Juhun Lee
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Juhun Lee has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Dae Gy Hong
Department of Obstetrics and Gynecology, School of Medicine, Kyungpook National University, Daegu, Republic of Korea
4 shared publications - Hyun Jung Lee
Department of Obstetrics and Gynecology, School of Medicine, Kyungpook National University, Daegu, Korea
4 shared publications - Gun Oh Chong
Clinical Omics Institute, Kyungpook National University
2 shared publications - Yoon Hee Lee
Department of Obstetrics and Gynecology, Kyungpook National University Chilgok Hospital, School of Medicine, Kyungpook National University, Daegu 41404, Republic of Korea
2 shared publications - Dae-Hyung Lee
Department of Obstetrics and Gynecology, Yeungnam University Medical Center, 170, Hyeonchung-ro, Nam-gu, Daegu 42415, Republic of Korea
1 shared publication - Dong Ja Kim
Department of Forensic Medicine, School of Medicine, Kyungpook National University, Daegu, Korea
1 shared publication
Latest Publications
- Open Access
- 7 May 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Adnexal torsion diagnosis framework with CT-based adaptive preprocessing and deep neural networks
Adnexal cystic torsion is a gynecological emergency that requires prompt and accurate diagnosis followed by immediate surgical intervention to preserve ovarian function. Although ultrasound or computed tomography (CT) are commonly employed, their diagnostic accuracy is often limited by image quality and interobserver variability. This study aimed to evaluate the feasibility of applying deep learning models for the automated diagnosis of adnexal torsion using abdominopelvic CT. We retrospectively collected 1,191 CT scans from 514 women at two tertiary hospitals in Korea. The disease group included 259 surgically confirmed cases of adnexal torsion, while the control group comprised 255 cases of adnexal cysts without torsion, evaluated during diagnostic exploration for acute abdominal pain. Preprocessing included histogram-based clustering of Hounsfield unit distributions for adaptive windowing and anatomical region selection using the pretrained segmentation model TotalSegmentator. We trained both slice-based 2D multiple instance learning frameworks and fully volumetric 3D convolutional neural networks were using stratified five-fold cross-validation. Among the models tested, the 3D EfficientNet architecture demonstrated the best performance, with 79.25% AUC, 73.4% accuracy, 73.77% specificity. These findings highlight the potential of deep learning-assisted CT interpretation as a tool in gynecologic emergency settings.
- 7 May 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessAdnexal torsion diagnosis framework with CT-based adaptive preprocessing and deep neural networks
Adnexal cystic torsion is a gynecological emergency that requires prompt and accurate diagnosis followed by immediate surgical intervention to preserve ovarian function. Although ultrasound or computed tomography (CT) are commonly employed, their diagnostic accuracy is often limited by image quality and interobserver variability. This study aimed to evaluate the feasibility of applying deep learning models for the automated diagnosis of adnexal torsion using abdominopelvic CT. We retrospectively collected 1,191 CT scans from 514 women at two tertiary hospitals in Korea. The disease group included 259 surgically confirmed cases of adnexal torsion, while the control group comprised 255 cases of adnexal cysts without torsion, evaluated during diagnostic exploration for acute abdominal pain. Preprocessing included histogram-based clustering of Hounsfield unit distributions for adaptive windowing and anatomical region selection using the pretrained segmentation model TotalSegmentator. We trained both slice-based 2D multiple instance learning frameworks and fully volumetric 3D convolutional neural networks were using stratified five-fold cross-validation. Among the models tested, the 3D EfficientNet architecture demonstrated the best performance, with 79.25% AUC, 73.4% accuracy, 73.77% specificity. These findings highlight the potential of deep learning-assisted CT interpretation as a tool in gynecologic emergency settings.
- Open Access
Faster R-CNN Model for Predicting Endometrial Cancer Recurrence: A Five-Subclass Approach with Texture Analysis and Validation
Purpose: This study investigates the prediction of endometrial cancer (EC) recurrence using MRI-based texture analysis and deep learning. EC recurrence remains a major clinical challenge, and accurate prognostic tools are needed. The study aims to identify significant imaging biomarkers for recurrence risk and evaluate the performance of a Faster R-CNN deep learning model in classifying recurrence-free survival (RFS) and recurrence (REC) groups. Materials and Methods: A total of 112 patients with histopathologically confirmed EC who underwent pre-treatment MRI between 2009 and 2018 were included. MRI texture analysis was performed using histogram and gray-level co-occurrence matrix (GLCM) methods, analyzing perimeter, integrated density, entropy, and homogeneity. A Faster R-CNN deep learning model was trained and tested on five image subclasses, using 20,000 iterations to optimize performance. REC and RFS were classified based on clinical outcomes, and probability scores were calculated to define the REC score. Feature importance was assessed using a Random Forest model. Results: Significant clinical predictors of recurrence included high-grade tumors, FIGO stage III-IV, and cervical invasion. Texture analysis revealed notable differences in perimeter, integrated density, contrast, entropy, and homogeneity between REC and RFS groups. Faster R-CNN achieved an average precision of 0.86 and an AUC of 0.61. Feature importance analysis identified area as the most significant factor, while REC_score was independent of texture-based parameters. Conclusion: MRI texture analysis combined with deep learning provides valuable predictive insights for EC recurrence. The study highlights the potential of integrating machine learning with imaging biomarkers for personalized risk assessment, offering a promising approach for improving EC patient management..
Open AccessFaster R-CNN Model for Predicting Endometrial Cancer Recurrence: A Five-Subclass Approach with Texture Analysis and Validation
Purpose: This study investigates the prediction of endometrial cancer (EC) recurrence using MRI-based texture analysis and deep learning. EC recurrence remains a major clinical challenge, and accurate prognostic tools are needed. The study aims to identify significant imaging biomarkers for recurrence risk and evaluate the performance of a Faster R-CNN deep learning model in classifying recurrence-free survival (RFS) and recurrence (REC) groups. Materials and Methods: A total of 112 patients with histopathologically confirmed EC who underwent pre-treatment MRI between 2009 and 2018 were included. MRI texture analysis was performed using histogram and gray-level co-occurrence matrix (GLCM) methods, analyzing perimeter, integrated density, entropy, and homogeneity. A Faster R-CNN deep learning model was trained and tested on five image subclasses, using 20,000 iterations to optimize performance. REC and RFS were classified based on clinical outcomes, and probability scores were calculated to define the REC score. Feature importance was assessed using a Random Forest model. Results: Significant clinical predictors of recurrence included high-grade tumors, FIGO stage III-IV, and cervical invasion. Texture analysis revealed notable differences in perimeter, integrated density, contrast, entropy, and homogeneity between REC and RFS groups. Faster R-CNN achieved an average precision of 0.86 and an AUC of 0.61. Feature importance analysis identified area as the most significant factor, while REC_score was independent of texture-based parameters. Conclusion: MRI texture analysis combined with deep learning provides valuable predictive insights for EC recurrence. The study highlights the potential of integrating machine learning with imaging biomarkers for personalized risk assessment, offering a promising approach for improving EC patient management..
- Open Access
- 9 May 2024
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 13 (10) , 2791
Primary Solid Pseudopapillary Tumor of the Ovary: A Case Report and Review of the Literature
Introduction: Solid pseudopapillary neoplasms (SPNs) are rare and mainly originate from the pancreas. SPNs originating from the ovary (SPN-O) are extremely rare, and only 13 cases have been reported in the English literature since 2010. Case: We report a 31-year-old woman with SPN-O accompanied by multiple metastases in the abdominal cavity. The patient underwent staging surgery and cytoreduction. Furthermore, the multidisciplinary board decided on adjuvant chemotherapy with an FP regimen (fluorouracil plus cisplatin) because a microscopic metastasis was discovered in the peritoneum near the appendix. Next-generation sequencing showed some pathologic mutations of oncogenes/cancer-associated genes, including CTNNB1 and TP53. This is the fourteenth case of SPN-O and the first one to demonstrate the TP53 pathogenic mutant variant in SPN-O. The patient showed 8 months of disease-free survival until February 2024. Conclusion: The combination of R0 cytoreduction with FOLFIRI chemotherapy appears to be an effective and feasible treatment option.
- 9 May 2024
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 13 (10) , 2791
Open AccessPrimary Solid Pseudopapillary Tumor of the Ovary: A Case Report and Review of the Literature
Introduction: Solid pseudopapillary neoplasms (SPNs) are rare and mainly originate from the pancreas. SPNs originating from the ovary (SPN-O) are extremely rare, and only 13 cases have been reported in the English literature since 2010. Case: We report a 31-year-old woman with SPN-O accompanied by multiple metastases in the abdominal cavity. The patient underwent staging surgery and cytoreduction. Furthermore, the multidisciplinary board decided on adjuvant chemotherapy with an FP regimen (fluorouracil plus cisplatin) because a microscopic metastasis was discovered in the peritoneum near the appendix. Next-generation sequencing showed some pathologic mutations of oncogenes/cancer-associated genes, including CTNNB1 and TP53. This is the fourteenth case of SPN-O and the first one to demonstrate the TP53 pathogenic mutant variant in SPN-O. The patient showed 8 months of disease-free survival until February 2024. Conclusion: The combination of R0 cytoreduction with FOLFIRI chemotherapy appears to be an effective and feasible treatment option.
- Open Access
- 1 January 2024
- Published by XMLink in Investigative Magnetic Resonance Imaging
- Vol. 28 (2) , 76-86
Texture Analysis in Preoperative Magnetic Resonance Imaging for Assessing Recurrence Risk in Endometrial Cancer After Hysterectomy
- 1 January 2024
- Published by XMLink in Investigative Magnetic Resonance Imaging
- Vol. 28 (2) , 76-86
Open AccessTexture Analysis in Preoperative Magnetic Resonance Imaging for Assessing Recurrence Risk in Endometrial Cancer After Hysterectomy
- Open Access
- 30 November 2023
- Published by Yeungnam University College of Medicine in Journal of Yeungnam Medical Science
- Vol. 40 (Suppl) , S65-S72
Comparison of serum anti-Müllerian hormone between unilateral and bilateral ovarian endometriomas during follicular, luteal, and random menstrual phases: a retrospective study
Background: Over the last two decades, serum levels of anti-Müllerian hormone (AMH) have been shown to be reliable markers of ovarian reserve. This study aimed to compare baseline serum AMH levels and well-controlled clinical factors between patients with unilateral and bilateral ovarian endometriomas during the menstrual phase.Methods: We conducted a retrospective study. We enrolled 136 patients aged 18 to 36 years who were diagnosed with unilateral or bilateral ovarian endometriomas. Serum AMH levels of all patients and their latest two to three menstrual cycles were measured before surgery for ovarian endometriomas. The latest menstrual cycle length ranged from 26 to 30 days. Patients with irregular menstruation, a recent medication history of hormonal drugs other than oral contraceptive pills, a previous history of ovarian surgery, or any medical history influencing ovarian function were excluded. Results: Of the 136 patients, 76 (55.9%) had unilateral ovarian endometriomas and 60 (44.1%) had bilateral ovarian endometriomas. Serum AMH levels were not significantly different between the two groups in the follicular phase, luteal phase, or at any random time point. Conclusion: Serum AMH levels were not significantly different between unilateral and bilateral ovarian endometriomas in the follicular and luteal phases, or at any random time during the menstrual cycle when various confounding factors were excluded.
- 30 November 2023
- Published by Yeungnam University College of Medicine in Journal of Yeungnam Medical Science
- Vol. 40 (Suppl) , S65-S72
Open AccessComparison of serum anti-Müllerian hormone between unilateral and bilateral ovarian endometriomas during follicular, luteal, and random menstrual phases: a retrospective study
Background: Over the last two decades, serum levels of anti-Müllerian hormone (AMH) have been shown to be reliable markers of ovarian reserve. This study aimed to compare baseline serum AMH levels and well-controlled clinical factors between patients with unilateral and bilateral ovarian endometriomas during the menstrual phase.Methods: We conducted a retrospective study. We enrolled 136 patients aged 18 to 36 years who were diagnosed with unilateral or bilateral ovarian endometriomas. Serum AMH levels of all patients and their latest two to three menstrual cycles were measured before surgery for ovarian endometriomas. The latest menstrual cycle length ranged from 26 to 30 days. Patients with irregular menstruation, a recent medication history of hormonal drugs other than oral contraceptive pills, a previous history of ovarian surgery, or any medical history influencing ovarian function were excluded. Results: Of the 136 patients, 76 (55.9%) had unilateral ovarian endometriomas and 60 (44.1%) had bilateral ovarian endometriomas. Serum AMH levels were not significantly different between the two groups in the follicular phase, luteal phase, or at any random time point. Conclusion: Serum AMH levels were not significantly different between unilateral and bilateral ovarian endometriomas in the follicular and luteal phases, or at any random time during the menstrual cycle when various confounding factors were excluded.
- Open Access
- 30 November 2023
- Published by Yeungnam University College of Medicine in Journal of Yeungnam Medical Science
- Vol. 40 (Suppl) , S87-S92
Primary diffuse large B-cell lymphoma of the vulva: a case report
Diffuse large B-cell lymphoma (DLBCL) is a subtype of non-Hodgkin lymphoma (NHL) and is estimated to account for approximately 30% of all NHL cases. NHL can also occur in the female genital tract and accounts for approximately 1.5% of all NHL cases. Many doctors have difficulty diagnosing or treating vulvar DLBCL because of its very low prevalence. A 55-year-old woman presented with a solid mass on the right side of the vulva. No significantly enlarged lymph nodes were observed in the inguinal region. She underwent excisional biopsy at our institution. DLBCL was diagnosed based on histological examination. According to the Hans algorithm, the lesion was diagnosed as a non-germinal center B-cell-like subtype. The patient was referred to a hematologic oncologist. The disease stage was classified as IE according to the Ann Arbor staging classification. The patient received four cycles of chemotherapy with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisolone and localized radiation therapy with 36 Gy in 20 fractions. She showed complete remission and maintained this status on the latest computed tomography scan. Gynecologists should rule out lymphoma in patients presenting with a vulvar mass.
- 30 November 2023
- Published by Yeungnam University College of Medicine in Journal of Yeungnam Medical Science
- Vol. 40 (Suppl) , S87-S92
Open AccessPrimary diffuse large B-cell lymphoma of the vulva: a case report
Diffuse large B-cell lymphoma (DLBCL) is a subtype of non-Hodgkin lymphoma (NHL) and is estimated to account for approximately 30% of all NHL cases. NHL can also occur in the female genital tract and accounts for approximately 1.5% of all NHL cases. Many doctors have difficulty diagnosing or treating vulvar DLBCL because of its very low prevalence. A 55-year-old woman presented with a solid mass on the right side of the vulva. No significantly enlarged lymph nodes were observed in the inguinal region. She underwent excisional biopsy at our institution. DLBCL was diagnosed based on histological examination. According to the Hans algorithm, the lesion was diagnosed as a non-germinal center B-cell-like subtype. The patient was referred to a hematologic oncologist. The disease stage was classified as IE according to the Ann Arbor staging classification. The patient received four cycles of chemotherapy with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisolone and localized radiation therapy with 36 Gy in 20 fractions. She showed complete remission and maintained this status on the latest computed tomography scan. Gynecologists should rule out lymphoma in patients presenting with a vulvar mass.
- Open Access
- 24 September 2023
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 12 (19) , 6155
Do Concurrent Multiple Infections with High-Risk HPVs Carry a More Malignant Potential than a Single Infection in the Uterine Cervix?
The high-risk human papillomavirus (HR-HPV) has been known as the most important carcinogen in uterine cervical carcinoma. However, there is limited evidence of the malignant potential of these concurrent multiple infections. This study included women who had undergone cervical conization. They underwent an HPV test by cervical swab within 12 months before the surgery. They were divided into two groups: one with a single infection with HR-HPV16 and the other with concurrent multiple infections with HR-HPVs, including genotype 16. Pathologic examination classified cases as CIS+ to assess and compare the malignant potential in both groups, including carcinoma in situ (CIS) and invasive carcinoma. Of the 220 patients infected with HR-HPV16, the single infection group consisted of 120 patients (54.5%), whereas the concurrent multiple infections consisted of 100 (45.5%) patients. The rates of HSIL were significantly higher in the concurrent multiple infection group. However, the odds ratio for CIS+ did not show a significant difference between both groups (1.417, 95% CI = 0.831–2.414, p = 0.200). The malignant potential was not significantly different between concurrent multiple infections with HR-HPVs, including 16, and a single infection with 16 in Korean women.
- 24 September 2023
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 12 (19) , 6155
Open AccessDo Concurrent Multiple Infections with High-Risk HPVs Carry a More Malignant Potential than a Single Infection in the Uterine Cervix?
The high-risk human papillomavirus (HR-HPV) has been known as the most important carcinogen in uterine cervical carcinoma. However, there is limited evidence of the malignant potential of these concurrent multiple infections. This study included women who had undergone cervical conization. They underwent an HPV test by cervical swab within 12 months before the surgery. They were divided into two groups: one with a single infection with HR-HPV16 and the other with concurrent multiple infections with HR-HPVs, including genotype 16. Pathologic examination classified cases as CIS+ to assess and compare the malignant potential in both groups, including carcinoma in situ (CIS) and invasive carcinoma. Of the 220 patients infected with HR-HPV16, the single infection group consisted of 120 patients (54.5%), whereas the concurrent multiple infections consisted of 100 (45.5%) patients. The rates of HSIL were significantly higher in the concurrent multiple infection group. However, the odds ratio for CIS+ did not show a significant difference between both groups (1.417, 95% CI = 0.831–2.414, p = 0.200). The malignant potential was not significantly different between concurrent multiple infections with HR-HPVs, including 16, and a single infection with 16 in Korean women.
- Open Access
- 7 October 2022
- Published by Wolters Kluwer Health in Medicine
- Vol. 101 (40) , e30977
Serum anti-Müllerian hormone recovery after ovarian cystectomy for endometriosis: A retrospective study among Korean women
Serum anti-Müllerian hormone (sAMH) is a useful marker of ovarian reserve. In many patients, the sAMH levels gradually recover after falling to the lowest level after surgery (nadir phenomenon). This study aimed to analyze the factors related to sAMH recovery from the nadir after an ovarian cystectomy for endometriosis among Korean women. A total of 159 patients with ovarian endometriosis were included in the study. The sAMH levels were measured before surgery and at least twice within 12 months after the surgery. The patients were divided into two groups: those with recovery (nadir group) and those without recovery (reduction group). Postoperative recovery of the sAMH levels from the nadir was not related to the stage and bilaterality of the lesion, surgical methods, such as robot-assisted or laparoscopic surgery, or surgical time. In the nadir group, the level of preoperative cancer antigen 125 was significantly higher, and it decreased significantly after the surgery than in the reduction group (P = .02 and P = .02). Additionally, the postoperative C-reactive protein (CRP) level was significantly higher in the nadir group, and it increased significantly after the surgery than in the reduction group (P = .03 and P = .04). The increasing degree of perioperative CRP level showed a cutoff value on the receiver operating characteristic curve (0.735 mg/dL, area under curve = 0.604; P = .04). Increased serum CRP levels after surgery are significantly related to the recovery of sAMH levels from the nadir. Therefore, postoperative serum CRP level could be used as a marker to predict the sAMH nadir after surgery.
- 7 October 2022
- Published by Wolters Kluwer Health in Medicine
- Vol. 101 (40) , e30977
Open AccessSerum anti-Müllerian hormone recovery after ovarian cystectomy for endometriosis: A retrospective study among Korean women
Serum anti-Müllerian hormone (sAMH) is a useful marker of ovarian reserve. In many patients, the sAMH levels gradually recover after falling to the lowest level after surgery (nadir phenomenon). This study aimed to analyze the factors related to sAMH recovery from the nadir after an ovarian cystectomy for endometriosis among Korean women. A total of 159 patients with ovarian endometriosis were included in the study. The sAMH levels were measured before surgery and at least twice within 12 months after the surgery. The patients were divided into two groups: those with recovery (nadir group) and those without recovery (reduction group). Postoperative recovery of the sAMH levels from the nadir was not related to the stage and bilaterality of the lesion, surgical methods, such as robot-assisted or laparoscopic surgery, or surgical time. In the nadir group, the level of preoperative cancer antigen 125 was significantly higher, and it decreased significantly after the surgery than in the reduction group (P = .02 and P = .02). Additionally, the postoperative C-reactive protein (CRP) level was significantly higher in the nadir group, and it increased significantly after the surgery than in the reduction group (P = .03 and P = .04). The increasing degree of perioperative CRP level showed a cutoff value on the receiver operating characteristic curve (0.735 mg/dL, area under curve = 0.604; P = .04). Increased serum CRP levels after surgery are significantly related to the recovery of sAMH levels from the nadir. Therefore, postoperative serum CRP level could be used as a marker to predict the sAMH nadir after surgery.
- Open Access
- 12 August 2022
- Published by Wolters Kluwer Health in Medicine
- Vol. 101 (32) , e29967
Does protracted chemotherapy have an influence on the clinical outcomes in advanced epithelial ovarian cancer?
In epithelial ovarian cancer, first-line adjuvant chemotherapy is necessary, and patients sometimes require protraction; however, there are only a few recent studies to show its influence. In this study, we investigated whether the protraction of the total period of first-line chemotherapy has a negative influence on the survival outcomes. Of the 101 patients we recruited from February 2011 to February 2021, 70 (69.3%) and 31 (30.7%) were classified into the not protracted and protracted groups, respectively. They underwent surgery and adjuvant chemotherapy for epithelial ovarian cancer. Protraction was defined as the overall duration of the first-line chemotherapy being more than 20 days longer than intended. Number of patients who underwent additional treatments such as bevacizumab or poly(adenosine diphosphate ribose) polymerase inhibitors or pembrolizumab was compared between both groups. Kaplan–Meier survival analysis and Cox regression analysis were used for survival outcomes. There was no significant difference for additional treatments. The progression-free survival (PFS) in the total follow-up period in the protracted group was significantly shorter than that in the not protracted group (P = .037); however, the difference in the overall survival between the 2 groups was not significant (P = .223). For the PFS, the hazard ratio of protraction was 1.646 in the univariate analysis (95% confidence interval, 1.020–2.658; P = .041). Excessive protraction of chemotherapy over 20 days or more can result in significantly shorter PFS within 5 years. A better therapeutic strategy is required for patients requiring protracted first-line chemotherapy in advanced epithelial ovarian cancer.
- 12 August 2022
- Published by Wolters Kluwer Health in Medicine
- Vol. 101 (32) , e29967
Open AccessDoes protracted chemotherapy have an influence on the clinical outcomes in advanced epithelial ovarian cancer?
In epithelial ovarian cancer, first-line adjuvant chemotherapy is necessary, and patients sometimes require protraction; however, there are only a few recent studies to show its influence. In this study, we investigated whether the protraction of the total period of first-line chemotherapy has a negative influence on the survival outcomes. Of the 101 patients we recruited from February 2011 to February 2021, 70 (69.3%) and 31 (30.7%) were classified into the not protracted and protracted groups, respectively. They underwent surgery and adjuvant chemotherapy for epithelial ovarian cancer. Protraction was defined as the overall duration of the first-line chemotherapy being more than 20 days longer than intended. Number of patients who underwent additional treatments such as bevacizumab or poly(adenosine diphosphate ribose) polymerase inhibitors or pembrolizumab was compared between both groups. Kaplan–Meier survival analysis and Cox regression analysis were used for survival outcomes. There was no significant difference for additional treatments. The progression-free survival (PFS) in the total follow-up period in the protracted group was significantly shorter than that in the not protracted group (P = .037); however, the difference in the overall survival between the 2 groups was not significant (P = .223). For the PFS, the hazard ratio of protraction was 1.646 in the univariate analysis (95% confidence interval, 1.020–2.658; P = .041). Excessive protraction of chemotherapy over 20 days or more can result in significantly shorter PFS within 5 years. A better therapeutic strategy is required for patients requiring protracted first-line chemotherapy in advanced epithelial ovarian cancer.