• Journal of Biomedical Engineering Research
• /
• v.19 no.5
• /
• pp.469-476
• /
• 1998

In the case of acquisition and displaying medical Images such as ultrasonography and endoscopy on VGA monitor of PC system, image degradation of tear-drop appears through scan conversion. In this study, we compare several methods which can solve this degradation and implement the hardware system that resolves this problem in real-time with PC. It is possible to represent high quality image display and real-time processing and acquisition with specific de-interlacing device and PCI bridge on our hardware system. Image quality is improved remarkably on our hardware system. It is implemented as PC-based system, so acquiring, saving images and describing text comment on those images and PACS networking can be easily implemented.metabolism. All images were spatially normalized to MNI standard PET template and smoothed with 16mm FWHM Gaussian kernel using SPM96. Mean count in cerebral region was normalized. The VOls for 34 cerebral regions were previously defined on the standard template and 17 different counts of mirrored regions to hemispheric midline were extracted from spatially normalized images. A three-layer feed-forward error back-propagation neural network classifier with 7 input nodes and 3 output nodes was used. The network was trained to interpret metabolic patterns and produce identical diagnoses with those of expert viewers. The performance of the neural network was optimized by testing with 5~40 nodes in hidden layer. Randomly selected 40 images from each group were used to train the network and the remainders were used to test the learned network. The optimized neural network gave a maximum agreement rate of 80.3% with expert viewers. It used 20 hidden nodes and was trained for 1508 epochs. Also, neural network gave agreement rates of 75~80% with 10 or 30 nodes in hidden layer. We conclude that artificial neural network performed as well as human experts and could be potentially useful as clinical decision support tool for the localization of epileptogenic zones.

• Radiation Oncology Journal
• /
• v.21 no.1
• /
• pp.100-106
• /
• 2003

Purpose : To develop a patients' setup verification tool (PSVT) to verify the alignment of the machine and the target isocenters, and the reproduclbility of patients' setup for three dimensional conformal radiotherapy (3DCRT) and intensity modulated radiotherapy (IMRT). The utilization of this system is evaluated through phantom and patient case studies. Materials and methods : We developed and clinically tested a new method for patients' setup verification, using digitally reconstructed radiography (DRR), simulation, porial and digital images. The PSVT system was networked to a Pentium PC for the transmission of the acquired images to the PC for analysis. To verify the alignment of the machine and target isocenters, orthogonal pairs of simulation images were used as verification images. Errors in the isocenter alignment were measured by comparing the verification images with DRR of CT Images. Orthogonal films were taken of all the patients once a week. These verification films were compared with the DRR were used for the treatment setup. By performing this procedure every treatment, using humanoid phantom and patient cases, the errors of localization can be analyzed, with adjustments made from the translation. The reproducibility of the patients' setup was verified using portal and digital images. Results : The PSVT system was developed to verify the alignment of the machine and the target isocenters, and the reproducibility of the patients' setup for 3DCRT and IMRT. The results show that the localization errors are 0.8$\pm$0.2 mm (AP) and 1.0$\pm$0.3 mm (Lateral) in the cases relating to the brain and 1.1$\pm$0.5 mm (AP) and 1.0$\pm$0.6 mm (Lateral) in the cases relating to the pelvis. The reproducibility of the patients' setup was verified by visualization, using real-time image acquisition, leading to the practical utilization of our software Conclusions : A PSVT system was developed for the verification of the alignment between machine and the target isocenters, and the reproduclbility of the patients' setup in 3DCRT and IMRT. With adjustment of the completed GUI-based algorithm, and a good quality DRR image, our software may be used for clinical applications.

• Nuclear Engineering and Technology
• /
• v.52 no.10
• /
• pp.2250-2261
• /
• 2020

Accurate localization of radioactive materials is crucial in homeland security and radiological emergencies. Coded-aperture gamma camera is an interesting solution for such applications and can be developed into portable real-time imaging devices. However, traditional reconstruction methods cannot effectively deal with signal-independent noise, thereby hindering low-noise real-time imaging. In this study, a novel reconstruction method with excellent noise-suppression capability based on a multi-layer perceptron (MLP) is proposed. A coded-aperture gamma camera based on pixel detector and coded-aperture mask was constructed, and the process of radioactive source imaging was simulated. Results showed that the MLP method performs better in noise suppression than the traditional correlation analysis method. When the Co-57 source with an activity of 1 MBq was at 289 different positions within the field of view which correspond to 289 different pixels in the reconstructed image, the average contrast-to-noise ratio (CNR) obtained by the MLP method was 21.82, whereas that obtained by the correlation analysis method was 5.85. The variance in CNR of the MLP method is larger than that of correlation analysis, which means the MLP method has some instability in certain conditions.

• Archives of Plastic Surgery
• /
• v.51 no.1
• /
• pp.30-35
• /
• 2024

Background Breast aesthetics evaluation often relies on subjective assessments, leading to the need for objective, automated tools. We developed the Seoul Breast Esthetic Scoring Tool (S-BEST), a photometric analysis software that utilizes a DenseNet-264 deep learning model to automatically evaluate breast landmarks and asymmetry indices. Methods S-BEST was trained on a dataset of frontal breast photographs annotated with 30 specific landmarks, divided into an 80-20 training-validation split. The software requires the distances of sternal notch to nipple or nipple-to-nipple as input and performs image preprocessing steps, including ratio correction and 8-bit normalization. Breast asymmetry indices and centimeter-based measurements are provided as the output. The accuracy of S-BEST was validated using a paired t-test and Bland-Altman plots, comparing its measurements to those obtained from physical examinations of 100 females diagnosed with breast cancer. Results S-BEST demonstrated high accuracy in automatic landmark localization, with most distances showing no statistically significant difference compared with physical measurements. However, the nipple to inframammary fold distance showed a significant bias, with a coefficient of determination ranging from 0.3787 to 0.4234 for the left and right sides, respectively. Conclusion S-BEST provides a fast, reliable, and automated approach for breast aesthetic evaluation based on 2D frontal photographs. While limited by its inability to capture volumetric attributes or multiple viewpoints, it serves as an accessible tool for both clinical and research applications.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.11
• /
• pp.439-448
• /
• 2021

Driver negligence accounts for the largest proportion of the causes of traffic accidents, and research to detect them is continuously being conducted. This paper proposes a method to accurately detect a distracted driver and localize the most characteristic parts of the driver. The proposed method hierarchically constructs a CNN basic model that classifies 10 classes based on CAM in order to detect driver distration and 4 subclass models for detailed classification of classes having a confusing or common feature area in this model. The classification result output from each model can be considered as a new feature indicating the degree of matching with the CNN feature maps, and the accuracy of classification is improved by horizontally combining and learning them. In addition, by combining the heat map results reflecting the classification results of the basic and detailed classification models, the characteristic areas of attention in the image are found. The proposed method obtained an accuracy of 95.14% in an experiment using the State Farm data set, which is 2.94% higher than the 92.2%, which is the highest accuracy among the results using this data set. Also, it was confirmed by the experiment that more meaningful and accurate attention areas were found than the results of the attention area found when only the basic model was used.

• The Southeast Asian review
• /
• v.20 no.3
• /
• pp.263-297
• /
• 2010

The relics of the Southeast Asian civilizations in the first phase are found with the relics from India, China, and even further West of Persia and Rome. These relics are the historic marks of the ancient interactions of various continents, mainly through the maritime trade. The traces of the indic culture, which appears in the historic age, are represented in the textual records and arts, regarded as the essence of the India itself. The ancient Hindu arts found in various locations of Southeast Asia were thought to be transplanted directly from India. However, Neither did the Gupta Hindu Art of India form the mainstream of the Gupta Art, nor did it play an influential role in the adjacent areas. The Indian culture was transmitted to Southeast Asia rather intermittently than consistently. If we thoroughly compare the early Hindu art of India and that of Southeast Asia, we can find that the latter was influenced by the former, but still sustained Southeast Asian originality. The reason that the earliest Southeast Asian Hindu art is discovered mostly in continental Southeast Asia is resulted from the fact that the earliest networks between India and the region were constructed in this region. Among the images of Hindu gods produced before the 7th century are Shiva, Vishnu, Harihara, and Skanda(the son of Shiva), and Ganesha(the god of wealth). The earliest example of Vishnu was sculpted according to the Kushan style. After that, most of the sculptures came to have robust figures and graceful proportions. There are a small number of images of Ganesha and Skanda. These images strictly follow the iconography of the Indian sculpture. This shows that Southeast Asians chose their own Hindu gods from the Hindu pantheon selectively and devoted their faiths to them. Their basic iconography obediently followed the Indian model, but they tried to transform parts of the images within the Southeast Asian contexts. However, it is very difficult to understand the process of the development of the Hindu faith and its contents in the ancient Southeast Asia. It is because there are very few undamaged Hindu temples left in Southeast Asia. It is also difficult to make sure that the Hindu religion of India, which was based on the complex rituals and the caste system, was transplanted to Southeast Asia, because there were no such strong basis of social structure and religion in the region. "Indianization" is an organized expansion of the Indian culture based on the sense of belonging to an Indian context. This can be defined through the process of transmission and progress of the Hindu or Buddhist religions, legends about purana, and the influx of various epic expression and its development. Such conditions are represented through the Sanskrit language and the art. It is the element of the Indian culture to fabricate an image of god as a devotional object. However, if we look into details of the iconography, style, and religious culture, these can be understood as a "selective reception of foreign religious culture." There were no sophisticated social structure yet to support the Indian culture to continue in Southeast Asia around the 7th century. Whether this phenomena was an "Indianization" or the "influx of elements of Indian culture," it was closely related to the matter of 'localization.' The regional character of each local region in Southeast Asia is partially shown after the 8th century. However it is not clear whether this culture was settled in each region as its dominant culture. The localization of the Indian culture in Southeast Asia which acted as a network connecting ports or cities was a part of the process of localization of Indian culture in pan-Southeast Asian region, and the process of the building of the basis for establishing an identity for each Southeast Asian region.

• Nuclear Medicine and Molecular Imaging
• /
• v.42 no.2
• /
• pp.153-163
• /
• 2008

Medical imaging modalities to image either anatomical structure or functional processes have developed along somewhat independent paths. Functional images with single photon emission computed tomography (SPECT) and positron emission tomography (PET) are playing an increasingly important role in the diagnosis and staging of malignant disease, image-guided therapy planning, and treatment monitoring. SPECT and PET complement the more conventional anatomic imaging modalities of computed tomography (CT) and magnetic resonance (MR) imaging. When the functional imaging modality was combined with the anatomic imaging modality, the multimodality can help both identify and localize functional abnormalities. Combining PET with a high-resolution anatomical imaging modality such as CT can resolve the localization issue as long as the images from the two modalities are accurately coregistered. Software-based registration techniques have difficulty accounting for differences in patient positioning and involuntary movement of internal organs, often necessitating labor-intensive nonlinear mapping that may not converge to a satisfactory result. These challenges have recently been addressed by the introduction of the combined PET/CT scanner and SPECT/CT scanner, a hardware-oriented approach to image fusion. Combined PET/CT and SPECT/CT devices are playing an increasingly important role in the diagnosis and staging of human disease. The paper will review the development of multi modality instrumentations for clinical use from conception to present-day technology and the application software.

• Journal of Internet Computing and Services
• /
• v.23 no.3
• /
• pp.35-45
• /
• 2022

Even in recent years, treatment of first-aid patients is still often delayed due to a shortage of medical resources in marginalized areas. Research on automating the analysis of medical data to solve the problems of inaccessibility for medical services and shortage of medical personnel is ongoing. Computer vision-based medical inspection automation requires a lot of cost in data collection and labeling for training purposes. These problems stand out in the works of classifying lesion that are rare, or pathological features and pathogenesis that are difficult to clearly define visually. Anomaly detection is attracting as a method that can significantly reduce the cost of data collection by adopting an unsupervised learning strategy. In this paper, we propose methods for detecting abnormal images on chest X-RAY images as follows based on existing anomaly detection techniques. (1) Normalize the brightness range of medical images resampled as optimal resolution. (2) Some feature vectors with high representative power are selected in set of patch features extracted as intermediate-level from lesion-free images. (3) Measure the difference from the feature vectors of lesion-free data selected based on the nearest neighbor search algorithm. The proposed system can simultaneously perform anomaly classification and localization for each image. In this paper, the anomaly detection performance of the proposed system for chest X-RAY images of PA projection is measured and presented by detailed conditions. We demonstrate effect of anomaly detection for medical images by showing 0.705 classification AUROC for random subset extracted from the PadChest dataset. The proposed system can be usefully used to improve the clinical diagnosis workflow of medical institutions, and can effectively support early diagnosis in medically poor area.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.5
• /
• pp.18-24
• /
• 2010

In this paper, a new matrix decoding system using vector based Virtual Source Location Information (VSLI) is proposed as an alternative to the conventional Dolby Pro logic II/IIx system for reconstructing multi-channel output signals from matrix encoded two channel signals, Lt/Rt. This new matrix decoding system is composed of passive decoding part and active part. The passive part makes crude multi-channel signals using linear combination of the two encoded signals(Lt/Rt) and the active part enhances each channel regarding to the virtual source which is emergent in each inter channel. Since the virtual sources are related to the perceptual sound images in virtual sound field, the reconstructed multi-channel sound results in good dynamic perception and stable image localization. Moreover, the good channel separation is maintained with nonlinear trigonometric enhancing function.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.2 s.302
• /
• pp.121-130
• /
• 2005

In this paper, we propose a feature extraction method for iris recognition using wavelet transforms. The wavelet transform is fast and has a good localization characteristic. In particular, the low frequency band can be used as an effective feature vector. In iris recognition, the noise caused by eyelid the eyebrow, glint, etc may be included in iris. The iris pattern is distorted by noises by itself, and a feature extraction algorithm based on filter such as Wavelets, Gabor transform spreads noises into whole iris region. Namely, such noises degrade the performance of iris recognition systems a major problem. This kind of noise has adverse effect on performance. In order to solve these problems, we propose to divide the iris image into a number of sub-region and apply the wavelet transform to each sub-region. Experimental results show that the performance of proposed method is comparable to existing methods using Gabor transform and region division noticeably improves recognition performance. However, it is noted that the processing time of the wavelet transform is much faster than that of the existing methods.