Jeong Sub Lee

Mr. Jeong Sub Lee

Medical Device Development, Seoul National University College of Medicine, Seoul, Republic of Korea
orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2025 in American Journal of Roentgenology
Verified
Journal Article
Closed access
SJSowon Jang
Sowon Jang
JKJunghoon Kim
Junghoon Kim
Jeong Sub LeeJeong Sub Lee
Jeong Sub Lee
YJYounbeom Jeong
Younbeom Jeong
JNJu Gang Nam
Ju Gang Nam
KLKyung Won Lee
Kyung Won Lee
Background: Studies of artificial intelligence (AI) for lung nodule detection on CT have primarily been conducted in investigational settings and/or focused on lung cancer screening. Objective: To evaluate the impact of AI assistance on radiologists' diagnostic performance for detecting lung metastases on chest CT in patients with colorectal cancer (CRC) in real-world clinical practice and to assess the clinical utility of AI assistance in this setting. Methods: This retrospective study included patients with CRC who underwent chest CT as surveillance for lung metastasis from May 2020 to December 2020 (conventional interpretation) or May 2022 to December 2022 (AI-assisted interpretation). Between periods, the institution implemented a commercial AI lung nodule detection system. During the second period, radiologists interpreted examinations concurrently with AI-generated reports, using clinical judgment regarding whether to report AI-detected nodules. The reference standard for metastasis incorporated pathologic and clinical follow-up criteria. Diagnostic performance (sensitivity, specificity, accuracy), and clinical utility (diagnostic yield, false-referral rate, management changes after positive reports) were compared between groups based on clinical radiology reports. Net benefit was estimated using decision curve analysis equation. Standalone AI interpretation was evaluated. Results: The conventional interpretation group included 647 patients (mean age, 64±11 years; 394 men, 253 women; metastasis prevalence, 4.3%); AI-assisted interpretation group included 663 patients (mean age, 63±12 years; 381 men, 282 women; metastasis prevalence, 4.4%). The AI-assisted interpretation group compared with the conventional interpretation group showed higher sensitivity (72.4% vs 32.1%; p=.008), accuracy (98.5% vs 96.0%; p=.005), and frequency of management changes (55.2% vs 25.0%, p=.02), without significant difference in specificity (99.7% vs 98.9%; p=.11), diagnostic yield (3.2% vs 1.4%, p=.30) or false-referral rate (0.3% vs 1.1%, p=.10). AI-assisted interpretation had positive estimated net benefit across outcome ratios. Standalone AI correctly detected metastasis in 24 of 29 patients but had 381 false-positive detections in 634 patients without metastasis; only one AI false-positive was reported as positive by interpretating radiologists. Conclusion: AI assistance yielded increased sensitivity, accuracy, and frequency of management changes, without significantly changed specificity. False-positive AI results minimally impacted radiologists' interpretations. Clinical Impact: The findings support clinical utility of AI assistance for CRC metastasis surveillance.

24 October 2025 in Journal of Thoracic Imaging
Verified
Journal Article
Closed access
SJSowon Jang
Sowon Jang
MKMinseon Kim
Minseon Kim
Jeong Sub LeeJeong Sub Lee
Jeong Sub Lee
SYSung Hyun Yoon
Sung Hyun Yoon
JKJunghoon Kim
Junghoon Kim
KLKyung Won Lee
Kyung Won Lee
Purpose: To develop and validate a nomogram to predict hemoptysis after percutaneous transthoracic needle biopsy (PTNB) by integrating clinical and radiologic data, facilitating pre-biopsy decision-making. Materials and Methods: This single-center, retrospective cohort study included 1383 patients who underwent 1389 PTNB procedures between 2020 and 2022. The participants were randomly allocated to the training and validation cohorts. Logistic regression was performed to discern the independent predictors of hemoptysis within the clinical and radiologic variables. A nomogram was developed based on pre-biopsy variables obtained before the biopsy, and its performance was subsequently evaluated. The goodness of fit of the nomogram was compared with that of another model, which integrated pre-biopsy and post-biopsy variables. Results: Among the 1389 procedures, hemoptysis was observed in 128 (9.2%) cases. Current smoking status, lesion size of <25 mm, consolidation-type lesions, presence of the computed tomography bronchus sign, and perilesional vascularity were independent predictors of PTNB-related hemoptysis. The nomogram based on the pre-biopsy variables showed fair discrimination abilities (area under the receiver operating characteristic curve = 0.79 and 0.76 in the training and validation cohorts, respectively) and strong calibration agreement in the training and validation cohorts. The model fit was good in both cohorts ( P = 0.41 and 0.55 in the training and validation cohorts, respectively). No significant difference was observed in the model fit between the pre-biopsy nomogram and the model incorporating pre-biopsy and post-biopsy variables ( P = 0.88). Conclusion: The proposed nomogram utilizing pre-biopsy variables could predict hemoptysis before PTNB.

9 February 2026 in Diagnostics
Verified
Journal Article
Open access
PDF file
Jeong Sub LeeJeong Sub Lee
Jeong Sub Lee
EHEuijin Hwang
Euijin Hwang
CJChanggyun Jin
Changgyun Jin
Kyong Joon LeeKyong Joon Lee
Kyong Joon Lee
YCYe Ra Choi
Ye Ra Choi
SCSang Il Choi
Sang Il Choi
Background: High false-positive rates remain a significant challenge in the automated detection of pulmonary embolism (PE) using Computed Tomography Pulmonary Angiography (CTPA). This study evaluated the additional value of a False-Positive Reduction (FPR) module integrated into a Region Proposal Network (RPN). Methods: A retrospective analysis of 303 CTPA scans (163 PE-positive and 140 PE-negative) was conducted from a single tertiary institution. Both models were additionally validated on an independent external cohort of 100 CTPA scans (50 PE-positive and 50 PE-negative) from the RSNA PE Challenge dataset. The diagnostic performance of the one-stage RPN-only model was compared with that of a two-stage Modified Mask R-CNN (Region-based Convolutional Neural Network) incorporating the FPR module. Results: The Modified Mask R-CNN exhibited significant improvement in terms of specificity. The false-positive rate per scan decreased by 31% in comparison to the RPN-only model. Although there was a slight reduction in patient-level sensitivity, the Positive Predictive Value significantly increased by 10.5%. Additionally, patient-level specificity for emboli with a volume ≥ 1000 mm3 increased, reflecting a 7.4% relative improvement in detecting clinically significant emboli. Conclusions: The Modified Mask R-CNN significantly reduced false positives while maintaining high sensitivity over a region proposal network.

20 May 2026 in PLOS ONE
Verified
Journal Article
Open access
JKJunghoon Kim
Junghoon Kim
Jeong Sub LeeJeong Sub Lee
Jeong Sub Lee
CCChoong Guen Chee
Choong Guen Chee
JWJihoon Woo
Jihoon Woo
YKYoungjune Kim
Youngjune Kim
SCSang Il Choi
Sang Il Choi
Little information is available regarding the factors that contribute to the occurrence of right innominate vein stenosis in patients with chemoports. Thus, the aims of this study were to evaluate the prevalence of chemoport-related right innominate vein stenosis (CR-RIVS) in patients with colorectal cancer and to identify factors associated with increased CR-RIVS risk. This retrospective study included patients with colorectal cancer (age, ≥ 19 years) who underwent their first chemoport insertion via the right internal jugular vein between August 2015 and August 2020. The prevalence of CR-RIVS was evaluated using follow-up contrast-enhanced chest computed tomography post-chemoport insertion. Risk factors were analyzed using Classification and Regression Tree, Kaplan-Meier survival, and Cox regression analyses. A subgroup analysis was also conducted among patients in palliative care to assess the impact of bevacizumab and cetuximab treatment on the prevalence of CR-RIVS. Among 1,442 patients with colorectal cancer (mean age, 62.1 ± 11.8 years; 850 males), the prevalence of CR-RIVS was 6.6%. The median time from chemoport insertion to CR-RIVS occurrence was 405 days (interquartile range, 256–908 days). The Classification and Regression Tree analysis identified palliative settings as the key predictor of CR-RIVS. Age, palliative settings, and a time×palliative setting interaction were identified as statistically significant variables ( p < 0.001). In the 728 palliative patients, bevacizumab treatment was significantly associated with a higher risk of CR-RIVS (hazard ratio = 1.51, 95% confidence interval: 1.25–1.82, P < 0.001), whereas cetuximab was not ( p = 0.85). The CR-RIVS-free probability was lower in patients receiving palliative chemotherapy than in those undergoing adjuvant chemotherapy ( p < 0.001); it was also lower in the bevacizumab-treated patients ( p < 0.001) in the palliative setting subgroup analysis. In summary, the prevalence of CR-RIVS was 6.6% in patients with colorectal cancer. Bevacizumab use showed the strongest association with CR-RIVS occurrence in the palliative subgroup analysis.