Chae Jung Park
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Chae Jung Park has a total of 73 co-authors. Here are the top co-authors by number of shared publications:
- Seung-Koo Lee
Department of Radiology and Research Institute of Radiological Science and Center for Clinical Imaging Data Science, Yonsei University College of Medicine, Seoul, Republic of Korea
8 shared publications - Yae Won Park
Department of Radiology and Research Institute of Radiological Science and Center for Clinical Imaging Data Science, Yonsei University College of Medicine , Seoul
7 shared publications - Jong Hee Chang
Department of Neurosurgery, Yonsei University College of Medicine, Seoul, Republic of Korea
4 shared publications - Phil Hyu Lee
Department of Neurology, Yonsei University College of Medicine, Seoul, Korea
4 shared publications - Se-Hoon Lee
Division of Hematology-Oncology, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
4 shared publications - Seok Jong Chung
Department of Neurology, Yonsei University College of Medicine, Seoul, South Korea
4 shared publications
Latest Publications
- Open Access
- 31 October 2026
- Published by The Korean Movement Disorder Society in Journal of Movement Disorders
- Vol. 19 (4) , 431-441
Quantitative Analysis of Automatically Segmented Perivascular Space in Subcortical White Matter of Parkinson’s Disease: Its Clinical Relevance
Objective To examine the relationship between automatically quantified perivascular space (PVS) extent located in the subcortical white matter (WM), baseline motor deficits and cognitive function, as well as longitudinal motor and cognitive outcomes in patients with Parkinson’s disease (PD).Methods This retrospective cohort study included 270 consecutive patients with newly diagnosed PD. Subcortical WM PVS was automatically segmented using high-resolution T2-weighted images, and automated PVS (aPVS) volume and count were calculated. Multivariate linear regression analyses evaluated the association between subcortical WM aPVS and baseline Unified Parkinson’s Disease Rating Scale Part III, Mini-Mental State Examination, and Montreal Cognitive Assessment scores. Cox regression models assessed the prognostic value of baseline subcortical WM aPVS for the risk of developing freezing of gait (FOG) and dementia.Results Greater subcortical WM aPVS volume, particularly in the frontal and parietal regions, correlated with milder parkinsonian motor deficits and better global cognitive function at baseline. Higher subcortical WM aPVS volume, especially in the frontal region, was linked to a lower risk of FOG (hazard ratio [HR], 0.53; 95% confidence interval [CI] 0.22–0.93; p=0.028). Additionally, greater subcortical WM aPVS volume, especially in the occipital region, was associated with a reduced risk of dementia conversion (HR, 0.38; 95% CI 0.10–0.95; p=0.041) in Cox regression models. Higher subcortical WM aPVS count was linked to better baseline motor symptoms, cognitive function, and lower risk of FOG.Conclusion These findings indicate that larger subcortical WM aPVS volume or higher aPVS count can be an indicator of favorable clinical outcomes in PD.
- 31 October 2026
- Published by The Korean Movement Disorder Society in Journal of Movement Disorders
- Vol. 19 (4) , 431-441
Open AccessQuantitative Analysis of Automatically Segmented Perivascular Space in Subcortical White Matter of Parkinson’s Disease: Its Clinical Relevance
Objective To examine the relationship between automatically quantified perivascular space (PVS) extent located in the subcortical white matter (WM), baseline motor deficits and cognitive function, as well as longitudinal motor and cognitive outcomes in patients with Parkinson’s disease (PD).Methods This retrospective cohort study included 270 consecutive patients with newly diagnosed PD. Subcortical WM PVS was automatically segmented using high-resolution T2-weighted images, and automated PVS (aPVS) volume and count were calculated. Multivariate linear regression analyses evaluated the association between subcortical WM aPVS and baseline Unified Parkinson’s Disease Rating Scale Part III, Mini-Mental State Examination, and Montreal Cognitive Assessment scores. Cox regression models assessed the prognostic value of baseline subcortical WM aPVS for the risk of developing freezing of gait (FOG) and dementia.Results Greater subcortical WM aPVS volume, particularly in the frontal and parietal regions, correlated with milder parkinsonian motor deficits and better global cognitive function at baseline. Higher subcortical WM aPVS volume, especially in the frontal region, was linked to a lower risk of FOG (hazard ratio [HR], 0.53; 95% confidence interval [CI] 0.22–0.93; p=0.028). Additionally, greater subcortical WM aPVS volume, especially in the occipital region, was associated with a reduced risk of dementia conversion (HR, 0.38; 95% CI 0.10–0.95; p=0.041) in Cox regression models. Higher subcortical WM aPVS count was linked to better baseline motor symptoms, cognitive function, and lower risk of FOG.Conclusion These findings indicate that larger subcortical WM aPVS volume or higher aPVS count can be an indicator of favorable clinical outcomes in PD.
- Open Access
- 27 November 2024
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 26, e54641
Development and Validation of a Prediction Model Using Sella Magnetic Resonance Imaging–Based Radiomics and Clinical Parameters for the Diagnosis of Growth Hormone Deficiency and Idiopathic Short Stature: Cross-Sectional, Multicenter Study
Background: Growth hormone deficiency (GHD) and idiopathic short stature (ISS) are the major etiologies of short stature in children. For the diagnosis of GHD and ISS, meticulous evaluations are required, including growth hormone provocation tests, which are invasive and burdensome for children. Additionally, sella magnetic resonance imaging (MRI) is necessary for assessing etiologies of GHD, which cannot evaluate hormonal secretion. Recently, radiomics has emerged as a revolutionary technique that uses mathematical algorithms to extract various features for the quantitative analysis of medical images. Objective: This study aimed to develop a machine learning–based model using sella MRI–based radiomics and clinical parameters to diagnose GHD and ISS. Methods: A total of 293 children with short stature who underwent sella MRI and growth hormone provocation tests were included in the training set, and 47 children who met the same inclusion criteria were enrolled in the test set from different hospitals for this study. A total of 186 radiomic features were extracted from the pituitary glands using a semiautomatic segmentation process for both the T2-weighted and contrast-enhanced T1-weighted image. The clinical parameters included auxological data, insulin-like growth factor-I, and bone age. The extreme gradient boosting algorithm was used to train the prediction models. Internal validation was conducted using 5-fold cross-validation on the training set, and external validation was conducted on the test set. Model performance was assessed by plotting the area under the receiver operating characteristic curve. The mean absolute Shapley values were computed to quantify the impact of each parameter. Results: The area under the receiver operating characteristic curves (95% CIs) of the clinical, radiomics, and combined models were 0.684 (0.590-0.778), 0.691 (0.620-0.762), and 0.830 (0.741-0.919), respectively, in the external validation. Among the clinical parameters, the major contributing factors to prediction were BMI SD score (SDS), chronological age–bone age, weight SDS, growth velocity, and insulin-like growth factor-I SDS in the clinical model. In the combined model, radiomic features including maximum probability from a T2-weighted image and run length nonuniformity normalized from a T2-weighted image added incremental value to the prediction (combined model vs clinical model, P=.03; combined model vs radiomics model, P=.02). The code for our model is available in a public repository on GitHub. Conclusions: Our model combining both radiomics and clinical parameters can accurately predict GHD from ISS, which was also proven in the external validation. These findings highlight the potential of machine learning–based models using radiomics and clinical parameters for diagnosing GHD and ISS.
- 27 November 2024
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 26, e54641
Open AccessDevelopment and Validation of a Prediction Model Using Sella Magnetic Resonance Imaging–Based Radiomics and Clinical Parameters for the Diagnosis of Growth Hormone Deficiency and Idiopathic Short Stature: Cross-Sectional, Multicenter Study
Background: Growth hormone deficiency (GHD) and idiopathic short stature (ISS) are the major etiologies of short stature in children. For the diagnosis of GHD and ISS, meticulous evaluations are required, including growth hormone provocation tests, which are invasive and burdensome for children. Additionally, sella magnetic resonance imaging (MRI) is necessary for assessing etiologies of GHD, which cannot evaluate hormonal secretion. Recently, radiomics has emerged as a revolutionary technique that uses mathematical algorithms to extract various features for the quantitative analysis of medical images. Objective: This study aimed to develop a machine learning–based model using sella MRI–based radiomics and clinical parameters to diagnose GHD and ISS. Methods: A total of 293 children with short stature who underwent sella MRI and growth hormone provocation tests were included in the training set, and 47 children who met the same inclusion criteria were enrolled in the test set from different hospitals for this study. A total of 186 radiomic features were extracted from the pituitary glands using a semiautomatic segmentation process for both the T2-weighted and contrast-enhanced T1-weighted image. The clinical parameters included auxological data, insulin-like growth factor-I, and bone age. The extreme gradient boosting algorithm was used to train the prediction models. Internal validation was conducted using 5-fold cross-validation on the training set, and external validation was conducted on the test set. Model performance was assessed by plotting the area under the receiver operating characteristic curve. The mean absolute Shapley values were computed to quantify the impact of each parameter. Results: The area under the receiver operating characteristic curves (95% CIs) of the clinical, radiomics, and combined models were 0.684 (0.590-0.778), 0.691 (0.620-0.762), and 0.830 (0.741-0.919), respectively, in the external validation. Among the clinical parameters, the major contributing factors to prediction were BMI SD score (SDS), chronological age–bone age, weight SDS, growth velocity, and insulin-like growth factor-I SDS in the clinical model. In the combined model, radiomic features including maximum probability from a T2-weighted image and run length nonuniformity normalized from a T2-weighted image added incremental value to the prediction (combined model vs clinical model, P=.03; combined model vs radiomics model, P=.02). The code for our model is available in a public repository on GitHub. Conclusions: Our model combining both radiomics and clinical parameters can accurately predict GHD from ISS, which was also proven in the external validation. These findings highlight the potential of machine learning–based models using radiomics and clinical parameters for diagnosing GHD and ISS.
- Open Access
- 1 January 2024
- Published by XMLink in Yonsei Medical Journal
- Vol. 65 (5) , 283-292
Diffusion- and Perfusion-Weighted MRI Radiomics for Survival Prediction in Patients with Lower-Grade Gliomas
Park CJ, et al. Yonsei Med J. 2024 Mar;65:e10. https://doi.org/10.3349/ymj.2023.0323
- 1 January 2024
- Published by XMLink in Yonsei Medical Journal
- Vol. 65 (5) , 283-292
Open AccessDiffusion- and Perfusion-Weighted MRI Radiomics for Survival Prediction in Patients with Lower-Grade Gliomas
Park CJ, et al. Yonsei Med J. 2024 Mar;65:e10. https://doi.org/10.3349/ymj.2023.0323
- Open Access
- 16 November 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Development and Validation of a Prediction Model Using Sella Magnetic Resonance Imaging–Based Radiomics and Clinical Parameters for the Diagnosis of Growth Hormone Deficiency and Idiopathic Short Stature: Cross-Sectional, Multicenter Study (Preprint)
BACKGROUND Growth hormone deficiency (GHD) and idiopathic short stature (ISS) are the major etiologies of short stature in children. For the diagnosis of GHD and ISS, meticulous evaluations are required, including growth hormone provocation tests, which are invasive and burdensome for children. Additionally, sella magnetic resonance imaging (MRI) is necessary for assessing etiologies of GHD, which cannot evaluate hormonal secretion. Recently, radiomics has emerged as a revolutionary technique that uses mathematical algorithms to extract various features for the quantitative analysis of medical images. OBJECTIVE This study aimed to develop a machine learning–based model using sella MRI–based radiomics and clinical parameters to diagnose GHD and ISS. METHODS A total of 293 children with short stature who underwent sella MRI and growth hormone provocation tests were included in the training set, and 47 children who met the same inclusion criteria were enrolled in the test set from different hospitals for this study. A total of 186 radiomic features were extracted from the pituitary glands using a semiautomatic segmentation process for both the T2-weighted and contrast-enhanced T1-weighted image. The clinical parameters included auxological data, insulin-like growth factor-I, and bone age. The extreme gradient boosting algorithm was used to train the prediction models. Internal validation was conducted using 5-fold cross-validation on the training set, and external validation was conducted on the test set. Model performance was assessed by plotting the area under the receiver operating characteristic curve. The mean absolute Shapley values were computed to quantify the impact of each parameter. RESULTS The area under the receiver operating characteristic curves (95% CIs) of the clinical, radiomics, and combined models were 0.684 (0.590-0.778), 0.691 (0.620-0.762), and 0.830 (0.741-0.919), respectively, in the external validation. Among the clinical parameters, the major contributing factors to prediction were BMI SD score (SDS), chronological age–bone age, weight SDS, growth velocity, and insulin-like growth factor-I SDS in the clinical model. In the combined model, radiomic features including maximum probability from a T2-weighted image and run length nonuniformity normalized from a T2-weighted image added incremental value to the prediction (combined model vs clinical model, P=.03; combined model vs radiomics model, P=.02). The code for our model is available in a public repository on GitHub. CONCLUSIONS Our model combining both radiomics and clinical parameters can accurately predict GHD from ISS, which was also proven in the external validation. These findings highlight the potential of machine learning–based models using radiomics and clinical parameters for diagnosing GHD and ISS.
- 16 November 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessDevelopment and Validation of a Prediction Model Using Sella Magnetic Resonance Imaging–Based Radiomics and Clinical Parameters for the Diagnosis of Growth Hormone Deficiency and Idiopathic Short Stature: Cross-Sectional, Multicenter Study (Preprint)
BACKGROUND Growth hormone deficiency (GHD) and idiopathic short stature (ISS) are the major etiologies of short stature in children. For the diagnosis of GHD and ISS, meticulous evaluations are required, including growth hormone provocation tests, which are invasive and burdensome for children. Additionally, sella magnetic resonance imaging (MRI) is necessary for assessing etiologies of GHD, which cannot evaluate hormonal secretion. Recently, radiomics has emerged as a revolutionary technique that uses mathematical algorithms to extract various features for the quantitative analysis of medical images. OBJECTIVE This study aimed to develop a machine learning–based model using sella MRI–based radiomics and clinical parameters to diagnose GHD and ISS. METHODS A total of 293 children with short stature who underwent sella MRI and growth hormone provocation tests were included in the training set, and 47 children who met the same inclusion criteria were enrolled in the test set from different hospitals for this study. A total of 186 radiomic features were extracted from the pituitary glands using a semiautomatic segmentation process for both the T2-weighted and contrast-enhanced T1-weighted image. The clinical parameters included auxological data, insulin-like growth factor-I, and bone age. The extreme gradient boosting algorithm was used to train the prediction models. Internal validation was conducted using 5-fold cross-validation on the training set, and external validation was conducted on the test set. Model performance was assessed by plotting the area under the receiver operating characteristic curve. The mean absolute Shapley values were computed to quantify the impact of each parameter. RESULTS The area under the receiver operating characteristic curves (95% CIs) of the clinical, radiomics, and combined models were 0.684 (0.590-0.778), 0.691 (0.620-0.762), and 0.830 (0.741-0.919), respectively, in the external validation. Among the clinical parameters, the major contributing factors to prediction were BMI SD score (SDS), chronological age–bone age, weight SDS, growth velocity, and insulin-like growth factor-I SDS in the clinical model. In the combined model, radiomic features including maximum probability from a T2-weighted image and run length nonuniformity normalized from a T2-weighted image added incremental value to the prediction (combined model vs clinical model, P=.03; combined model vs radiomics model, P=.02). The code for our model is available in a public repository on GitHub. CONCLUSIONS Our model combining both radiomics and clinical parameters can accurately predict GHD from ISS, which was also proven in the external validation. These findings highlight the potential of machine learning–based models using radiomics and clinical parameters for diagnosing GHD and ISS.
- Open Access
- 25 September 2023
- Published by Korean Society of Otorhinolaryngology-Head and Neck Surgery in Korean Journal of Otorhinolaryngology-Head and Neck Surgery
- Vol. 66 (9) , 618-623
A Case of Rapid Growing Myoepithelioma of the Maxillary Sinus
Myoepithelioma is a rare tumor of the salivary glands. Occurrence of myoepithelioma in the nasal cavity is even rarer, and there has been no case wherein the change in the size of the tumor was confirmed by imaging. In this report, a 84-year-old female was found to have a rapidly enlarged tumor on the preoperative image, but she had shown no significant size change in a medical examination six months ago or in the paranasal sinuses view three months ago. The image showed a rapid increase in the size of the cystic component, so the endoscopic medial maxillectomy was performed. There was no recurrence of the tumor at follow-up six months after surgery. Although myoepithelioma is rarely found in the maxillary sinus of the sinonasal cavity, it is important to suspect myoepithelioma and consult a pathologist if it shows a similar pattern to the current case.
- 25 September 2023
- Published by Korean Society of Otorhinolaryngology-Head and Neck Surgery in Korean Journal of Otorhinolaryngology-Head and Neck Surgery
- Vol. 66 (9) , 618-623
Open AccessA Case of Rapid Growing Myoepithelioma of the Maxillary Sinus
Myoepithelioma is a rare tumor of the salivary glands. Occurrence of myoepithelioma in the nasal cavity is even rarer, and there has been no case wherein the change in the size of the tumor was confirmed by imaging. In this report, a 84-year-old female was found to have a rapidly enlarged tumor on the preoperative image, but she had shown no significant size change in a medical examination six months ago or in the paranasal sinuses view three months ago. The image showed a rapid increase in the size of the cystic component, so the endoscopic medial maxillectomy was performed. There was no recurrence of the tumor at follow-up six months after surgery. Although myoepithelioma is rarely found in the maxillary sinus of the sinonasal cavity, it is important to suspect myoepithelioma and consult a pathologist if it shows a similar pattern to the current case.
- Open Access
- 30 August 2023
- Published by Springer Nature in npj Parkinson's Disease
- Vol. 9 (1) , 1-10
An interpretable multiparametric radiomics model of basal ganglia to predict dementia conversion in Parkinson’s disease
Cognitive impairment in Parkinson’s disease (PD) severely affects patients’ prognosis, and early detection of patients at high risk of dementia conversion is important for establishing treatment strategies. We aimed to investigate whether multiparametric MRI radiomics from basal ganglia can improve the prediction of dementia development in PD when integrated with clinical profiles. In this retrospective study, 262 patients with newly diagnosed PD (June 2008–July 2017, follow-up >5 years) were included. MRI radiomic features (n = 1284) were extracted from bilateral caudate and putamen. Two models were developed to predict dementia development: (1) a clinical model—age, disease duration, and cognitive composite scores, and (2) a combined clinical and radiomics model. The area under the receiver operating characteristic curve (AUC) were calculated for each model. The models’ interpretabilities were studied. Among total 262 PD patients (mean age, 68 years ± 8 [standard deviation]; 134 men), 51 (30.4%), and 24 (25.5%) patients developed dementia within 5 years of PD diagnosis in the training (n = 168) and test sets (n = 94), respectively. The combined model achieved superior predictive performance compared to the clinical model in training (AUCs 0.928 vs. 0.894, P = 0.284) and test set (AUCs 0.889 vs. 0.722, P = 0.016). The cognitive composite scores of the frontal/executive function domain contributed most to predicting dementia. Radiomics derived from the caudate were also highly associated with cognitive decline. Multiparametric MRI radiomics may have an incremental prognostic value when integrated with clinical profiles to predict future cognitive decline in PD.
- 30 August 2023
- Published by Springer Nature in npj Parkinson's Disease
- Vol. 9 (1) , 1-10
Open AccessAn interpretable multiparametric radiomics model of basal ganglia to predict dementia conversion in Parkinson’s disease
Cognitive impairment in Parkinson’s disease (PD) severely affects patients’ prognosis, and early detection of patients at high risk of dementia conversion is important for establishing treatment strategies. We aimed to investigate whether multiparametric MRI radiomics from basal ganglia can improve the prediction of dementia development in PD when integrated with clinical profiles. In this retrospective study, 262 patients with newly diagnosed PD (June 2008–July 2017, follow-up >5 years) were included. MRI radiomic features (n = 1284) were extracted from bilateral caudate and putamen. Two models were developed to predict dementia development: (1) a clinical model—age, disease duration, and cognitive composite scores, and (2) a combined clinical and radiomics model. The area under the receiver operating characteristic curve (AUC) were calculated for each model. The models’ interpretabilities were studied. Among total 262 PD patients (mean age, 68 years ± 8 [standard deviation]; 134 men), 51 (30.4%), and 24 (25.5%) patients developed dementia within 5 years of PD diagnosis in the training (n = 168) and test sets (n = 94), respectively. The combined model achieved superior predictive performance compared to the clinical model in training (AUCs 0.928 vs. 0.894, P = 0.284) and test set (AUCs 0.889 vs. 0.722, P = 0.016). The cognitive composite scores of the frontal/executive function domain contributed most to predicting dementia. Radiomics derived from the caudate were also highly associated with cognitive decline. Multiparametric MRI radiomics may have an incremental prognostic value when integrated with clinical profiles to predict future cognitive decline in PD.
- Top Cited Papers
- 30 June 2023
- Published by BMJ in Journal of Neurology, Neurosurgery & Psychiatry
- Vol. 94 (12) , 1047-1055
Association of choroid plexus volume with motor symptoms and dopaminergic degeneration in Parkinson’s disease
Background The choroid plexus (CP) is involved in the clearance of harmful metabolites from the brain, as a part of the glymphatic system. This study aimed to investigate the association between CP volume (CPV), nigrostriatal dopaminergic degeneration and motor outcomes in Parkinson’s disease (PD). Methods We retrospectively searched drug-naïve patients with early-stage PD who underwent dopamine transporter (DAT) scanning and MRI. Automatic CP segmentation was performed, and the CPV was calculated. The relationship between CPV, DAT availability and Unified PD Rating Scale Part III (UPDRS-III) scores was assessed using multivariate linear regression. We performed longitudinal analyses to assess motor outcomes according to CPV. Results CPV was negatively associated with DAT availability in each striatal subregion (anterior caudate, β=−0.134, p=0.012; posterior caudate, β=−0.162, p=0.002; anterior putamen, β=−0.133, p=0.024; posterior putamen, β=−0.125, p=0.039; ventral putamen, β=−0.125, p=0.035), except for the ventral striatum. CPV was positively associated with the UPDRS-III score even after adjusting for DAT availability in the posterior putamen (β=0.121; p=0.035). A larger CPV was associated with the future development of freezing of gait in the Cox regression model (HR 1.539, p=0.027) and a more rapid increase in dopaminergic medication in the linear mixed model (CPV×time, p=0.037), but was not associated with the risk of developing levodopa-induced dyskinesia or wearing off. Conclusion These findings suggest that CPV has the potential to serve as a biomarker for baseline and longitudinal motor disabilities in PD.
- 30 June 2023
- Published by BMJ in Journal of Neurology, Neurosurgery & Psychiatry
- Vol. 94 (12) , 1047-1055
Top Cited PapersAssociation of choroid plexus volume with motor symptoms and dopaminergic degeneration in Parkinson’s disease
Background The choroid plexus (CP) is involved in the clearance of harmful metabolites from the brain, as a part of the glymphatic system. This study aimed to investigate the association between CP volume (CPV), nigrostriatal dopaminergic degeneration and motor outcomes in Parkinson’s disease (PD). Methods We retrospectively searched drug-naïve patients with early-stage PD who underwent dopamine transporter (DAT) scanning and MRI. Automatic CP segmentation was performed, and the CPV was calculated. The relationship between CPV, DAT availability and Unified PD Rating Scale Part III (UPDRS-III) scores was assessed using multivariate linear regression. We performed longitudinal analyses to assess motor outcomes according to CPV. Results CPV was negatively associated with DAT availability in each striatal subregion (anterior caudate, β=−0.134, p=0.012; posterior caudate, β=−0.162, p=0.002; anterior putamen, β=−0.133, p=0.024; posterior putamen, β=−0.125, p=0.039; ventral putamen, β=−0.125, p=0.035), except for the ventral striatum. CPV was positively associated with the UPDRS-III score even after adjusting for DAT availability in the posterior putamen (β=0.121; p=0.035). A larger CPV was associated with the future development of freezing of gait in the Cox regression model (HR 1.539, p=0.027) and a more rapid increase in dopaminergic medication in the linear mixed model (CPV×time, p=0.037), but was not associated with the risk of developing levodopa-induced dyskinesia or wearing off. Conclusion These findings suggest that CPV has the potential to serve as a biomarker for baseline and longitudinal motor disabilities in PD.
- Open Access
- 7 March 2023
- Published by Springer Nature in Research Square
An Interpretable Multiparametric Radiomics Model of Basal Ganglia to Predict Dementia Conversion in Parkinson's Disease
Objectives: Cognitive impairment in Parkinson’s disease (PD) severely affects patients’ prognosis, and early detection of patients at high risk of dementia conversion is important for establishing treatment strategies. We aimed to investigate whether multiparametric MRI radiomics from basal ganglia can improve the prediction of dementia development in PD when integrated with clinical profiles. Methods: In this retrospective study, 262 patients with newly diagnosed PD (June 2008 – July 2017, follow-up >5 years) were included. MRI radiomic features (n = 1,284) were extracted from bilateral caudate and putamen. Two models were developed to predict dementia development: 1) a clinical model – age, disease duration, and cognitive composite scores, and 2) a combined clinical and radiomics model. The area under the receiver operating characteristic curve (AUC) were calculated for each model. The models’ interpretabilities were studied. Results: Among total 262 PD patients (mean age, 68 years ± 8 [standard deviation]; 134 men), 51 (30.4%) and 24 (25.5%) patients developed dementia within 5 years of PD diagnosis in the training (n = 168) and test sets (n = 94), respectively. The combined model achieved superior predictive performance compared to the clinical model in the training (AUCs 0.928 vs. 0.894, p= 0.284) and test set (AUCs 0.889 vs. 0.722, p = 0.016). The cognitive composite scores of the frontal/executive function domain contributed most to predicting dementia. Radiomics derived from the caudate were also highly associated with cognitive decline. Conclusion: Multiparametric MRI radiomics may have an incremental prognostic value when integrated with clinical profiles to predict future cognitive decline in PD.
- 7 March 2023
- Published by Springer Nature in Research Square
Open AccessAn Interpretable Multiparametric Radiomics Model of Basal Ganglia to Predict Dementia Conversion in Parkinson's Disease
Objectives: Cognitive impairment in Parkinson’s disease (PD) severely affects patients’ prognosis, and early detection of patients at high risk of dementia conversion is important for establishing treatment strategies. We aimed to investigate whether multiparametric MRI radiomics from basal ganglia can improve the prediction of dementia development in PD when integrated with clinical profiles. Methods: In this retrospective study, 262 patients with newly diagnosed PD (June 2008 – July 2017, follow-up >5 years) were included. MRI radiomic features (n = 1,284) were extracted from bilateral caudate and putamen. Two models were developed to predict dementia development: 1) a clinical model – age, disease duration, and cognitive composite scores, and 2) a combined clinical and radiomics model. The area under the receiver operating characteristic curve (AUC) were calculated for each model. The models’ interpretabilities were studied. Results: Among total 262 PD patients (mean age, 68 years ± 8 [standard deviation]; 134 men), 51 (30.4%) and 24 (25.5%) patients developed dementia within 5 years of PD diagnosis in the training (n = 168) and test sets (n = 94), respectively. The combined model achieved superior predictive performance compared to the clinical model in the training (AUCs 0.928 vs. 0.894, p= 0.284) and test set (AUCs 0.889 vs. 0.722, p = 0.016). The cognitive composite scores of the frontal/executive function domain contributed most to predicting dementia. Radiomics derived from the caudate were also highly associated with cognitive decline. Conclusion: Multiparametric MRI radiomics may have an incremental prognostic value when integrated with clinical profiles to predict future cognitive decline in PD.
- 9 May 2022
- Published by SAGE Publications in The Neuroradiology Journal
- Vol. 36 (1) , 49-58
Association of dynamic susceptibility contrast- and dynamic contrast-enhanced magnetic resonance imaging parameters with molecular marker status in lower-grade gliomas: A retrospective study
Purpose: Molecular marker status is clinically relevant for treatment planning and predicting the prognosis of gliomas. This study aimed to assess whether quantitative imaging parameters from dynamic susceptibility contrast- (DSC-) and dynamic contrast-enhanced (DCE)-magnetic resonance imaging (MRI) can predict the molecular marker status of lower-grade gliomas (LGGs). Materials and methods: Overall, 132 patients with LGGs who underwent DSC- and DCE-MRI were retrospectively enrolled. Statuses of relevant molecular markers including isocitrate dehydrogenase isoenzyme (IDH), 1p19q codeletion, epidermal growth factor receptor (EGFR), O6-methylguanine-DNA methyltransferase (MGMT), and telomerase reverse transcriptase (TERT) were collected. For each molecular marker, age, tumor diameter and location, and DSC- and DCE-MRI parameters, including the normalized cerebral blood volume (nCBV), volume transfer constant (Ktrans), rate transfer coefficient (Kep), extravascular extracellular volume fraction (Ve), and plasma volume fraction (Vp), were compared. Multivariable logistic regression analyses were performed. Results: The nCBV was significantly lower in LGGs with IDH mutation ( p = .001) and TERT mutation ( p = .027) than those without these mutations. Ktrans ( p = .034), Ve ( p = .023), and Vp ( p = .044) values were significantly lower in MGMT methylated LGGs than in MGMT unmethylated LGGs. Perfusion parameters were not significantly associated with EGFR amplification and 1p19q codeletion. Young age ( p < .001) and small diameter ( p = .001) were significantly associated with IDH mutation. The nCBV was independently associated with IDH status (AUC, 0.817; 95% CI: 0.739–0.894). Conclusions: DSC- and DCE-MRI parameters demonstrated correlations with molecular markers of LGGs. Especially, the nCBV can be helpful in predicting the IDH mutation status.
- 9 May 2022
- Published by SAGE Publications in The Neuroradiology Journal
- Vol. 36 (1) , 49-58
Association of dynamic susceptibility contrast- and dynamic contrast-enhanced magnetic resonance imaging parameters with molecular marker status in lower-grade gliomas: A retrospective study
Purpose: Molecular marker status is clinically relevant for treatment planning and predicting the prognosis of gliomas. This study aimed to assess whether quantitative imaging parameters from dynamic susceptibility contrast- (DSC-) and dynamic contrast-enhanced (DCE)-magnetic resonance imaging (MRI) can predict the molecular marker status of lower-grade gliomas (LGGs). Materials and methods: Overall, 132 patients with LGGs who underwent DSC- and DCE-MRI were retrospectively enrolled. Statuses of relevant molecular markers including isocitrate dehydrogenase isoenzyme (IDH), 1p19q codeletion, epidermal growth factor receptor (EGFR), O6-methylguanine-DNA methyltransferase (MGMT), and telomerase reverse transcriptase (TERT) were collected. For each molecular marker, age, tumor diameter and location, and DSC- and DCE-MRI parameters, including the normalized cerebral blood volume (nCBV), volume transfer constant (Ktrans), rate transfer coefficient (Kep), extravascular extracellular volume fraction (Ve), and plasma volume fraction (Vp), were compared. Multivariable logistic regression analyses were performed. Results: The nCBV was significantly lower in LGGs with IDH mutation ( p = .001) and TERT mutation ( p = .027) than those without these mutations. Ktrans ( p = .034), Ve ( p = .023), and Vp ( p = .044) values were significantly lower in MGMT methylated LGGs than in MGMT unmethylated LGGs. Perfusion parameters were not significantly associated with EGFR amplification and 1p19q codeletion. Young age ( p < .001) and small diameter ( p = .001) were significantly associated with IDH mutation. The nCBV was independently associated with IDH status (AUC, 0.817; 95% CI: 0.739–0.894). Conclusions: DSC- and DCE-MRI parameters demonstrated correlations with molecular markers of LGGs. Especially, the nCBV can be helpful in predicting the IDH mutation status.
- Open Access
- 1 January 2022
- Published by XMLink in Korean Journal of Radiology
- Vol. 23 (1) , 77-88
Quality of Radiomics Research on Brain Metastasis: A Roadmap to Promote Clinical Translation
Park CJ, et al. Korean J Radiol. 2022 Jan;23(1):77-88. https://doi.org/10.3348/kjr.2021.0421
- 1 January 2022
- Published by XMLink in Korean Journal of Radiology
- Vol. 23 (1) , 77-88
Open AccessQuality of Radiomics Research on Brain Metastasis: A Roadmap to Promote Clinical Translation
Park CJ, et al. Korean J Radiol. 2022 Jan;23(1):77-88. https://doi.org/10.3348/kjr.2021.0421