• 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 Biomedical Engineering Research
• /
• v.19 no.3
• /
• pp.305-312
• /
• 1998

A small PACS based on PC is designed for CR. To receive the digital image from CR, a DICOM Interface Unit (DIU) is designed that complied with the medical image standard, DICOM V3.0. The CR images acquired through the DIU are stored in a file-server; the patient information of the images is stored in a database. To improve the performance of PC and to use it easily, multiresolution images are constructed by wavelet transform and displayed progressively. Wavelet compression method is newly adopted to store the images hierarchically to storage units. In this compression method, the image is decomposed into subclasses of image by wavelet transform, and then the subclasses of the image are vector quantized using a multiresolution codebook. The storage units for CR images were divided into the short-term storage in file-server and the harddisk in viewing station. Image processing tools supported by general PACS is implemented based on PC.

• Journal of the Korean Society of Radiology
• /
• v.13 no.5
• /
• pp.757-763
• /
• 2019

Recently, Following the recent development of flat panel detector with wide dynamic ranges, increasing numbers of healthcare providers have begun to use digital radiography. As a result, filter thickness standards should be reestablished, as current clinical practice requires the use of thicknesses recommended by the National Council on Radiation Protection and Measurements, which are based on information, acquired using conventional analog systems. Here we investigated the possibility of minimizing dose creep and optimizing patient dose using Al filters in digital radiography. The use of thicker Al filters resulted in a maximum 19.3% reduction in the entrance skin exposure dose when medical images with similar sharpness values were compared. However, resolution, which is a critical factor in imaging, had a significant change of 1.01 lp/mm. This change in resolution is thought to be due to the increased amount of scattered rays generated from the object due to the X-ray beam hardening effect. The increase in the number of scattered rays was verified using the scattering degradation factor. However, the FPD, which has recently been developed and is widely used in various areas, has greater response to radiation than analog devices and has a wide dynamic range. Therefore, the FPD is expected to maintain an appropriate level of resolution corresponding to the increase in the scattered-ray content ratio, which depends on filter thickness. Use of the FPD is also expected to minimize dose creep by reducing the exposure dose.

• The Journal of the Korea Contents Association
• /
• v.10 no.7
• /
• pp.253-258
• /
• 2010

The image intensifier tube that is clinically used gets changed depending on its terms of use. As the brightness gets deteriorated and the performance gets decrease, the X-ray output needs to be increased to maintain the identical diagnosis outcome. As a consequence, the patients are exposed to the increased radiation. Since not many accurate studies have been made, this study measured the conversion factor and resolution depending on the use of the devices that were currently used in Korea. When comparison was made with a device with less than 1 year-use, the conversion factor of a device with 6 year-use got decreased up to 25% while its resolution was reduced from 3.4 lp/mm to 2.5 lp/mm. Such outcome highlights the necessary measures to be supplemented as the increased terms of use implied also the increased radiation exposure among the patients as well as difficulty to conduct an accurate exam.

• Journal of Biomedical Engineering Research
• /
• v.23 no.3
• /
• pp.243-251
• /
• 2002

Synthetic zinc wave employs Pulsed plane wave as transmit beam with linear time delay curve. The received echoes in different transmit directions at different transmit times are superposed at imaging Points with Proper time delay compensation using synthetic focusing scheme. This scheme. which uses full aperture in transmit, obtains a high SNR image, and also features high lateral resolution by using two way dynamic focusing at all imaging depths. In this Paper, we consider the Problems in realization of synthetic zinc wave. Also. we have applied the scheme to obtain phantom and in-vivo images using a linear array of 5 MHz. In phantom test. experimental images show high resolution over a more extended imaging depth than conventional fixed Point transmit and receive dynamic focusing schemes In-vivo images show that the resolution could not overcome conventional focusing systems because of motion blurring and(or) aberration of tissue. but the frame rate tan be increased by a factor of more than 5 compared to conventional focusing schemes. with competitive resolution at all imaging depths .

• Phonetics and Speech Sciences
• /
• v.12 no.4
• /
• pp.81-90
• /
• 2020

This paper describes a method of building Korean conversational speech data in the emergency medical domain and proposes an annotation method for the collected data in order to improve speech recognition performance. To suggest future research directions, baseline speech recognition experiments were conducted by using partial data that were collected and annotated. All voices were recorded at 16-bit resolution at 16 kHz sampling rate. A total of 166 conversations were collected, amounting to 8 hours and 35 minutes. Various information was manually transcribed such as orthography, pronunciation, dialect, noise, and medical information using Praat. Baseline speech recognition experiments were used to depict problems related to speech recognition in the emergency medical domain. The Korean conversational speech data presented in this paper are first-stage data in the emergency medical domain and are expected to be used as training data for developing conversational systems for emergency medical applications.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.510-512
• /
• 2018

The proposed method enhances the resolution of images by removing noise using wavelet transform to remove noise from images generated by ultrasound diagnosis. We propose an algorithm to reduce the speckle noise and enhance the edge of the ultrasound image. The proposed algorithm can enhance edges of various sizes by using wavelet transform which can use both frequency and spatial information. Experimental results show that the performance of the algorithm for noise reduction of ultrasound images is about 0.45ms for $520{\times}440$ images.

• The Journal of the Korea Contents Association
• /
• v.2 no.4
• /
• pp.23-29
• /
• 2002

This paper suggest the image compression that is needed to process PACS in medical information system. The image decoding method is used Linear-predictor and Lloyd-Max quantizer(quantization) in the Wavelet transform domain. Wavelet Transform Method is processed the multi-resolution by dividing image into 10 sub-bands of 3 levels. Low frequency domain that is sensitive to human visual characteristic is encoded by DPCM which is lossless encoding methods, and Lloyed-Max quantizer, the optimal quantizer for reducing ringing and aliasing in the image of inter sub-band, is used in the remaining high frequency domain of sub-band. The examination verifies that decompressed images are superior by the result that PSNR is 28.53dB on the input image, 512$\times$152 abdominal CT image and Chest image.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.480-481
• /
• 2018

치아는 식사 등의 일상생활에서 중요 역할을 하는 부위로 구강 건강을 지속적으로 관리하는 것이 중요하다. 최근 모바일에서는 의료 영상을 얻을 수 있을 만큼의 해상도를 갖춘 이미지를 얻을 수 있으며 디지털 의료기기로의 가능성을 보여준다. 본 연구에서는 스마트폰을 사용한 가시광선 영역의 파장대를 이용하여 치아의 초기 우식병소 진단에 도움을 받고 효율적인 관리를 할 수 있도록 하는 형광영상기반 치아분석시스템을 개발하였다. 시스템은 405 nm 파장대의 가시광선과 515 nm 이하의 단파장을 차단하는 필터를 고정하는 장치를 3D 프린터를 기술을 통해 고안 및 제작하여 스마트폰으로 치아의 영상을 얻고, 병소를 검출하여 정량적인 데이터를 제공한다. 제안한 시스템은 사용자의 시간적, 공간적 제약 없이 객관적인 데이터를 기반으로 사용자의 구강위생관리상태를 제시하여 조기진단 및 예방이 가능하다.

• Journal of the Korean Society of Radiology
• /
• v.15 no.2
• /
• pp.153-158
• /
• 2021

X-ray image analysis is a very important field to improve the early diagnosis rate and prediction accuracy of periodontal disease. Research on the development and application of artificial intelligence-based algorithms to improve the quality of such dental X-ray images is being widely conducted worldwide. Thus, the aim of this study was to design a super-resolution algorithm for predicting periodontal disease and to evaluate its applicability in dental X-ray images. The super-resolution algorithm was constructed based on the convolution layer and ReLU, and an image obtained by up-sampling a low-resolution image by 2 times was used as an input data. Also, 1,500 dental X-ray data used for deep learning training were used. Quantitative evaluation of images used root mean square error and structural similarity, which are factors that can measure similarity through comparison of two images. In addition, the recently developed no-reference based natural image quality evaluator and blind/referenceless image spatial quality evaluator were additionally analyzed. According to the results, we confirmed that the average similarity and no-reference-based evaluation values were improved by 1.86 and 2.14 times, respectively, compared to the existing bicubic-based upsampling method when the proposed method was used. In conclusion, the super-resolution algorithm for predicting periodontal disease proved useful in dental X-ray images, and it is expected to be highly applicable in various fields in the future.