Gilsoon 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
Research Subjects
Research Interests
Gilsoon Park has a total of 88 co-authors. Here are the top co-authors by number of shared publications:
Hosung Kim
USC Stevens Neuroimaging and Informatics Institute, Keck School of Medicine of USC, University of Southern California , Los Angeles, CA
3 shared publications- Hosung Kim
University of Southern California
3 shared publications - Jong-Min Lee
Hanyang University
2 shared publications Adriana B Conforto
LIM-44, Laboratory of Magnetic Resonance Imaging in Neuroradiology, Hospital das Clinicas da Faculdade de Medicina da Universidade de Sao Paulo, Sao Paulo, Brazil; Hospital Israelita Albert Einstein, Sao Paulo, Brazil
1 shared publicationAmy Brodtmann
Cognitive Health Initiative, School of Translational Medicine, Monash University, Melbourne, VIC, Australia
1 shared publicationAndrew D Robertson
Kinesiology and Health Sciences, University of Waterloo, Waterloo, ON, Canada
1 shared publication
Latest Publications
- Open Access
- 22 January 2026
- Published by Elsevier in The Lancet Digital Health
- Vol. 8 (1) , 100942
Associations between contralesional neuroplasticity and motor impairment through deep learning-derived MRI regional brain age in chronic stroke (ENIGMA): a multicohort, retrospective, observational study
- 22 January 2026
- Published by Elsevier in The Lancet Digital Health
- Vol. 8 (1) , 100942
Open AccessAssociations between contralesional neuroplasticity and motor impairment through deep learning-derived MRI regional brain age in chronic stroke (ENIGMA): a multicohort, retrospective, observational study
- 28 June 2025
- Published by Oxford University Press (OUP) in SLEEP
- Vol. 48 (11)
Chronotype prior to shift work influences outcomes: resilience as a mediator of sleep and mental health
Study Objectives: Despite persistent nursing shortages, newly hired nurses inevitably engage in shift work, which harms their sleep and mental health, accelerating their early-career turnover. This study examined whether pre-shift work chronotype is associated with post-shift work sleep/mental health outcomes and whether resilience mediates these relationships. Methods: A prospective cohort of 595 newly hired nurses, of whom 331 were included in the final analysis, was recruited from two national tertiary hospitals. Participants provided baseline and 6–12 months follow-up data after SW exposure. Chronotype, resilience, sleep outcomes (sleep quality and daytime sleepiness), and mental health outcomes (depression, anxiety, and stress scales) were assessed at both baseline and follow-up using validated self-report questionnaires. Structural equation modeling examined direct and indirect pathways linking baseline chronotype to follow-up resilience and sleep/mental health outcomes. Results: Among 331 nurses (age: 23.4 ± 1.2 years, 293 females, 26 morningness, 186 intermediate, 119 eveningness), the morningness group exhibited the highest resilience at both baseline (p = .0206) and follow-up (p = .0002). Baseline chronotype was significantly associated with resilience following SW (β = 0.259, p < .01). Baseline chronotype further indirectly affected both excessive daytime sleepiness (β = -0.013, 95% CI: [-0.025, -0.001]) and mental health (β = -0.045, 95% CI [-0.066, -0.009]) via resilience. Conclusion: Pre-shift work chronotype significantly influenced post-shift work sleep and mental health outcomes, with resilience serving as a key mediator. These findings highlight the potential for targeted interventions—such as resilience training or chronotype-based scheduling—to mitigate early shift work-related mental health challenges and reduce nurse turnover. Statement of Significance Previous studies examining the interaction between chronotype and shift work (SW) schedules among nurses have yielded inconsistent results, primarily because SW schedules transiently influence chronotype, and most research has focused on nurses already exposed to SW. Although chronotype is generally stable and tends to revert once an intervention (e.g. altered light exposure) ceases—highlighting its trait-like characteristic—it can still be temporarily influenced by SW exposure. Consequently, prior studies often assessed chronotypes after participants had already been exposed to SW, potentially affecting the reliability of their assessments. In our study, baseline chronotypes measured before SW exposure influenced sleep and mental health during the initial adaptation period, with resilience acting as a mediator between chronotype and these outcomes. Although we did not examine long-term outcomes following SW exposure, our findings suggest that baseline chronotype may serve as a predictive biomarker for early SW outcomes.
- 28 June 2025
- Published by Oxford University Press (OUP) in SLEEP
- Vol. 48 (11)
Chronotype prior to shift work influences outcomes: resilience as a mediator of sleep and mental health
Study Objectives: Despite persistent nursing shortages, newly hired nurses inevitably engage in shift work, which harms their sleep and mental health, accelerating their early-career turnover. This study examined whether pre-shift work chronotype is associated with post-shift work sleep/mental health outcomes and whether resilience mediates these relationships. Methods: A prospective cohort of 595 newly hired nurses, of whom 331 were included in the final analysis, was recruited from two national tertiary hospitals. Participants provided baseline and 6–12 months follow-up data after SW exposure. Chronotype, resilience, sleep outcomes (sleep quality and daytime sleepiness), and mental health outcomes (depression, anxiety, and stress scales) were assessed at both baseline and follow-up using validated self-report questionnaires. Structural equation modeling examined direct and indirect pathways linking baseline chronotype to follow-up resilience and sleep/mental health outcomes. Results: Among 331 nurses (age: 23.4 ± 1.2 years, 293 females, 26 morningness, 186 intermediate, 119 eveningness), the morningness group exhibited the highest resilience at both baseline (p = .0206) and follow-up (p = .0002). Baseline chronotype was significantly associated with resilience following SW (β = 0.259, p < .01). Baseline chronotype further indirectly affected both excessive daytime sleepiness (β = -0.013, 95% CI: [-0.025, -0.001]) and mental health (β = -0.045, 95% CI [-0.066, -0.009]) via resilience. Conclusion: Pre-shift work chronotype significantly influenced post-shift work sleep and mental health outcomes, with resilience serving as a key mediator. These findings highlight the potential for targeted interventions—such as resilience training or chronotype-based scheduling—to mitigate early shift work-related mental health challenges and reduce nurse turnover. Statement of Significance Previous studies examining the interaction between chronotype and shift work (SW) schedules among nurses have yielded inconsistent results, primarily because SW schedules transiently influence chronotype, and most research has focused on nurses already exposed to SW. Although chronotype is generally stable and tends to revert once an intervention (e.g. altered light exposure) ceases—highlighting its trait-like characteristic—it can still be temporarily influenced by SW exposure. Consequently, prior studies often assessed chronotypes after participants had already been exposed to SW, potentially affecting the reliability of their assessments. In our study, baseline chronotypes measured before SW exposure influenced sleep and mental health during the initial adaptation period, with resilience acting as a mediator between chronotype and these outcomes. Although we did not examine long-term outcomes following SW exposure, our findings suggest that baseline chronotype may serve as a predictive biomarker for early SW outcomes.
- Open Access
- 22 November 2024
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 13 (23) , e035881
P‐Wave Duration Is Associated With Aging Patterns in Structural Brain Networks
Background: Impaired cardiac function is associated with cognitive impairment and brain imaging features of aging. Cardiac arrhythmias, including atrial fibrillation, are implicated in clinical and subclinical brain injuries. Even in the absence of a clinical diagnosis, subclinical or prodromal substrates of arrhythmias, including an abnormally long or short P‐wave duration (PWD), a measure associated with atrial abnormalities, have been associated with stroke and cognitive decline. However, the extent to which PWD has subclinical influences on overall aging patterns of the brain is not clearly understood. Methods and Results: Here, using neuroimaging and ECG data from the UK Biobank, we use a novel regional “brain age” method to identify the brain aging networks associated with abnormal PWD. We find associations between short PWD and accelerated brain aging in the sensorimotor, frontoparietal, ventral attention, and dorsal attention networks, even in the absence of overt cardiac diseases. Conclusions: These findings contribute to our understanding of the relationship between PWD and structural brain aging. This work emphasizes the need for continued study designs that consider brain‐based outcomes related to abnormally short PWD.
- 22 November 2024
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 13 (23) , e035881
Open AccessP‐Wave Duration Is Associated With Aging Patterns in Structural Brain Networks
Background: Impaired cardiac function is associated with cognitive impairment and brain imaging features of aging. Cardiac arrhythmias, including atrial fibrillation, are implicated in clinical and subclinical brain injuries. Even in the absence of a clinical diagnosis, subclinical or prodromal substrates of arrhythmias, including an abnormally long or short P‐wave duration (PWD), a measure associated with atrial abnormalities, have been associated with stroke and cognitive decline. However, the extent to which PWD has subclinical influences on overall aging patterns of the brain is not clearly understood. Methods and Results: Here, using neuroimaging and ECG data from the UK Biobank, we use a novel regional “brain age” method to identify the brain aging networks associated with abnormal PWD. We find associations between short PWD and accelerated brain aging in the sensorimotor, frontoparietal, ventral attention, and dorsal attention networks, even in the absence of overt cardiac diseases. Conclusions: These findings contribute to our understanding of the relationship between PWD and structural brain aging. This work emphasizes the need for continued study designs that consider brain‐based outcomes related to abnormally short PWD.
- 23 July 2024
- Published by Wolters Kluwer Health in Neurology
- Vol. 103 (2) , e209498
Spatiotemporal Progression Patterns of Dopamine Availability and Deep Gray Matter Volume in Parkinson Disease–Related Cognitive Impairment
Cognitive impairment is a frequent nonmotor symptom in patients with Parkinson disease (PD), and early cognitive decline is often attributed to dopaminergic system dysfunction. We aimed to explore spatiotemporal progression patterns of striatal dopamine availability and regional brain volume based on cognitive status among patients with PD. This retrospective, cross-sectional study included patients with newly diagnosed PD who were not taking medication for this condition who visited a university-affiliated hospital in Seoul between January 2018 and December 2020. Patients were classified as having normal cognition (PD-NC), mild cognitive impairment (PD-MCI), or PD dementia (PDD) based on Seoul Neuropsychological Screening Battery-II, which includes 31 subsets covering activities of daily living and 5 cognitive domains. They all had brain imaging with MRI and PET with 18F-N-(3-fluoropropyl)-2beta-carbon ethoxy-3beta-(4-iodophenyl) nortropane at baseline. Subsequently, standardized uptake value ratios (SUVRs) for regional dopamine availability and regional gray matter volumes were obtained using automated segmentation. These metrics were compared across cognitive status groups, and spatiotemporal progression patterns were analyzed using the Subtype and Stage Inference machine learning technique. Among 168 patients (mean age, 73.3 ± 6.1 years; 81 [48.2%] women), 65 had PD-NC, 65 had PD-MCI, and 38 had PDD. Patients with PD-MCI exhibited lower SUVRs (3.61 ± 1.31, p < 0.001) in the caudate than patients with PD-NC (4.43 ± 1.21) but higher SUVRs than patients with PDD (2.39 ± 1.06). Patients with PD-NC had higher thalamic SUVRs (1.55 ± 0.16, p < 0.001) than patients with both PD-MCI (1.45 ± 0.16) and PDD (1.38 ± 0.19). Regional deep gray matter volumes of the caudate (p = 0.015), putamen (p = 0.012), globus pallidus (p < 0.001), thalamus (p < 0.001), hippocampus (p < 0.001), and amygdala (p < 0.001) were more reduced in patients with PD-MCI or PDD than in patients with PD-NC, and the SUVR of the caudate correlated with caudate volume (r = 0.187, p = 0.015). Hippocampal atrophy was the initial change influencing cognitive impairment. The reduced dopamine availability of the thalamus preceded reductions in volume across most deep gray matter regions. Our finding underscores the association between decreased dopamine availability and volume of the caudate and thalamus with cognitive dysfunction in PD. The dopamine availability of the caudate and thalamus was reduced before the volume of the caudate and thalamus was decreased, highlighting the spatiotemporal association between dopaminergic and structural pathology in cognitive impairment in PD.
- 23 July 2024
- Published by Wolters Kluwer Health in Neurology
- Vol. 103 (2) , e209498
Spatiotemporal Progression Patterns of Dopamine Availability and Deep Gray Matter Volume in Parkinson Disease–Related Cognitive Impairment
Cognitive impairment is a frequent nonmotor symptom in patients with Parkinson disease (PD), and early cognitive decline is often attributed to dopaminergic system dysfunction. We aimed to explore spatiotemporal progression patterns of striatal dopamine availability and regional brain volume based on cognitive status among patients with PD. This retrospective, cross-sectional study included patients with newly diagnosed PD who were not taking medication for this condition who visited a university-affiliated hospital in Seoul between January 2018 and December 2020. Patients were classified as having normal cognition (PD-NC), mild cognitive impairment (PD-MCI), or PD dementia (PDD) based on Seoul Neuropsychological Screening Battery-II, which includes 31 subsets covering activities of daily living and 5 cognitive domains. They all had brain imaging with MRI and PET with 18F-N-(3-fluoropropyl)-2beta-carbon ethoxy-3beta-(4-iodophenyl) nortropane at baseline. Subsequently, standardized uptake value ratios (SUVRs) for regional dopamine availability and regional gray matter volumes were obtained using automated segmentation. These metrics were compared across cognitive status groups, and spatiotemporal progression patterns were analyzed using the Subtype and Stage Inference machine learning technique. Among 168 patients (mean age, 73.3 ± 6.1 years; 81 [48.2%] women), 65 had PD-NC, 65 had PD-MCI, and 38 had PDD. Patients with PD-MCI exhibited lower SUVRs (3.61 ± 1.31, p < 0.001) in the caudate than patients with PD-NC (4.43 ± 1.21) but higher SUVRs than patients with PDD (2.39 ± 1.06). Patients with PD-NC had higher thalamic SUVRs (1.55 ± 0.16, p < 0.001) than patients with both PD-MCI (1.45 ± 0.16) and PDD (1.38 ± 0.19). Regional deep gray matter volumes of the caudate (p = 0.015), putamen (p = 0.012), globus pallidus (p < 0.001), thalamus (p < 0.001), hippocampus (p < 0.001), and amygdala (p < 0.001) were more reduced in patients with PD-MCI or PDD than in patients with PD-NC, and the SUVR of the caudate correlated with caudate volume (r = 0.187, p = 0.015). Hippocampal atrophy was the initial change influencing cognitive impairment. The reduced dopamine availability of the thalamus preceded reductions in volume across most deep gray matter regions. Our finding underscores the association between decreased dopamine availability and volume of the caudate and thalamus with cognitive dysfunction in PD. The dopamine availability of the caudate and thalamus was reduced before the volume of the caudate and thalamus was decreased, highlighting the spatiotemporal association between dopaminergic and structural pathology in cognitive impairment in PD.
- Open Access
- 15 November 2022
- Published by Elsevier in NeuroImage
- Vol. 264, 119753
Novel neuroelectrophysiological age index associated with imaging features of brain aging and sleep disorders.
- 15 November 2022
- Published by Elsevier in NeuroImage
- Vol. 264, 119753
Open AccessNovel neuroelectrophysiological age index associated with imaging features of brain aging and sleep disorders.
- 1 March 2022
- Published by Wolters Kluwer Health in Stroke
- Vol. 53 (3) , 698-709
Cilostazol Versus Aspirin on White Matter Changes in Cerebral Small Vessel Disease: A Randomized Controlled Trial
Background and Purpose: Cerebral small vessel disease is characterized by progressive cerebral white matter changes (WMCs). This study aimed to compare the effects of cilostazol and aspirin on changes in WMC volume in patients with cerebral small vessel disease. Methods: In a multicenter, double-blind, randomized controlled trial, participants with moderate or severe WMCs and at least one lacunar infarction detected on brain magnetic resonance imaging were randomly assigned to the cilostazol and aspirin groups in a 1:1 ratio. Cilostazol slow release (200 mg) or aspirin (100 mg) capsules were administered once daily for 2 years. The primary outcome was the change in WMC volume on magnetic resonance images from baseline to 2 years. Secondary imaging outcomes include changes in the number of lacunes or cerebral microbleeds, fractional anisotropy, and mean diffusivity on diffusion tensor images, and brain atrophy. Secondary clinical outcomes include all ischemic strokes, all ischemic vascular events, and changes in cognition, motor function, mood, urinary symptoms, and disability. Results: Between July 2013 and August 2016, 256 participants were randomly assigned to the cilostazol (n=127) and aspirin (n=129) groups. Over 2 years, the percentage of WMC volume to total WM volume and the percentage of WMC volume to intracranial volume increased in both groups, but neither analysis showed significant differences between the groups. The peak height of the mean diffusivity histogram in normal-appearing WMs was significantly reduced in the aspirin group compared with the cilostazol group. Cilostazol significantly reduced the risk of ischemic vascular event compared with aspirin (0.5 versus 4.5 cases per 100 person-years; hazard ratio, 0.11 [95% CI, 0.02–0.89]). Conclusions: There was no significant difference between the effects of cilostazol and aspirin on WMC progression in patients with cerebral small vessel disease. Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT01932203.
- 1 March 2022
- Published by Wolters Kluwer Health in Stroke
- Vol. 53 (3) , 698-709
Cilostazol Versus Aspirin on White Matter Changes in Cerebral Small Vessel Disease: A Randomized Controlled Trial
Background and Purpose: Cerebral small vessel disease is characterized by progressive cerebral white matter changes (WMCs). This study aimed to compare the effects of cilostazol and aspirin on changes in WMC volume in patients with cerebral small vessel disease. Methods: In a multicenter, double-blind, randomized controlled trial, participants with moderate or severe WMCs and at least one lacunar infarction detected on brain magnetic resonance imaging were randomly assigned to the cilostazol and aspirin groups in a 1:1 ratio. Cilostazol slow release (200 mg) or aspirin (100 mg) capsules were administered once daily for 2 years. The primary outcome was the change in WMC volume on magnetic resonance images from baseline to 2 years. Secondary imaging outcomes include changes in the number of lacunes or cerebral microbleeds, fractional anisotropy, and mean diffusivity on diffusion tensor images, and brain atrophy. Secondary clinical outcomes include all ischemic strokes, all ischemic vascular events, and changes in cognition, motor function, mood, urinary symptoms, and disability. Results: Between July 2013 and August 2016, 256 participants were randomly assigned to the cilostazol (n=127) and aspirin (n=129) groups. Over 2 years, the percentage of WMC volume to total WM volume and the percentage of WMC volume to intracranial volume increased in both groups, but neither analysis showed significant differences between the groups. The peak height of the mean diffusivity histogram in normal-appearing WMs was significantly reduced in the aspirin group compared with the cilostazol group. Cilostazol significantly reduced the risk of ischemic vascular event compared with aspirin (0.5 versus 4.5 cases per 100 person-years; hazard ratio, 0.11 [95% CI, 0.02–0.89]). Conclusions: There was no significant difference between the effects of cilostazol and aspirin on WMC progression in patients with cerebral small vessel disease. Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT01932203.
- Open Access
- 3 May 2021
- Published by Elsevier in NeuroImage
- Vol. 237, 118140
White matter hyperintensities segmentation using the ensemble U-Net with multi-scale highlighting foregrounds
- 3 May 2021
- Published by Elsevier in NeuroImage
- Vol. 237, 118140
Open AccessWhite matter hyperintensities segmentation using the ensemble U-Net with multi-scale highlighting foregrounds