• Korean Journal of Radiology
• /
• v.22 no.3
• /
• pp.476-488
• /
• 2021

Objective: We aimed to develop a deep neural network for segmenting lung parenchyma with extensive pathological conditions on non-contrast chest computed tomography (CT) images. Materials and Methods: Thin-section non-contrast chest CT images from 203 patients (115 males, 88 females; age range, 31-89 years) between January 2017 and May 2017 were included in the study, of which 150 cases had extensive lung parenchymal disease involving more than 40% of the parenchymal area. Parenchymal diseases included interstitial lung disease (ILD), emphysema, nontuberculous mycobacterial lung disease, tuberculous destroyed lung, pneumonia, lung cancer, and other diseases. Five experienced radiologists manually drew the margin of the lungs, slice by slice, on CT images. The dataset used to develop the network consisted of 157 cases for training, 20 cases for development, and 26 cases for internal validation. Two-dimensional (2D) U-Net and three-dimensional (3D) U-Net models were used for the task. The network was trained to segment the lung parenchyma as a whole and segment the right and left lung separately. The University Hospitals of Geneva ILD dataset, which contained high-resolution CT images of ILD, was used for external validation. Results: The Dice similarity coefficients for internal validation were 99.6 ± 0.3% (2D U-Net whole lung model), 99.5 ± 0.3% (2D U-Net separate lung model), 99.4 ± 0.5% (3D U-Net whole lung model), and 99.4 ± 0.5% (3D U-Net separate lung model). The Dice similarity coefficients for the external validation dataset were 98.4 ± 1.0% (2D U-Net whole lung model) and 98.4 ± 1.0% (2D U-Net separate lung model). In 31 cases, where the extent of ILD was larger than 75% of the lung parenchymal area, the Dice similarity coefficients were 97.9 ± 1.3% (2D U-Net whole lung model) and 98.0 ± 1.2% (2D U-Net separate lung model). Conclusion: The deep neural network achieved excellent performance in automatically delineating the boundaries of lung parenchyma with extensive pathological conditions on non-contrast chest CT images.

• Korean Journal of Radiology
• /
• v.25 no.2
• /
• pp.189-198
• /
• 2024

Objective: To investigate the prognostic utility of radiomics features extracted from ¹⁸F-fluorodeoxyglucose (FDG) PET/CT combined with clinical factors and metabolic parameters in predicting progression-free survival (PFS) and overall survival (OS) in individuals diagnosed with extranodal nasal-type NK/T cell lymphoma (ENKTCL). Materials and Methods: A total of 126 adults with ENKTCL who underwent ¹⁸F-FDG PET/CT examination before treatment were retrospectively included and randomly divided into training (n = 88) and validation cohorts (n = 38) at a ratio of 7:3. Least absolute shrinkage and selection operation Cox regression analysis was used to select the best radiomics features and calculate each patient's radiomics scores (RadPFS and RadOS). Kaplan-Meier curve and Log-rank test were used to compare survival between patient groups risk-stratified by the radiomics scores. Various models to predict PFS and OS were constructed, including clinical, metabolic, clinical + metabolic, and clinical + metabolic + radiomics models. The discriminative ability of each model was evaluated using Harrell's C index. The performance of each model in predicting PFS and OS for 1-, 3-, and 5-years was evaluated using the time-dependent receiver operating characteristic (ROC) curve. Results: Kaplan-Meier curve analysis demonstrated that the radiomics scores effectively identified high- and low-risk patients (all P < 0.05). Multivariable Cox analysis showed that the Ann Arbor stage, maximum standardized uptake value (SUVmax), and RadPFS were independent risk factors associated with PFS. Further, β2-microglobulin, Eastern Cooperative Oncology Group performance status score, SUVmax, and RadOS were independent risk factors for OS. The clinical + metabolic + radiomics model exhibited the greatest discriminative ability for both PFS (Harrell's C-index: 0.805 in the validation cohort) and OS (Harrell's C-index: 0.833 in the validation cohort). The time-dependent ROC analysis indicated that the clinical + metabolic + radiomics model had the best predictive performance. Conclusion: The PET/CT-based clinical + metabolic + radiomics model can enhance prognostication among patients with ENKTCL and may be a non-invasive and efficient risk stratification tool for clinical practice.

• Journal of the Korean Society of Radiology
• /
• v.3 no.1
• /
• pp.29-39
• /
• 2009

In this paper has proposed to the recognition of the disease on medical images using neural network. The neural network is constructed as three-layers of the input-layer, the hidden-layer and the output-layer. The training method applied for the recognition of disease region is adaptive error back-propagation. The low-frequency region analyzed by DWT are expressed by matrix. The coefficient-values of the characteristic polynomial applied are n+1. The normalized maximum value +1 and minimum value -1 in the range of tangent-sigmoid transfer function are applied to be use as the input vector of the neural network. To prove the validity of the proposed methods used in the experiment with a simulation experiment, the input medical image recognition rate the evaluation of areas of disease. As a result of the experiment, the characteristic polynomial coefficient of low-frequency area matrix, conversed to 4 level DWT, was proved to be optimum to be applied to the feature parameter. As for the number of training, it was marked fewest in 0.01 of learning coefficient and 0.95 of momentum, when the adaptive error back-propagation was learned by inputting standardized feature parameter into organized neural network. As to the training result when the learning coefficient was 0.01, and momentum was 0.95, it was 100% recognized in fifty-five times of the stomach image, fifty-five times of the chest image, forty-six times of the CT image, fifty-five times of ultrasonogram, and one hundred fifty-seven times of angiogram.

• Journal of Korea Multimedia Society
• /
• v.23 no.3
• /
• pp.468-475
• /
• 2020

In this paper, we proposed a three-dimensional visualization system for medical images in augmented reality based on deep learning. In the proposed system, the artificial neural network model performed fully automatic segmentation of the region of lung and pulmonary nodule from chest CT images. After applying the three-dimensional volume rendering method to the segmented images, it was visualized in augmented reality devices. As a result of the experiment, when nodules were present in the region of lung, it could be easily distinguished with the naked eye. Also, the location and shape of the lesions were intuitively confirmed. The evaluation was accomplished by comparing automated segmentation results of the test dataset to the manual segmented image. Through the evaluation of the segmentation model, we obtained the region of lung DSC (Dice Similarity Coefficient) of 98.77%, precision of 98.45%, recall of 99.10%. And the region of pulmonary nodule DSC of 91.88%, precision of 93.05%, recall of 90.94%. If this proposed system will be applied in medical fields such as medical practice and medical education, it is expected that it can contribute to custom organ modeling, lesion analysis, and surgical education and training of patients.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.57-61
• /
• 2005

This paper presents a new colonoscopy training simulator that includes a specialized haptic device and graphics algorithms to transfer haptic sensation through a long and flexible tube, and manage large number of polygons. The developed haptic device makes the colonoscope tube move along the two guiding rods in the translational direction. The torque of the roll motion is transferred by a timing belt and pulleys. A special guide is developed, which allows the force and torque from the motors to be transmitted to the user without loss. The haptic device is evaluated by physicians. One of the important skills of the colonoscopy, jiggling is incorporated for the first time by the developed sensor mechanism using photo-sensors. A colonoscope handle that shares the look, feel, and functions with the actual colonoscope, is developed with the necessary electronics inside. The number of polygons is reduced by an edge-collapse algorithm for real-time simulation. The algorithms to import CT data, to segment the colon image, to extract centerline of the colon, and to construct the colon surface, are integrated into a Colon Modeling Kit system that performs all these processes in real-time.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.9
• /
• pp.4085-4089
• /
• 2014

Background: Atelectasis is an important prognostic factor that can cause pleuritic chest pain, coughing or dyspnea, and even may be a cause of death. In this study, we aimed to investigate the potential impact of atelectasis and PET parameters on survival and the relation between atelectasis and PET parameters. Materials and Methods: The study consisted of patients with lung cancer with or without atelectasis who underwent $^{18}F$-FDG PET/CT examination before receiving any treatment. $^{18}F$-FDG PET/CT derived parameters including tumor size, SUVmax, SUVmean, MTV, total lesion glycosis (TLG), SUV mean of atelectasis area, atelectasis volume, and histological and TNM stage were considered as potential prognostic factors for overall survival. Results: Fifty consecutive lung cancer patients (22 patients with atelectasis and 28 patients without atelectasis, median age of 65 years) were evaluated in the present study. There was no relationship between tumor size and presence or absence of atelectasis, nor between presence/absence of atelectasis and TLG of primary tumors. The overall one-year survival rate was 83% and median survival was 20 months (n=22) in the presence of atelectasis; the overall one-year survival rate was 65.7% (n=28) and median survival was 16 months (p=0.138) in the absence of atelectasis. With respect to PFS; the one-year survival rate of AT+ patients was 81.8% and median survival was 19 months; the one-year survival rate of AT-patients was 64.3% and median survival was 16 months (p=0.159). According to univariate analysis, MTV, TLG and tumor size were significant risk factors for PFS and OS (p<0.05). However, SUVmax was not a significant factor for PFS and OS (p>0.05). Conclusions: The present study suggested that total lesion glycolysis and metabolic tumor volume were important predictors of survival in lung cancer patients, in contrast to SUVmax. In addition, having a segmental lung atelectasis seems not to be a significant factor on survival.

• Proceedings of the Korea Contents Association Conference
• /
• 2012.05a
• /
• pp.203-204
• /
• 2012

한류 문화산업은 드라마로부터 시작하여 K-Pop으로 영역이 넓어지는 장르의 다양화와 함께 동남아에서 남미, 유럽까지 지역적 영향력도 커져가고 있다. 문화산업 관련 직종 중 기획-경영 분야에는 문화 예술에 대한 지식과 함께 경제학, 경영학, 기술 분야(CT: Culture Technology)의 다양하고 고도의 융합지식 필요하다. 한류 문화산업 전문가는 또한 한류 문화상품이 영향력을 미치는 중국, 일본, 베트남, 서남 아시아권의 지역정보와 해당 국가의 문화, 법, 제도 이해가 필요하다. 이 연구에서는 한류 문화산업의 전문가를 양성하기 위한 산업계의 수요를 분석하고 이들을 양성하기 위한 체계적인 교육과정을 도출하는 프로세스를 제시한다.

• Korean Journal of Bronchoesophagology
• /
• v.7 no.1
• /
• pp.76-79
• /
• 2001

A 60-year-old male was admitted due to cervical spine injury (C7-T1 fracture dislocation) and quadriparesis after slip down. During conservative management in department of neurologic surgery, he complainted of fever, dyspnea, neck swelling. Follow up cervicothoracic CT revealed abscess pocket in paraglottic, retropharyngeal, anterior cervical spaces and mediastinum. Also noted bilateral pleural effusions. Under impression of descending necrotizing mediastinitis (DNM). cervical drainage and bilateral chest tube insertion was performed immediately. On next day. mediastinal drainage through mediastinotomy was performed with careful handling of cervical spine. Escherichia coli was identified in bacteriologic culture. Wire fixation of dislocated C7-T1 spine through Posterior approach was performed on 30th days after mediastinotomy. Right chest tube was removed on 40th days. At now, the patient is on rehabilitation and physical training program. DNM is relatively rare, but lethal disease with high mortality. Immedate and sufficient mediastinal drainage is essential in treatment.

• Proceedings of the KIEE Conference
• /
• summer
• /
• pp.384-387
• /
• 2004

The direction and the type of a fault on a transmission line need to be identified rapidly and correctly, The work described in this paper addresses the problem encountered by a conventional algorithm in a fault type classification in double circuit line, this arises due to a mutual coupling and CT saturation under the fault condition. We present an approach to identify fault type with novel neural network on double circuit transmission line. The neural network based on combined unsupervised training method provides the ability classify the fault type by different patterns of the associated voltages and currents.

• Progress in Medical Physics
• /
• v.26 no.3
• /
• pp.127-136
• /
• 2015

The purpose of this study is to perform a dosimetric evaluation of amplitude-based respiratory gating for the delivery of volumetric modulated arc therapy (VMAT). We selected two types of breathing patterns, subjectively among patients with respiratory-gated treatment log files. For patients that showed consistent breathing patterns (CBP) relative to the 4D CT respiration patterns, the variability of the breath-holding position during treatment was observed within the thresholds. However, patients with inconsistent breathing patterns (IBP) show differences relative to those with CBP. The relative isodose distribution was evaluated using an EBT3 film by comparing gated delivery to static delivery, and an absolute dose measurement was performed with a $0.6cm^3$ Farmer-type ion chamber. The passing rate percentages under the 3%/3 mm gamma analysis for Patients 1, 2 and 3 were respectively 93.18%, 91.16%, and 95.46% for CBP, and 66.77%, 48.79%, and 40.36% for IBP. Under the more stringent criteria of 2%/2 mm, passing rates for Patients 1, 2 and 3 were respectively 73.05%, 67.14%, and 86.85% for CBP, and 46.53%, 32.73%, and 36.51% for IBP. The ion chamber measurements were within 3.5%, on average, of those calculated by the TPS and within 2.0%, on average, when compared to the static-point dose measurements for all cases of CBP. Inconsistent breathing patterns between 4D CT simulation and treatment may cause considerable dosimetric differences. Therefore, patient training is important to maintain consistent breathing amplitude during CT scan acquisition and treatment delivery.