• The Korean Journal of Nuclear Medicine Technology
• /
• v.12 no.3
• /
• pp.214-221
• /
• 2008

Purpose: We make a qualitative analysis of whether Fusion SPECT/CT can find lesion's anatomical sites better than existing SPECT or not, and we want to show the usefulness of SPECT/CT through finding out effects of CT attenuation correction on SPECT images. Materials and Method: 1. The evaluation of fusion images: This study comprised patients who was tested $^{131}I$-MIBG, Bone, $^{111}In$-Octreotide, Meckel's diverticulum, Parathyroid MIBI with Precedence 16 or Symbia T2 from 2008 Jan to Aug. We compared SPECT/CT image with non fusion image and make a qualitative analysis. 2. The evaluation of attenuation correction: We classified 38 patients who was tested 201Tl myocardial exam with Symbia T2 into 5 sections by using Cedars Sinai' QPS program - Ant, Inf, Lat, Septum, Apex. And we showed each section's perfusion states by percentage. We compared the each section's perfusion-states differences between CT AC and Non AC by average${\pm}$standard deviation. Results: 1. The evaluation of fusion images : In high energy $^{131}I$ cases, it was hard to grasp exact anatomical lesions due to difference between regions and surrounding lesions' uptake level. After combining with CT, we could grabs anatomical lesion more exactly. And in meckel's diverticulum case or to find lesions around bowels or organs with $^{111}In$ cases, it demonstrates its superiority. Bone SPECT/CT images help to distinguish between disk spaces certainly and give correct results. 2. The evaluation of attenuation correction: There is no significant difference statistically in Ant and Lat (p>0.05), but there is a meaningful difference in Inferior, Apex and Septum (p<0.05). AC perfusion at inferior wall in the 5 sections of myocardium: The perfusion difference between Non AC perfusion image ($68.58{\pm}7.55$) and CT corrected perfusion image ($76.84{\pm}6.52$) was the largest by $8.26{\pm}4.95$ (p<0.01, t=10.29). Conclusion: Nuclear medicine physicians can identify not only molecular image which shows functional activity of lesions but also anatomical location information of lesions with more accuracy using the combination of SPECT and CT systems. Of course this combination helps nuclear medicine physician find out the abnormal parts. Moreover combined data sets help separate between normal group and abnormal group in complicated body part. So clinicians can carry out diagnosis and treatment planning at the same time with a single test image. In addition, when we examine a myocardium in thorax where attenuation can occur easily, we can trust perfusion more in a certain region in SPECT test because CT provides the capability for accurate attenuation correction. In these reasons, we think we can prove the justice after treatment fusion image.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.5 no.1
• /
• pp.14-22
• /
• 2001

Off-line handwritten numeral recognition is a very difficult task and hard to achieve high recognition results using a single feature and a single classifier, since handwritten numerals contain many pattern variations which mostly depend upon individual writing styles. In this paper, we propose handwritten numeral recognition system using hybrid features and combined classifier. To improve recognition rate, we select mutually helpful features -directional features, crossing point feature and mesh features- and make throe new hybrid feature sets by using these features. These hybrid feature sets hold the local and global characteristics of input numeral images. And we implement combined classifier by combining three neural network classifiers to achieve high recognition rate, where fuzzy integral is used for multiple network fusion. In order to verify the performance of the proposed recognition system, experiments with the unconstrained handwritten numeral database of Concordia University, Canada were performed. As a result, our method has produced 97.85% of the recognition rate.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.1
• /
• pp.50-56
• /
• 2020

With the rapid development of deep neural network theory and application research, the effect of face detection has been improved. However, due to the complexity of deep neural network calculation and the high complexity of the detection environment, how to detect face quickly and accurately becomes the main problem. This paper is based on the relatively simple model of the MTCNN model, using FDDB (Face Detection Dataset and Benchmark Homepage), LFW (Field Label Face) and FaceScrub public datasets as training samples. At the same time of sorting out and introducing MTCNN(Multi-Task Cascaded Convolutional Neural Network) model, it explores how to improve training speed and Increase performance at the same time. In this paper, the dynamic image pyramid technology is used to replace the traditional image pyramid technology to segment samples, and OHEM (the online hard example mine) function in MTCNN model is deleted in training, so as to improve the training speed.

• Journal of KIISE:Software and Applications
• /
• v.29 no.1_2
• /
• pp.1-9
• /
• 2002

Evolutionary artificial neural networks (EANNs) are towards the near optimal ANN using the global search of evolutionary instead of trial-and-error process. However, many real-world problems are too hard to be solved by only one ANN. Recently there has been plenty of interest on combining ANNs in the last generation to improve the performance and reliability. This paper proposes a new approach of constructing multiple ANNs which complement each other by speciation. Also, we develop a multiple ANN to combine the results in abstract, rank, and measurement levels. The experimental results on Australian credit approval data from UCI benchmark data set have shown that combining of the speciated EANNs have better recognition ability than EANNs which are not speciated, and the average error rate of 0.105 proves the superiority of the proposed EANNs.

• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.8
• /
• pp.315-320
• /
• 2014

Composition is a creative activity of a composer in order to express his or her emotion into melody based on their experience. However, it is very hard to implement an automatic composition program whose composition process is the same as the composer. On the basis that the creative activity is possible from the imitation we propose a method to implement an automatic composition system using the learning capability of ANN(Artificial Neural Networks). First, we devise a method to convert a melody into time series that ANN can train and then another method to learn the repeated melody with melody bar for correct training of ANN. After training of the time series to ANN, we feed a new time series into the ANN, then the ANN produces a full new time series which is converted a new melody. But post processing is necessary because the produced melody does not fit to the tempo and harmony of music theory. In this paper, we applied a tempo post processing using tempo post processing program, but the harmony post processing is done by human because it is difficult to implement. We will realize the harmony post processing program as a further work.