• Korean Journal of Radiology
• /
• 제25권4호
• /
• pp.384-394
• /
• 2024

Objective: To evaluate the image quality of novel dark-blood computed tomography angiography (CTA) imaging combined with deep learning reconstruction (DLR) compared to delayed-phase CTA images with hybrid iterative reconstruction (HIR), to visualize the cervical artery wall in patients with Takayasu arteritis (TAK). Materials and Methods: This prospective study continuously recruited 53 patients with TAK (mean age: 33.8 ± 10.2 years; 49 females) between January and July 2022 who underwent head-neck CTA scans. The arterial- and delayed-phase images were reconstructed using HIR and DLR. Subtracted images of the arterial-phase from the delayed-phase were then added to the original delayed-phase using a denoising filter to generate the final-dark-blood images. Qualitative image quality scores and quantitative parameters were obtained and compared among the three groups of images: Delayed-HIR, Dark-blood-HIR, and Dark-blood-DLR. Results: Compared to Delayed-HIR, Dark-blood-HIR images demonstrated higher qualitative scores in terms of vascular wall visualization and diagnostic confidence index (all P < 0.001). These qualitative scores further improved after applying DLR (Dark-blood-DLR compared to Dark-blood-HIR, all P < 0.001). Dark-blood DLR also showed higher scores for overall image noise than Dark-blood-HIR (P < 0.001). In the quantitative analysis, the contrast-to-noise ratio (CNR) values between the vessel wall and lumen for the bilateral common carotid arteries and brachiocephalic trunk were significantly higher on Dark-blood-HIR images than on Delayed-HIR images (all P < 0.05). The CNR values were significantly higher for Dark-blood-DLR than for Dark-blood-HIR in all cervical arteries (all P < 0.001). Conclusion: Compared with Delayed-HIR CTA, the dark-blood method combined with DLR improved CTA image quality and enhanced visualization of the cervical artery wall in patients with TAK.

• Korean Journal of Radiology
• /
• 제24권6호
• /
• pp.541-552
• /
• 2023

Objective: Detection of pneumoperitoneum using abdominal radiography, particularly in the supine position, is often challenging. This study aimed to develop and externally validate a deep learning model for the detection of pneumoperitoneum using supine and erect abdominal radiography. Materials and Methods: A model that can utilize "pneumoperitoneum" and "non-pneumoperitoneum" classes was developed through knowledge distillation. To train the proposed model with limited training data and weak labels, it was trained using a recently proposed semi-supervised learning method called distillation for self-supervised and self-train learning (DISTL), which leverages the Vision Transformer. The proposed model was first pre-trained with chest radiographs to utilize common knowledge between modalities, fine-tuned, and self-trained on labeled and unlabeled abdominal radiographs. The proposed model was trained using data from supine and erect abdominal radiographs. In total, 191212 chest radiographs (CheXpert data) were used for pre-training, and 5518 labeled and 16671 unlabeled abdominal radiographs were used for fine-tuning and self-supervised learning, respectively. The proposed model was internally validated on 389 abdominal radiographs and externally validated on 475 and 798 abdominal radiographs from the two institutions. We evaluated the performance in diagnosing pneumoperitoneum using the area under the receiver operating characteristic curve (AUC) and compared it with that of radiologists. Results: In the internal validation, the proposed model had an AUC, sensitivity, and specificity of 0.881, 85.4%, and 73.3% and 0.968, 91.1, and 95.0 for supine and erect positions, respectively. In the external validation at the two institutions, the AUCs were 0.835 and 0.852 for the supine position and 0.909 and 0.944 for the erect position. In the reader study, the readers' performances improved with the assistance of the proposed model. Conclusion: The proposed model trained with the DISTL method can accurately detect pneumoperitoneum on abdominal radiography in both the supine and erect positions.

• Korean Journal of Radiology
• /
• 제24권4호
• /
• pp.294-304
• /
• 2023

Objective: We aimed to investigate whether image standardization using deep learning-based computed tomography (CT) image conversion would improve the performance of deep learning-based automated hepatic segmentation across various reconstruction methods. Materials and Methods: We collected contrast-enhanced dual-energy CT of the abdomen that was obtained using various reconstruction methods, including filtered back projection, iterative reconstruction, optimum contrast, and monoenergetic images with 40, 60, and 80 keV. A deep learning based image conversion algorithm was developed to standardize the CT images using 142 CT examinations (128 for training and 14 for tuning). A separate set of 43 CT examinations from 42 patients (mean age, 10.1 years) was used as the test data. A commercial software program (MEDIP PRO v2.0.0.0, MEDICALIP Co. Ltd.) based on 2D U-NET was used to create liver segmentation masks with liver volume. The original 80 keV images were used as the ground truth. We used the paired t-test to compare the segmentation performance in the Dice similarity coefficient (DSC) and difference ratio of the liver volume relative to the ground truth volume before and after image standardization. The concordance correlation coefficient (CCC) was used to assess the agreement between the segmented liver volume and ground-truth volume. Results: The original CT images showed variable and poor segmentation performances. The standardized images achieved significantly higher DSCs for liver segmentation than the original images (DSC [original, 5.40%-91.27%] vs. [standardized, 93.16%-96.74%], all P < 0.001). The difference ratio of liver volume also decreased significantly after image conversion (original, 9.84%-91.37% vs. standardized, 1.99%-4.41%). In all protocols, CCCs improved after image conversion (original, -0.006-0.964 vs. standardized, 0.990-0.998). Conclusion: Deep learning-based CT image standardization can improve the performance of automated hepatic segmentation using CT images reconstructed using various methods. Deep learning-based CT image conversion may have the potential to improve the generalizability of the segmentation network.

• Korean Journal of Radiology
• /
• 제23권8호
• /
• pp.811-820
• /
• 2022

Objective: To develop a model incorporating radiomic features and clinical factors to accurately predict acute ischemic stroke (AIS) outcomes. Materials and Methods: Data from 522 AIS patients (382 male [73.2%]; mean age ± standard deviation, 58.9 ± 11.5 years) were randomly divided into the training (n = 311) and validation cohorts (n = 211). According to the modified Rankin Scale (mRS) at 6 months after hospital discharge, prognosis was dichotomized into good (mRS ≤ 2) and poor (mRS > 2); 1310 radiomics features were extracted from diffusion-weighted imaging and apparent diffusion coefficient maps. The minimum redundancy maximum relevance algorithm and the least absolute shrinkage and selection operator logistic regression method were implemented to select the features and establish a radiomics model. Univariable and multivariable logistic regression analyses were performed to identify the clinical factors and construct a clinical model. Ultimately, a multivariable logistic regression analysis incorporating independent clinical factors and radiomics score was implemented to establish the final combined prediction model using a backward step-down selection procedure, and a clinical-radiomics nomogram was developed. The models were evaluated using calibration, receiver operating characteristic (ROC), and decision curve analyses. Results: Age, sex, stroke history, diabetes, baseline mRS, baseline National Institutes of Health Stroke Scale score, and radiomics score were independent predictors of AIS outcomes. The area under the ROC curve of the clinical-radiomics model was 0.868 (95% confidence interval, 0.825-0.910) in the training cohort and 0.890 (0.844-0.936) in the validation cohort, which was significantly larger than that of the clinical or radiomics models. The clinical radiomics nomogram was well calibrated (p > 0.05). The decision curve analysis indicated its clinical usefulness. Conclusion: The clinical-radiomics model outperformed individual clinical or radiomics models and achieved satisfactory performance in predicting AIS outcomes.

• Korean Journal of Radiology
• /
• 제24권11호
• /
• pp.1151-1163
• /
• 2023

Objective: To develop a deep-learning-based bone age prediction model optimized for Korean children and adolescents and evaluate its feasibility by comparing it with a Greulich-Pyle-based deep-learning model. Materials and Methods: A convolutional neural network was trained to predict age according to the bone development shown on a hand radiograph (bone age) using 21036 hand radiographs of Korean children and adolescents without known bone development-affecting diseases/conditions obtained between 1998 and 2019 (median age [interquartile range {IQR}], 9 [7-12] years; male:female, 11794:9242) and their chronological ages as labels (Korean model). We constructed 2 separate external datasets consisting of Korean children and adolescents with healthy bone development (Institution 1: n = 343; median age [IQR], 10 [4-15] years; male: female, 183:160; Institution 2: n = 321; median age [IQR], 9 [5-14] years; male: female, 164:157) to test the model performance. The mean absolute error (MAE), root mean square error (RMSE), and proportions of bone age predictions within 6, 12, 18, and 24 months of the reference age (chronological age) were compared between the Korean model and a commercial model (VUNO Med-BoneAge version 1.1; VUNO) trained with Greulich-Pyle-based age as the label (GP-based model). Results: Compared with the GP-based model, the Korean model showed a lower RMSE (11.2 vs. 13.8 months; P = 0.004) and MAE (8.2 vs. 10.5 months; P = 0.002), a higher proportion of bone age predictions within 18 months of chronological age (88.3% vs. 82.2%; P = 0.031) for Institution 1, and a lower MAE (9.5 vs. 11.0 months; P = 0.022) and higher proportion of bone age predictions within 6 months (44.5% vs. 36.4%; P = 0.044) for Institution 2. Conclusion: The Korean model trained using the chronological ages of Korean children and adolescents without known bone development-affecting diseases/conditions as labels performed better in bone age assessment than the GP-based model in the Korean pediatric population. Further validation is required to confirm its accuracy.

• Korean Journal of Radiology
• /
• 제23권11호
• /
• pp.1055-1066
• /
• 2022

Objective: The clinical relevance of myosteatosis has not been well evaluated in patients with pancreatic ductal adenocarcinoma (PDAC), although sarcopenia has been extensively researched. Therefore, we evaluated the prognostic value of muscle quality, including myosteatosis, in patients with resectable PDAC treated surgically. Materials and Methods: We retrospectively evaluated 347 patients with resectable PDAC who underwent curative surgery (mean age ± standard deviation, 63.6 ± 9.6 years; 202 male). Automatic muscle segmentation was performed on preoperative computed tomography (CT) images using an artificial intelligence program. A single axial image of the portal phase at the inferior endplate level of the L3 vertebra was used for analysis in each patient. Sarcopenia was evaluated using the skeletal muscle index, calculated as the skeletal muscle area (SMA) divided by the height squared. The mean SMA attenuation was used to evaluate myosteatosis. Diagnostic cutoff values for sarcopenia and myosteatosis were devised using the Contal and O'Quigley methods, and patients were classified according to normal (nMT), sarcopenic (sMT), myosteatotic (mMT), or combined (cMT) muscle quality types. Multivariable Cox regression analyses were conducted to assess the effects of muscle type on the overall survival (OS) and recurrence-free survival (RFS) after surgery. Results: Eighty-four (24.2%), 73 (21.0%), 75 (21.6%), and 115 (33.1%) patients were classified as having nMT, sMT, mMT, and cMT, respectively. Compared to nMT, mMT and cMT were significantly associated with poorer OS, with hazard ratios (HRs) of 1.49 (95% confidence interval, 1.00-2.22) and 1.68 (1.16-2.43), respectively, while sMT was not (HR of 1.40 [0.94-2.10]). Only mMT was significantly associated with poorer RFS, with an HR of 1.59 (1.07-2.35), while sMT and cMT were not. Conclusion: Myosteatosis was associated with poor OS and RFS in patients with resectable PDAC who underwent curative surgery.

• 한국컴퓨터정보학회논문지
• /
• 제29권4호
• /
• pp.73-82
• /
• 2024

2017년 UN에서는 전 세계적으로 60세 이상 인구는 모든 젊은 연령층보다 빠르게 증가하고 있으며, 2050년까지 60세 이상 인구는 아프리카를 제외한 전 세계 인구의 최소 25%를 구성할 것으로 예상하였다. 세계는 전반적으로 고령화로 인해 일을 할 수 있는 인구의 증가율이 감소하고 있으며, 청년층은 힘들고 어려운 직업을 선호하지 않고 있다. 이론적으로는 인공지능을 겸비한 AI가 모든 분야에서 사람을 대신할 수 있다고 하지만 윤리적인 판단 등 현실 세계의 정보보호 분야에서는 사람의 판단과 노하우가 절대적으로 필요하다. 이에, 본 논문에서는 IT 종사자 중 50대 이상 퇴직자 또는 전직을 희망하는 사람을 대상으로 재교육을 통해 현업으로 유입시키는 방법을 제안하고자 한다. 연구를 위해 수요 부분의 정부·공공기관 21곳과 공급 부분의 보안관제전문업체 9곳을 대상으로 설문하였으며 설문 결과 공급(78%)와 수요(90%) 모두가 절대적으로 필요하다는 데 의견을 모았다. 향후 이 연구 결과를 토대로 현장에 적용한다면 인구 저출산 100세 시대에 정보보호분야 시니어의 전략적 육성으로 대한민국 정보보호산업의 초석이 될 신규시장을 발굴할 수 있을 것이다.

• 문화기술의 융합
• /
• 제10권2호
• /
• pp.565-574
• /
• 2024

챗GPT는 우리 사회의 모든 분야에 커다란 반향을 일으키고 있으며, 이는 교육 분야에서도 마찬가지이다. 하지만 학교 현장에서의 챗GPT에 관한 학계 및 사회적 논의는 주로 표절과 같은 부정행위에 초점이 맞춰져있으며, 학생들의 인공지능 기술 활용으로 인한 긍정적인 효과에 대해서는 상대적으로 연구가 부족한 실정이다. 그와 더불어, 코로나 시대의 교육 위기 속에서 학업열의를 증진시키는 연구의 필요성이 제기되고 있다. 이러한 문제인식에 따라 우리는 학생들의 챗GPT 리터러시 수준에 따라 학업열의가 어떻게 나타나는지에 대해 알아보고, 그 과정에서 학생들의 학업 자신감 및 지각된 학업역량이 매개효과를 나타내는지를 확인했다. 이를 위해 406명의 대학생을 대상으로 수집한 데이터를 SPSS로 분석했다. 그 결과, 챗GPT 리터러시는 학업열의에 긍정적인 영향을 미치는 것이 확인되었으며, 학업자신감은 챗GPT 리터러시와 학업열의 사이를 매개했다. 또한 학업자신감과 지각된 학업역량은 이중으로 매개되었을 때 유의한 결과를 도출했다. 이러한 결과를 바탕으로 우리는 챗GPT 리터러시와 학업열의 사이의 이론적 매커니즘을 명확히 규명할 수 있다는 이론적 함의를 논하였다. 또한 이와 관련하여 챗GPT 리터러시 교육의 중요성에 관한 실용적 함의에 대해도 기술하였다.

• 문화기술의 융합
• /
• 제10권2호
• /
• pp.493-498
• /
• 2024

최근 생성형 인공지능 기술이 급속도로 발전함에 따라, 이를 실무에 활용하는 방법에 대한 관심이 높아지고 있다. 또한 사용자 요구에 부합하는 결과물을 생성하기 위한 프롬프트 엔지니어링의 중요성이 새롭게 조명되고 있다. 이러한 생성형 인공지능의 새로운 활용 가능성을 탐구하는 것은 중요한 가치를 지닐 수 있다. 본 연구는 대표적인 생성형 인공지능인 챗지피티 4.0을 활용하여 온라인 고객 리뷰 데이터 분석을 통한 효과적인 사용자 경험 평가 방법을 제안하는 것을 목적으로 한다. 사용자 경험 평가 방법은 사용자 경험 계층의 6단계 요소인 '기능성', '신뢰성', '사용성', '편의성', '감성', '의미성'을 기반으로 수행되었다. 본 연구를 위해 프롬프트 엔지니어링의 이해도를 높이고 사용자경험 계층의 명확한 개념을 파악하는 문헌연구를 수행하고, 이를 기반으로 프롬프트를 작성 및 수집된 온라인 고객 리뷰 데이터 분석을 통한 사용자 경험 평가 방법을 위한 실험이 수행되었다. 본 연구에서 우리는 사용자 경험 요소에 대한 정확한 정의 및 분류 과정에 대한 설명 입력 시, 챗지피티는 사용자 경험 평가에 대한 우수한 성능을 나타냈으나, 시간적 제약으로 다량의 데이터 분석에 한계를 나타냈음을 밝힌다. 우리는 사용자 경험 평가에 챗지피티 4.0을 활용하는 방법을 소개하고 제안함으로써 UX 분야의 발전에 공헌할 수 있는 것으로 기대한다.

• 서비스연구
• /
• 제14권1호
• /
• pp.77-90
• /
• 2024

본 연구는 소비자의 감성과 경험이 중요한 서비스 분야에서 생성형 인공지능을 활용하는 방법에 대한 조사를 목표로 활용시의 환각 현상을 최소화하고, 소비자의 감성 및 경험에 대한 전략적 서비스를 개발하는 것에 초점을 맞추고 있다. 이를 위해 기계적인 방식의 접근과 사용자가 프롬프트를 직접 생성하는 방식을 검토하였고, 사업아이템 정의 제공, 페르소나 특성 값 제공, 예시와 맥락형 동사명령, 출력 포멧과 톤 컨셉 지정 등의 프롬프트 생성 요인을 중심으로 실험적으로 적용하였다. 연구는 생성형 AI가 제공하는 맞춤형 콘텐츠의 정확성과 사용자 만족도를 향상시키는 데 기여할 수 있는 방안을 탐색한다. 또한, 이러한 접근 방식은 생성형 인공지능을 실제 서비스에 적용 시 발생할 수 있는 환각 현상 중심의 문제들을 해결하는 데 중요한 역할을 하며, 생성형 인공지능을 통한 소비자 서비스 혁신에 기여할 것으로 기대한다. 연구 결과는 소비자의 감성과 경험을 풍부하게 해석하는데 생성형 인공지능이 중요한 역할을 할 수 있음을 보여주며, 이는 다양한 산업 분야에서의 활용 가능성을 넓히고, 기술 발전을 넘어 소비자 감성 및 경험 전략의 새로운 방향을 제시할 것으로 기대한다. 하지만, 아직은 연구가 생소한 생성형 AI 기술 기반의 연구를 진행함으로써 미흡한 부분이 많다. 향후 연구에서는 더 다양한 산업 환경 적용으로 연구요인들의 범용성과 조건별 효과를 더 깊이 탐구할 필요가 있다. 또한, AI 기술의 급속한 발전에 따라 새로운 형태의 환각 증상과 이에 대응하는 새로운 전략 개발에 관한 연구가 지속해서 이루어져야 할 것이다.