• Korean Journal of Radiology
• /
• v.23 no.3
• /
• pp.343-354
• /
• 2022

Objective: To develop and evaluate a deep learning-based artificial intelligence (AI) model for detecting skull fractures on plain radiographs in children. Materials and Methods: This retrospective multi-center study consisted of a development dataset acquired from two hospitals (n = 149 and 264) and an external test set (n = 95) from a third hospital. Datasets included children with head trauma who underwent both skull radiography and cranial computed tomography (CT). The development dataset was split into training, tuning, and internal test sets in a ratio of 7:1:2. The reference standard for skull fracture was cranial CT. Two radiology residents, a pediatric radiologist, and two emergency physicians participated in a two-session observer study on an external test set with and without AI assistance. We obtained the area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity along with their 95% confidence intervals (CIs). Results: The AI model showed an AUROC of 0.922 (95% CI, 0.842-0.969) in the internal test set and 0.870 (95% CI, 0.785-0.930) in the external test set. The model had a sensitivity of 81.1% (95% CI, 64.8%-92.0%) and specificity of 91.3% (95% CI, 79.2%-97.6%) for the internal test set and 78.9% (95% CI, 54.4%-93.9%) and 88.2% (95% CI, 78.7%-94.4%), respectively, for the external test set. With the model's assistance, significant AUROC improvement was observed in radiology residents (pooled results) and emergency physicians (pooled results) with the difference from reading without AI assistance of 0.094 (95% CI, 0.020-0.168; p = 0.012) and 0.069 (95% CI, 0.002-0.136; p = 0.043), respectively, but not in the pediatric radiologist with the difference of 0.008 (95% CI, -0.074-0.090; p = 0.850). Conclusion: A deep learning-based AI model improved the performance of inexperienced radiologists and emergency physicians in diagnosing pediatric skull fractures on plain radiographs.

• Korean Journal of Radiology
• /
• v.2 no.3
• /
• pp.138-144
• /
• 2001

Objective: Bronchogenic carcinoma can mimic or be masked by pulmonary tuberculosis (TB), and the aim of this study was to describe the radiologic findings and clinical significance of bronchogenic carcinoma and pulmonary TB which coexist in the same lobe. Materials and Methods: The findings of 51 patients (48 males and three females, aged 48-79 years) in whom pulmonary TB and bronchogenic carcinoma coexisted in the same lobe were analyzed. The morphologic characteristics of a tumor, such as its diameter and margin, the presence of calcification or cavitation, and mediastinal lymphadenopathy, as seen at CT, were retrospectively assessed, and the clinical stage of the lung cancer was also determined. Using the serial chest radiographs available for 21 patients, the possible causes of delay in the diagnosis of lung cancer were analyzed. Results: Lung cancers with coexisting pulmonary TB were located predominantly in the upper lobes (82.4%). The mean diameter of the mass was 5.3 cm, and most tumors (n=42, 82.4%) had a lobulated border. Calcification within the tumor was seen in 20 patients (39.2%), and cavitation in five (9.8%). Forty-two (82.4%) had mediastinal lymphadenopathy, and more than half the tumors (60.8%) were at an advanced stage [IIIB (n=11) or IV (n=20)]. The average delay in diagnosing lung cancer was 11.7 (range, 1-24) months, and the causes of this were failure to observe new nodules masked by coexisting stable TB lesions (n=8), misinterpretation of new lesions as aggravation of TB (n=5), misinterpretation of lung cancer as tuberculoma at initial radiography (n=4), masking of the nodule by an active TB lesion (n=3), and subtleness of the lesion (n=1). Conclusion: Most cancers concurrent with TB are large, lobulated masses with mediastinal lymphadenopathy, indicating that the morphologic characteristics of lung cancer with coexisting pulmonary TB are similar to those of lung cancer without TB. The diagnosis of lung cancer is delayed mainly because of masking by a tuberculous lesion, and this suggests that in patients in whom a predominant or growing nodule is present and who show little improvement of symptoms despite antituberculous or other medical therapy, coexisting cancer should be suspected.

• Journal of the Korean Society of Radiology
• /
• v.18 no.3
• /
• pp.309-315
• /
• 2024

The Study In order to obtain a Clear Image of fractures in Atlas and Axis of Cervical Spin, this Study obtained Images examined using a Cervical Spin Pantom and a Radiation device, and then Commissioned five Radiologists and three Orthopedic Regent Doctors for Subjective Psychological ROC (Receiver, Operation, Characteristic) Evaluation. As aRresult, the X-ray Tube was Tilted 15° toward the leg to Receive a high score of 29 Points during the Eamination, and the Objective Evaluation Signal-to-noise Ratio (SNR) was 6.032 points, which was high. In Addition, by Tilting the X-ray tube 10° toward the head, it received a high score of 33 points during the Examination, and a high score of 7.840 Points in the objective evaluation. And as a result of Subjectively and Objectively Statistically Evaluating the Examined Images, the Cronbach Alpha value was Calculated as 0.791. Since the Cronbach Alpha value was 0.7 or higher, the Reliability could be Evaluated as 'very good', and the significance probability (p) was Statistically Significant at 0.042 points.

• Journal of Korean society of Dental Hygiene
• /
• v.23 no.6
• /
• pp.431-439
• /
• 2023

Objectives: The aim of this study was to investigate the essential services, importance, interval of examinations, and role of dental hygienists in children's family dentistry program. Methods: A total of 124 participants completed a self-administered questionnaire. The data were analyzed using independent t-test, and ANOVA, and Pearson correlation analysis. Results: All participants stated that oral examination and panoramic radiography are required; children should be educated about brushing, use of oral products, and regular check-ups; and preventive treatments such as molar sealants and prophylaxis should be offered. They stated that light-curing resins and glass ionomer fillings should be offered in treatment services. They stated that examination intervals should be shorter for education and prevention rather than treatment. Dental hygiene students were more likely than dentists and dental hygienists to say that the program was more important. There was a correlation between oral examinations and education and treatment, and between essential services and their importance. Conclusions: Services considered essential and important should be provided first, education and prevention should be provided more frequently than treatment, and their importance should be emphasized not only to dental hygiene students but also to dental hygienists and dentists who are the main providers of services.

• Imaging Science in Dentistry
• /
• v.54 no.1
• /
• pp.81-91
• /
• 2024

Purpose: The objective of this study was to propose a deep-learning model for the detection of the mandibular canal on dental panoramic radiographs. Materials and Methods: A total of 2,100 panoramic radiographs (PANs) were collected from 3 different machines: RAYSCAN Alpha (n=700, PAN A), OP-100 (n=700, PAN B), and CS8100 (n=700, PAN C). Initially, an oral and maxillofacial radiologist coarsely annotated the mandibular canals. For deep learning analysis, convolutional neural networks (CNNs) utilizing U-Net architecture were employed for automated canal segmentation. Seven independent networks were trained using training sets representing all possible combinations of the 3 groups. These networks were then assessed using a hold-out test dataset. Results: Among the 7 networks evaluated, the network trained with all 3 available groups achieved an average precision of 90.6%, a recall of 87.4%, and a Dice similarity coefficient (DSC) of 88.9%. The 3 networks trained using each of the 3 possible 2-group combinations also demonstrated reliable performance for mandibular canal segmentation, as follows: 1) PAN A and B exhibited a mean DSC of 87.9%, 2) PAN A and C displayed a mean DSC of 87.8%, and 3) PAN B and C demonstrated a mean DSC of 88.4%. Conclusion: This multi-device study indicated that the examined CNN-based deep learning approach can achieve excellent canal segmentation performance, with a DSC exceeding 88%. Furthermore, the study highlighted the importance of considering the characteristics of panoramic radiographs when developing a robust deep-learning network, rather than depending solely on the size of the dataset.

• Journal of the Korean Society of Radiology
• /
• v.85 no.1
• /
• pp.240-246
• /
• 2024

Epithelioid angiosarcoma is a rare variant of angiosarcoma characterized by an epithelioid morphology that mimics carcinoma. Therefore, multicentric epithelioid angiosarcoma is easily misdiagnosed as bone metastasis from carcinoma and has an aggressive clinical course. Here, we present a rare case of a 61-year-old male with multicentric epithelioid angiosarcoma of the bone. Plain radiography, CT, and MRI revealed multiple osteolytic lesions in both femurs; some lesions showed soft tissue extension with cortical bone destruction. Interestingly, PET-CT revealed that the lesions were only distributed along the bones of the lower extremities, including the pelvic bones, femurs, and tibiae. Despite histological analysis initially suggesting metastatic carcinoma, after additional immunohistological staining, including that for vascular markers (CD31 and ERG), the final diagnosis was epithelioid angiosarcoma. A better understanding of the clinicoradiological features of this disease may help eliminate diagnostic confusion and provide better management.

• Korean Journal of Radiology
• /
• v.25 no.7
• /
• pp.613-622
• /
• 2024

Objective: In Korea, radiology has been positioned towards the early adoption of artificial intelligence-based software as medical devices (AI-SaMDs); however, little is known about the current usage, implementation, and future needs of AI-SaMDs. We surveyed the current trends and expectations for AI-SaMDs among members of the Korean Society of Radiology (KSR). Materials and Methods: An anonymous and voluntary online survey was open to all KSR members between April 17 and May 15, 2023. The survey was focused on the experiences of using AI-SaMDs, patterns of usage, levels of satisfaction, and expectations regarding the use of AI-SaMDs, including the roles of the industry, government, and KSR regarding the clinical use of AI-SaMDs. Results: Among the 370 respondents (response rate: 7.7% [370/4792]; 340 board-certified radiologists; 210 from academic institutions), 60.3% (223/370) had experience using AI-SaMDs. The two most common use-case of AI-SaMDs among the respondents were lesion detection (82.1%, 183/223), lesion diagnosis/classification (55.2%, 123/223), with the target imaging modalities being plain radiography (62.3%, 139/223), CT (42.6%, 95/223), mammography (29.1%, 65/223), and MRI (28.7%, 64/223). Most users were satisfied with AI-SaMDs (67.6% [115/170, for improvement of patient management] to 85.1% [189/222, for performance]). Regarding the expansion of clinical applications, most respondents expressed a preference for AI-SaMDs to assist in detection/diagnosis (77.0%, 285/370) and to perform automated measurement/quantification (63.5%, 235/370). Most respondents indicated that future development of AI-SaMDs should focus on improving practice efficiency (81.9%, 303/370) and quality (71.4%, 264/370). Overall, 91.9% of the respondents (340/370) agreed that there is a need for education or guidelines driven by the KSR regarding the use of AI-SaMDs. Conclusion: The penetration rate of AI-SaMDs in clinical practice and the corresponding satisfaction levels were high among members of the KSR. Most AI-SaMDs have been used for lesion detection, diagnosis, and classification. Most respondents requested KSR-driven education or guidelines on the use of AI-SaMDs.

• Journal of the Korean Society of Radiology
• /
• v.3 no.3
• /
• pp.5-11
• /
• 2009

When using Image Guided Radiation Therapy, the patient is placed using skin marker first and after confirming anatomical location using OBI, the couch is moved to correct the set up. Evaluation for the error made at that moment was done. Through comparing $0^{\circ}$ and $270^{\circ}$ direction DRR image and OBI image with 2D-2D matching when therapy planning, comparison between patient's therapy plan setup and actual treatment setup was made to observe the error. Treatment confirmation on important organs such as head, neck and spinal cord was done every time through OBI setup and other organs such as chest, abdomen and pelvis was done 2 ~ 3 times a week. But corrections were all recorded on OIS so that evaluation on accuracy could be made through using skin index which was divided into head, neck, chest and abdomen-pelvis on 160 patients. Average setup error for head and neck patient on each AP, SI, RL direction was $0.2{\pm}0.2cm$, $-0.1{\pm}0.1cm$, $-0.2{\pm}0.0cm$, chest patient was $-0.5{\pm}0.1cm$, $0.3{\pm}0.3cm$, $0.4{\pm}0.2cm$, and abdomen was $0.4{\pm}0.4cm$, $-0.5{\pm}0.1cm$, $-0.4{\pm}0.1cm$. In case of pelvis, it was $0.5{\pm}0.3cm$, $0.8{\pm}0.4cm$, $-0.3{\pm}0.2cm$. In rigid body parts such as head and neck showed lesser setup error compared to chest and abdomen. Error was greater on chest in horizontal axis and in AP direction, abdomen-pelvis showed greater error. Error was greater on chest in horizontal axis because of the curve in patient's body when the setup is made. Error was greater on abdomen in AP direction because of the change in front and back location due to breathing of patient. There was no systematic error on patient setup system. Since OBI confirms the anatomical location, when focus is located on the skin, it is more precise to use skin marker to setup. When compared with 3D-3D conformation, although 2D-2D conformation can't find out the rolling error, it has lesser radiation exposure and shorter setup confirmation time. Therefore, on actual clinic, 2D-2D conformation is more appropriate.

• Tuberculosis and Respiratory Diseases
• /
• v.56 no.6
• /
• pp.611-618
• /
• 2004

Background : Idiopathic pulmonary fibrosis(IPF), a subtype of IIP(idiopathic interstitial pneumonia), is a fatal disease with a 3-5 year median survival. Many attempts at treating this condition have failed to demonstrate a survival benefit in IPF. Recently Ziesche et $al^{12}$ reported the efficacy of IFN-${\gamma}$ for treating IPF but there is still some controversy. The aim of this study was to determine the efficacy of IFN-${\gamma}$ in patients with advanced IPF who had not been responsive to steroid and cytotoxic agents. Method : Nine patients with advanced IPF(age: $55.4{\pm}15.3$ years, Male: Female=8:1) were enrolled. One year treatment regime with 2 million IU of IFN-${\gamma}$ administered subcutaneously three times a week, and low dose prednisolone(10-30 mg/d) was also used. In the case of a definite aggravation and serious side effects, the IFN-${\gamma}$ was discontinued. During the IFN-${\gamma}$ trial, a pulmonary function test and chest radiography were checked every three month throughout the study. Result : 1) Among 9 patients, only 4 patients were able to complete the 12 month treatment with IFN-${\gamma}$, and 5 patients died during the treatment period. 2) No improvement either in the respiratory symptoms or pulmonary functions were observed any of the patients, even in those who completed the 12 months trial of IFN-${\gamma}$, 3) At the time of IFN-${\gamma}$ trial, the survivors who finished the IFN-${\gamma}$ treatment for 12 months had a higher oxygen level($81.3{\pm}2.8$ vs. $67.4{\pm}8.4$, P=0.024) and a better pulmonary function(FVC: $61.3{\pm}5.1$ % predicted vs. $45.7{\pm}12.3%$, P=0.048, and $D_Lco$: $45.0{\pm}5.0%$ predicted vs. $30.8{\pm}11.2%$, P=0.048) than the non-survivors. Conclusion : Our data suggested that IFN-${\gamma}$ therapy was not effective in the patients with advanced IPF refractory compared with other therapeutic agents. Furthermore, these results suggest that severe impairment of the pulmonary function and hypoxemia during the IFN-${\gamma}$ therapy requires special attention.

• The Journal of Korean Society for Radiation Therapy
• /
• v.26 no.2
• /
• pp.239-245
• /
• 2014

Purpose : The purpose of this study is to verify the accuracy of dose delivery according to the pitch and roll rotational setup error with 6D robotic couch in Intensity Modulated Radiation Therapy (IMRT) for pelvic region in patients. Materials and Methods : Trilogy(Varian, USA) and 6D robotic couch(ProturaTM 1.4, CIVCO, USA) were used to measure and analyze the rotational setup error of 14 patients (157 setup cases) for pelvic region. The total 157 Images(CBCT 78, Radiography 79) were used to calculate the mean value and the incidence of pitch and roll rotational setup error with Microsoft Office Excel 2007. The measured data (3 mm, 3%) at the reference angle ($0^{\circ}$) without couch rotation of pitch and roll direction was compared to the others at different pitch and roll angles ($1^{\circ}$, $1.5^{\circ}$, $2^{\circ}$, $2.5^{\circ}$) to verify the accuracy of dose delivery by using 2D array ionization chamber (I'mRT Matrixx, IBA Dosimetry, Germany) and MultiCube Phantom(IBA Dosimetry, Germany). Result from the data, gamma index was evaluated. Results : The mean values of pitch and roll rotational setup error were $0.9^{\circ}{\pm}0.7$, $0.5^{\circ}{\pm}0.6$. The maximum values of them were $2.8^{\circ}$, $2.0^{\circ}$. All of the minimum values were zero. The mean values of gamma pass rate at four different pitch angles ($1^{\circ}$, $1.5^{\circ}$, $2^{\circ}$, $2.5^{\circ}$) were 97.75%, 96.65%, 94.38% and 90.91%. The mean values of gamma pass rate at four different roll angles ($1^{\circ}$, $1.5^{\circ}$, $2^{\circ}$, $2.5^{\circ}$) were 93.68%, 93.05%, 87.77% and 84.96%. when the same angles ($1^{\circ}$, $1.5^{\circ}$, $2^{\circ}$) of pitch and roll were applied simultaneously, The mean values of each angle were 94.90%, 92.37% and 87.88%, respectively. Conclusion : As a result of this study, it was able to recognize that the accuracy of dose delivered is lowered gradually as pitch and roll increases. In order to increase the accuracy of delivered dose, therefore, it is recommended to perform IGRT or correct patient's position in the pitch and roll direction, to improve the quality of treatment.