65 no.6
• /
• pp.772-778
• /
• 2022

Objective : Endovascular treatment of large, wide-necked intracranial aneurysms by coil embolization is often complicated by low rates of complete occlusion and high rates of recurrence. A flow diverter device has been shown to be safe and effective for the treatment of not only large and giant unruptured aneurysms, but small and medium aneurysms. However, in Korea, its use has only recently been approved for aneurysms <10 mm. This study aims to compare the safety and efficacy of flow diversion and coil embolization for the treatment of unruptured aneurysms ≥7 mm. Methods : The participants will include patients aged between 19 and 75 years to be treated for unruptured cerebral aneurysms ≥7 mm for the first time or for recurrent aneurysms after initial endovascular coil embolization. Participants assigned to a flow diversion cohort will be treated using any of the following devices : Pipeline Flex Embolization Device with Shield Technology (Medtronic, Minneapolis, MN, USA), Surpass Evolve (Stryker Neurovascular, Fremont, CA, USA), and FRED or FRED Jr. (MicroVention, Tustin, CA, USA). Participants assigned to a coil embolization cohort will undergo traditional endovascular coiling. The primary endpoint will be complete occlusion confirmed by cerebral angiography at 12 months after treatment. Secondary safety outcomes will evaluate periprocedural and post-procedural complications for up to 12 months. Results : The trial will begin enrollment in 2022, and clinical data will be available after enrollment and follow-up. Conclusion : This article describes the aim and design of a multi-center, randomized, open-label trial to compare the safety and efficacy of flow diversion versus traditional endovascular treatment for unruptured cerebral aneurysms ≥7 mm.

• Smart Media Journal
• /
• v.11 no.3
• /
• pp.26-30
• /
• 2022

Spiking neural network is a neural network that applies the working principle of real brain neurons. Due to the biological mechanism of neurons, it consumes less power for training and reasoning than conventional neural networks. Recently, as deep learning models become huge and operating costs increase exponentially, the spiking neural network is attracting attention as a third-generation neural network that connects convolution neural networks and recurrent neural networks, and related research is being actively conducted. However, in order to apply the spiking neural network model to the industry, a lot of research still needs to be done, and the problem of model retraining to apply a new model must also be solved. In this paper, we propose a method to minimize the cost of model retraining by extracting the weights of the existing trained deep learning model and converting them into the weights of the spiking neural network model. In addition, it was found that weight conversion worked correctly by comparing the results of inference using the converted weights with the results of the existing model.

• Parasites, Hosts and Diseases
• /
• v.60 no.5
• /
• pp.309-315
• /
• 2022

Inflammatory bowel disease (IBD) is a chronic and recurrent illness of the gastrointestinal tract. Treatment of IBD traditionally involves the use of aminosalicylic acid and steroids, while these drugs has been associated with untoward effects and refractoriness. The absence of effective treatment regimen against IBD has led to the exploration of new targets. Parasites are promising as an alternative therapy for IBD. Recent studies have highlighted the use of parasite-derived substances, such as excretory secretory products, extracellular vesicles (EVs), and exosomes, for the treatment of IBD. In this report, we examined whether EVs secreted by Giardia lamblia could prevent colitis in a mouse model. G. lamblia EVs (GlEVs) were prepared from in vitro cultures of Giardia trophozoites. Clinical signs, microscopic colon tissue inflammation, and cytokine expression levels were detected to assess the effect of GlEV treatment on dextran sulfate sodium (DSS)-induced experimental murine colitis. The administration of GlEVs prior to DSS challenge reduced the expression levels of pro-inflammatory cytokines, including tumor necrosis factor alpha, interleukin 1 beta, and interferon gamma. Our results indicate that GlEV can exert preventive effects and possess therapeutic properties against DSS-induced colitis.