https://scholarworks.bwise.kr/sch/handle/2021.sw.sch/1071 2026-04-04T14:39:55Z https://scholarworks.bwise.kr/sch/handle/2021.sw.sch/27310 Title: Systematic construction of composite radiation therapy dataset using automated data pipeline for prognosis prediction Authors: Lim, June Hyuck; Kim, Seonhwa; Park, Jun Hyeong; Kim, Chul-Ho; Choi, Jeong-Seok; Chang, Jae Won; Kim, Sup; Park, Il-Seok; Ha, Boram; Jo, In Young; Byeon, Hyung Kwon; Park, Ki Nam; Kim, Han Su; Jung, Soo Yeon; Heo, Jaesung Abstract: Background: Existing research on medical data has primarily focused on single time-points or single-modality data. This study aims to collect all data generated during radiotherapy comprehensively to improve the treatment and prognosis of patients with malignant tumors. Methods: The data collected from each medical institution were transmitted to the lead organization, where they underwent a file integrity check and were processed using a data pipeline. The key metadata of the collected data were compiled into a database, which were examined by data analysts to identify outliers based on theoretical and institution-specific characteristics. Appropriate filters were applied and the filtered data were subsequently reviewed by artificial intelligence (AI)-based models and researchers for radiotherapy organ slides. Finally, they were annotated by specialists. Results: The final dataset included 30,136 three-dimensional cone-beam computed tomography scans and 5,019 tabular data entries collected from 5,019 patients. It comprised 2,043,162 Digital Imaging and Communications in Medicine-format files with a total file size of 832 GB. Quality verification of the data using AI models revealed high classification performance for most organs, with relatively poor performance for the rectum. Overall, the macro AUROC value was 0.947. Conclusions: This study implemented an automated data pipeline and AI-based verification to enhance the quality of collected radiotherapy data. The constructed dataset can be utilized for various types of future research and is expected to contribute to the improvement of radiotherapy efficiency. 2025-03-01T00:00:00Z https://scholarworks.bwise.kr/sch/handle/2021.sw.sch/27057 Title: Does Intralesional Steroid Injection Effectively Mitigate Vocal Fold Scarring in A Rabbit Model? Authors: Jeong, Jun-Yeong; Thapa, Samjhana; Lee, Seung-Won Abstract: ObjectivesTo assess the efficacy of intralesional steroid treatment in preventing vocal fold scarring following vocal fold surgery using a rabbit model.MethodsThe research involved 42 male New Zealand white rabbits. Fourteen rabbits underwent vocal fold scar surgery using a 532nm laser and served as controls (control group). The remaining rabbits were divided into two groups of 14: one group received vocal fold scar surgery followed by dexamethasone injection (Dexa group) and the other received the same surgery followed by triamcinolone injection (Triam group). Four weeks after surgery, histological examinations and high-speed video analyses of vocal fold vibration were conducted. The maximum amplitude of vibration was the primary measure for assessing vocal fold function. In addition, real-time polymerase chain reaction (PCR) studies were undertaken to analyze scar regeneration and remodeling.ResultsThe maximum amplitude differences were notably higher in the Dexa and Triam groups than in controls. Histologically, the collagen density (CD) ratios in both the Dexa and Triam groups were significantly reduced compared with controls. Real-time PCR analysis indicated marked elevations of Has-2 and Mmp-9 in the Dexa and Triam groups relative to controls.ConclusionsIntralesional steroid injections after vocal fold surgery are effective for reducing vocal fold scarring in a rabbit model.Level of EvidenceNA Laryngoscope, 2024 This article evaluated the preventive effects of steroid injections on vocal fold scarring at transcriptional, histological, and functional levels.image 2025-01-01T00:00:00Z https://scholarworks.bwise.kr/sch/handle/2021.sw.sch/27680 Title: A deep learning algorithm model to automatically score and grade obstructive sleep apnea in adult polysomnography Authors: Park, Marn Joon; Choi, Ji Ho; Kim, Shin Young; Ha, Tae Kyoung Abstract: Objective Polysomnography (PSG) is unique in diagnosing sleep disorders, notably obstructive sleep apnea (OSA). Despite its advantages, manual PSG data grading is time-consuming and laborious. Thus, this research evaluated a deep learning-based automated scoring system for respiratory events in sleep-disordered breathing patients.Methods A total of 1000 case PSG data were enrolled to develop a deep learning algorithm. Of the 1000 data, 700 were distributed for training, 200 for validation, and 100 for testing. The respiratory events scoring deep learning model is composed of five sequential layers: an initial layer of perceptrons, followed by three consecutive layers of long short-term memory cells, and ultimately, an additional two layers of perceptrons.Results The PSG data of 100 patients (simple snoring, mild, moderate, and severe OSA; n = 25 in each group) were selected for validation and testing of the deep learning model. The algorithm demonstrated high sensitivity (95% CI: 98.06-98.51) and specificity (95% CI: 95.46-97.79) across all OSA severities in detecting apnea/hypopnea events, compared to manual PSG analysis. The deep learning model's area under the curve values for predicting OSA in apnea-hypopnea index >= 5, 15, and 30 groups were 0.9402, 0.9388, and 0.9442, respectively, showing no significant differences between each group.Conclusion The deep learning algorithm employed in our study showed high accuracy in identifying apnea/hypopnea episodes and assessing the severity of OSA, suggesting the potential for enhancing both the efficiency and accuracy of automated respiratory event scoring in PSG through advanced deep learning techniques. 2024-10-01T00:00:00Z https://scholarworks.bwise.kr/sch/handle/2021.sw.sch/26382 Title: Long-term follow-up results of facial nerve schwannoma with good facial nerve function: a multicenter study Authors: Cho, Young Sang; Lee, Jong Dae; Cho, Yang-Sun; Lee, Jun Ho; Seo, Hee Won; Gwak, Jang Wook; Moon, In Seok; Choi, Jin Woong; Han, Gyu Cheol; Koo, Ja-Won; Chung, Jong Woo Abstract: PurposeFacial nerve schwannomas (FNSs) are rare intracranial tumors, and the optimal management of these tumors remains unclear. We investigated the long-term follow-up results of FNS with good facial nerve function.MethodsAt nine medical centers in the Korean Facial Nerve Study Group, 43 patients undergoing observation periods longer than 12 months for FNS with good facial nerve function (House-Brackmann grade <= II) were enrolled, and clinical and radiographic data were obtained for these cases.ResultsThe mean follow-up period was 63 months. In the majority of cases, tumors involved multiple segments (81.4%) and only eight cases were confined to a single site. There were no cases where the tumor was confined to the extratemporal region. Tumor size increased slightly, with an average estimated change of 0.48 mm/year. Twenty (46.5%) of 43 patients showed no change in tumor size. Seven patients (16.3%) showed worsening House-Brackmann (H-B) grade, of which two patients deteriorated from H-B grade I to II, four worsened to grade III, and one deteriorated to grade IV. The remaining 36 patients (83.7%) showed no change in facial nerve function. There was no difference in H-B grade according to tumor size at the time of diagnosis or change in tumor size.ConclusionWe conducted a large-scale observational study of FNS with good facial nerve function. Our study showed that many patients maintained facial nerve function during long-term follow-up. Conservative management with regular examination and imaging can be an appropriate option for managing FNS with good facial nerve function. 2024-09-01T00:00:00Z