• Korean Journal of Head & Neck Oncology
• /
• v.15 no.2
• /
• pp.166-171
• /
• 1999

Background and Objectives: The aim of our study was to describe the appearance of recurrent and residual lesions in the head and neck tumors, and to evaluate the usefullness of CT and MRI. Materials and Methods: CT(n=42) and MRI(n=4) of 44 patients with recurrent head and neck tumors were reviewed retrospectively. Primary tumor sites were larynx/hypopharynx in 15, oral cavity/floor of mouth in 13, base of tongue/tonsil in 5, nasopharynx in 4, palate in 2, and others in 5 patients. Therapeutic modalities included sugery and radiotherapy in 23, radiotherapy in 11, surgery in 5, chemotherapy and radiotherapy in 4, and chemotherapy in 1 patient. Results: The patterns of tumor recurrence were nodal recurrence(n=17), primary tumor bed recurrence combined with nodal recurrence(n=12), primary tumor bed recurrence(n=10) and residual primary tumors(n=5). The most common appearance of residual/recurrent primary tumor on CT was focal or diffuse heterogenous mass with or without surrounding fat or muscle infiltration(25/27). On MRI, the recurrent lesions showed intermediate signal intensity on T1 weighted image and high signal intensity on T2 weighted image with heterogenous enhancement in the most cases(n=3). 38 out of 44 nodal recurrences(86%) which had been pathologically or clinically proved were more than 1 cm in diameter or contained central low density on CT scan. Conclusion: Although CT and MRI findings of recurrent and residual tumors of the head and neck were nonspecific, in the majority the lesions manifested as a mass at primary tumor bed and/or nodal disease including contralateral side of the neck. And CT and MRI are valuable for revealing above lesions.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.3
• /
• pp.11-25
• /
• 2012

We present the method to infer Context-Free Grammars by applying genetic algorithm to the Binary Third-order Recurrent Neural Networks(BTRNN). BTRNN is a multiple-layered architecture of recurrent neural networks, each of which is corresponding to an input symbol, and is combined with external stack. All parameters of BTRNN are represented as binary numbers and each state transition is performed with any stack operation simultaneously. We apply Genetic Algorithm to BTRNN chromosomes and obtain the optimal BTRNN inferring context-free grammar of positive and negative input patterns. This proposed method infers BTRNN, which includes the number of its states equal to or less than those of existing methods of Discrete Recurrent Neural Networks, with less examples and less learning trials. Also BTRNN is superior to the recent method of chromosomes representing grammars at recognition time complexity because of performing deterministic state transitions and stack operations at parsing process. If the number of non-terminals is p, the number of terminals q, the length of an input string k, and the max number of BTRNN states m, the parallel processing time is O(k) and the sequential processing time is O(km).

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.2
• /
• pp.6-10
• /
• 2023

High-potential data can be predicted and stored in the cache to prevent cache misses, thus reducing the processor's request and wait times. As a result, the processor can work non-stop, hiding memory latency. By utilizing the temporal/spatial locality of memory access, the prefetcher introduced to improve the performance of these computers predicts the following memory address will be accessed. We propose a prefetcher that applies the GRU model, which is advantageous for handling time series data. Display the currently accessed address in binary and use it as training data to train the Gated Recurrent Unit model based on the difference (delta) between consecutive memory accesses. Finally, using a GRU model with learned memory access patterns, the proposed data prefetcher predicts the memory address to be accessed next. We have compared the model with the multi-layer perceptron, but our prefetcher showed better results than the Multi-Layer Perceptron.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.11
• /
• pp.1536-1541
• /
• 2018

For effective electricity consumption in urban railway station such as peak load shaving, it is important to know each electrical load pattern by various usage. The total electricity consumption in the urban railway substation is already measured in Korea, but the electricity consumption for each usage is not measured. The author proposed the deep learning method to estimate the electrical load pattern for each usage in the urban railway substation with public data such as weather data. GRU (gated recurrent unit), a variation on the LSTM (long short-term memory), was used, which aims to solve the vanishing gradient problem of standard a RNN (recursive neural networks). The optimal model was found and the estimation results with that were assessed.

• Journal of Korean Neurosurgical Society
• /
• v.45 no.6
• /
• pp.401-404
• /
• 2009

Arteriovenous malformations (AVM) are generally considered to be cured following angiographically proven complete resection. However, rare instances of AVM recurrence have been reported in both children and adults with negative findings on postoperative angiography. The authors present the case of a 12-year-old boy with recurrent AVM. The AVM was originally fed by the pericallosal arteries on both sides, and it showed changing patterns of supply at recurrence. The authors concluded that a negative postoperative angiogram is not necessarily indicative of a cure. Repeat angiography and regular follow-up examinations should be performed to exclude the possibility of recurrence, especially in children.

• Annals of Clinical Neurophysiology
• /
• v.1 no.1
• /
• pp.64-69
• /
• 1999

Background : Syncope was defined as transient loss of consciousness and postural tone. The mechanisms of changes in cerebral hemodynamics during syncope have not been fully evaluated. Transcranial Doppler Ultrasonography can continuously monitor the changes in cerebral hemodynamics during head-up tilt (HUT). TCD could reveal the different patterns of changes in cerebral hemodynamics during syncope. Syncope without hypotension or bradycardia could be detected by TCD. We investigated the changes in cerebral blood flow velocity during HUT using TCD in 33 patients with a history of recurrent syncope or presyncope of unknown origin. Methods & Results : The positive responses were defined as presyncope or syncope with hypotension, bradycardia, or both. During HUT without isoproterenol infusion, there were a $86{\pm}23%$ drop in DV and a $41{\pm}34%$ drop in SV in 5 patients with positive reponses, and mean changes in those were less than 10% in patients with negative reponses (p=.00, p=.00). During HUT with isoproterenol infusion, TCD showed a $80{\pm}18%$ drop in diastolic velocity in 14 patients with positive reponses, and a $47{\pm}10%$ drop in that in patients with negative reponses (p=.00), however the change in systolic velocity did not differ. TCD showed three patterns during positive responses; loss of all flow, loss of end diastolic flow, and a decrease in diastolic velocity. Loss of consciousness occurred in the patients with loss of all flow or end-diastolic flow during positive reponses. Conclusions : TCD shows different patterns of changes in cerebral hemodynamics during HUT. TCD can be used to investigate the pathophysiology of neurocardiogenic syncope.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.4
• /
• pp.203-210
• /
• 2017

In this paper, we propose an effective neural network model for image caption generation and model transfer. This model is a kind of multi-modal recurrent neural network models. It consists of five distinct layers: a convolution neural network layer for extracting visual information from images, an embedding layer for converting each word into a low dimensional feature, a recurrent neural network layer for learning caption sentence structure, and a multi-modal layer for combining visual and language information. In this model, the recurrent neural network layer is constructed by LSTM units, which are well known to be effective for learning and transferring sequence patterns. Moreover, this model has a unique structure in which the output of the convolution neural network layer is linked not only to the input of the initial state of the recurrent neural network layer but also to the input of the multimodal layer, in order to make use of visual information extracted from the image at each recurrent step for generating the corresponding textual caption. Through various comparative experiments using open data sets such as Flickr8k, Flickr30k, and MSCOCO, we demonstrated the proposed multimodal recurrent neural network model has high performance in terms of caption accuracy and model transfer effect.

• Journal of Korean Society on Water Environment
• /
• v.39 no.1
• /
• pp.46-60
• /
• 2023

Climate change causes fluctuations in water quality in the aquatic environment, which can cause changes in water circulation patterns and severe adverse effects on aquatic ecosystems in the future. Therefore, research is needed to predict and respond to water quality changes caused by climate change in advance. In this study, we tried to predict the dissolved oxygen (DO), chlorophyll-a, and turbidity of the Paldang reservoir for about two weeks using long short-term memory (LSTM) and gated recurrent units (GRU), which are deep learning algorithms based on recurrent neural networks. The model was built based on real-time water quality data and meteorological data. The observation period was set from July to September in the summer of 2021 (Period 1) and from March to May in the spring of 2022 (Period 2). We tried to select an algorithm with optimal predictive power for each water quality parameter. In addition, to improve the predictive power of the model, an important variable extraction technique using random forest was used to select only the important variables as input variables. In both Periods 1 and 2, the predictive power after extracting important variables was further improved. Except for DO in Period 2, GRU was selected as the best model in all water quality parameters. This methodology can be useful for preventive water quality management by identifying the variability of water quality in advance and predicting water quality in a short period.

• Research in Mathematical Education
• /
• v.18 no.3
• /
• pp.187-208
• /
• 2014

This study engages in the features of interaction in elementary school mathematics lessons as reflected in the class discourse. 28 pre-service teachers documented the discourse during observation of their tutor-teachers' lessons. Mapping the interaction patterns was performed by a unique graphic model developed for that purpose and enabled providing a spatial picture of the discourse conducted in the lesson. The research findings present the known discourse pattern "initiation-response-evaluation / feedback" (IRE/F) which is recurrent in all the lessons and the teacher's exclusive control over the class discourse patterns. Hence, the remaining time of the lesson for the pupils' discourse is short and meaningless.

• The Journal of the Acoustical Society of Korea
• /
• v.36 no.6
• /
• pp.407-412
• /
• 2017

In this paper, we propose a speech emotion recognition method using a deep neural network based on the attention mechanism. The proposed method consists of a combination of CNN (Convolution Neural Networks), GRU (Gated Recurrent Unit), DNN (Deep Neural Networks) and attention mechanism. The spectrogram of the speech signal contains characteristic patterns according to the emotion. Therefore, we modeled characteristic patterns according to the emotion by applying the tuned Gabor filters as convolutional filter of typical CNN. In addition, we applied the attention mechanism with CNN and FC (Fully-Connected) layer to obtain the attention weight by considering context information of extracted features and used it for emotion recognition. To verify the proposed method, we conducted emotion recognition experiments on six emotions. The experimental results show that the proposed method achieves higher performance in speech emotion recognition than the conventional methods.