Ho-Joon Lee

Daeseok Oh, Dong Ah Lee, Ho-Joon Lee, and Kang Min Park. Korean J Pain -0001;0:. https://doi.org/10.3344/kjp.26024

Inner ear segmentation supports the diagnosis and treatment planning of auditory‐related disorders such as Meniere's disease. However, manual annotation of inner ear structures on MR images is both time‐consuming and labor‐intensive. To address this limitation, we developed a deep learning method that takes T2‐weighted MR images as input and segments the inner ear into three subregions: the cochlear basal turn, the cochlear mid‐to‐apical turn, and the vestibule including the semicircular canals. A total of 74 three‐dimensional (3D) T2‐weighted MR images were retrospectively collected and divided into 50 training cases and 24 internal test cases. In addition, eight MR images from the publicly available Vestibular‐Schwannoma‐SEG dataset were used as an independent external test set. Ground‐truth segmentations for both the internal and external datasets were manually annotated by an experienced radiologist. A 3D transformer‐based deep learning model was trained using these annotated data. To further improve segmentation performance, particularly for the thin and intricate semicircular canals, a label‐preserving data augmentation strategy was introduced and compared with a conventional data augmentation approach. Segmentation performance was evaluated using quantitative and qualitative assessments. Quantitative evaluation included the Dice similarity coefficient (DSC), Intersection over union (IoU), and Hausdorff distance (HD). On the internal test set, the proposed method achieved a mean DSC of 0.905 (vs. 0.904), a mean IoU of 0.828 (vs. 0.824), and a mean HD of 2.75 (vs. 3.04) compared with the conventional augmentation approach. On the external dataset, the proposed method demonstrated larger improvements, achieving a mean DSC of 0.919 (vs. 0.847), a mean IoU of 0.852 (vs. 0.738), and a mean HD of 3.92 (vs. 4.45). These results demonstrate that the proposed label‐preserving data augmentation method improves the robustness and accuracy of inner ear subregion segmentation, particularly for thin and intricate structures.

Oh D, et al. J Clin Neurol. 2026 ;22:e48. https://doi.org/10.3988/jcn.2026.0005

Objectives: Peak width of skeletonized mean diffusivity (PSMD) serves as a new biomarker for diffuse white matter microstructural damage. This study is aimed at quantifying PSMD in patients with status epilepticus (SE) to investigate white matter microstructural abnormalities and their potential association with small vessel disease (SVD) mechanisms in SE. Methods: We retrospectively enrolled patients with SE and healthy controls matched for age and sex. Diffusion tensor imaging (DTI) was performed using a 3.0 Tesla magnetic resonance imaging scanner to measure PSMD. We compared the PSMD values between patients with SE and healthy controls. Additionally, we analyzed correlations between PSMD and clinical factors in patients with SE. Results: Twenty‐eight patients with SE and 31 healthy controls were included in this study. PSMD values were significantly higher in patients with SE compared with controls (3.024 vs. 2.257, p < 0.001). Additionally, in patients with SE, PSMD positively correlated with age ( r = 0.721, p < 0.001). However, PSMD was not significantly correlated with other clinical characteristics, including the duration of SE and the number of antiseizure medications ( r = 0.031, p = 0.876; r = −0.340, p = 0.076, respectively). Conclusion: Patients with SE exhibited significantly elevated PSMD, which was associated with white matter microstructural abnormalities. These findings suggested that PSMD may be a useful imaging marker for assessing white matter changes in SE and warrant further investigation into its association with underlying vascular and nonvascular mechanisms.

Park KM, et al. J Clin Neurol. 2025 Jul;21:e25. https://doi.org/10.3988/jcn.2025.0036

김지민, et al. J Korean Soc Radiol. 2025 Jan;86(1):34-44. https://doi.org/10.3348/jksr.2024.0147

Increased choroid plexus (ChP) volume is well known to be associated with glymphatic system dysfunction. This study aimed to investigate glymphatic system function in patients with transient global amnesia (TGA) compared to healthy controls through ChP volumes measurements. We retrospectively enrolled patients with TGA from our hospital, as well as healthy controls. This was retrospectively observational study followed STROBE guideline. All participants underwent brain magnetic resonance imaging, including three-dimensional T1-weighted imaging. We analyzed and compared ChP volumes between patients with TGA and healthy controls and investigated the relationship between ChP volumes and clinical characteristics in patients with TGA. We enrolled 44 patients with TGA and 47 healthy controls. Among the 44 patients with TGA, 38 experienced a single TGA event, while 6 had recurrent TGA events. ChP volumes did not significantly differ between patients with TGA and healthy controls (2.140% vs 2.089%, P = .568). However, ChP volumes were higher in patients with a single TGA event compared to those with recurrent events (2.204% vs 1.740%, P < .013). We observed a significant positive correlation between ChP volumes and age in patients with TGA (R = 0.282, P = .007). ChP volumes were not associated with the duration of amnesia in patients with TGA (R = 0.187, P = .274). We find no differences in the glymphatic system function, as demonstrated by ChP volume for the first time. This study also found a significant correlation between ChP volume and age in patients with TGA, indicating that aging influences glymphatic system function.

Objectives: This study aimed to investigate the differences in structural connectivity and glymphatic system function between patients with temporal lobe epilepsy (TLE) and hippocampal sclerosis (HS) and healthy controls. Additionally, we analyzed the association between structural connectivity, glymphatic system function, and antiseizure medication (ASM) response.Methods: We retrospectively enrolled patients with TLE and HS and healthy controls who underwent diffusion tensor imaging at our hospital. We assessed structural connectivity in patients with TLE and HS and healthy controls by calculating network measures using graph theory and evaluated glymphatic system function using the diffusion tensor image analysis along the perivascular space (DTI‐ALPS) index. Patients with TLE and HS were categorized into two groups: ASM poor and good responders.Results: We enrolled 55 patients with TLE and HS and 53 healthy controls. Of the 55 patients with TLE and HS, 39 were ASM poor responders, and 16 were ASM good responders. The assortativity coefficient in patients with TLE and HS was higher than that in healthy controls (0.004 vs. −0.007, p = 0.004), and the assortativity coefficient in ASM poor responders was lower than that in ASM good responders (−0.001 vs. −0.197, p = 0.003). The DTI‐ALPS index in patients with TLE and HS was lower than that in healthy controls (1.403 vs. 1.709, p < 0.001); however, the DTI‐ALPS index did not differ between ASM poor and good responders (1.411 vs. 1.385, p = 0.628). The DTI‐ALPS index had a significant negative correlation with age in patients with TLE and HS (r = −0.267, p = 0.049).Significance: We confirmed increased assortativity coefficient in structural connectivity and decreased DTI‐ALPS index in patients with TLE and HS compared with healthy controls. Additionally, we demonstrated an association between decreased assortativity coefficient in structural connectivity and ASM poor response in patients with TLE patients and HS.Plain Language Summary: This study investigates the relationship between brain connectivity changes and glymphatic system function with antiseizure medication response in patients with temporal lobe epilepsy and hippocampal sclerosis. The research reveals that these patients show altered brain connectivity and glymphatic function compared to healthy individuals. A key finding is the strong link between a specific connectivity measure (assortativity coefficient) and antiseizure medication response, providing valuable insights that could influence epilepsy treatment and future research directions.

BACKGROUND AND PURPOSE: The peak width of skeletonized mean diffusivity (PSMD) is a novel marker of small vessel disease. In this study, we aimed to investigate the presence of small vessel disease in patients with transient global amnesia (TGA) using the PSMD. MATERIALS AND METHODS: We enrolled 75 patients newly diagnosed with TGA and included 65 age-and sex-matched healthy controls. Diffusion tensor imaging (DTI) was performed using a 3T magnetic resonance imaging scanner. We measured the PSMD based on DTI using the FSL program. This measure was compared between patients with TGA and healthy controls. Additionally, we conducted a correlation analysis to explore the relationship between PSMD and clinical factors. RESULTS: A significant difference in the PSMD between patients with TGA and healthy controls was observed. Patients with TGA exhibited higher a PSMD compared to healthy controls (2.297±0.232 vs. 2.188±0.216 ×10⁻⁴ mm²/s, p=0.005). Additionally, patients with TGA but without any vascular risk factors, such as diabetes, hypertension or dyslipidemia, also exhibited higher a PSMD compared to healthy controls (2.278±0.253 vs. 2.188±0.216 ×10⁻⁴ mm²/s, p=0.036). The PSMD positively correlated with age (r=0.248, p=0.032); however, it was not associated with duration of amnesia. CONCLUSIONS: This finding underscores the feasibility of using PSMD as a marker for detecting small vessel diseases in patients with neurological disorders. Furthermore, our study also implies the presence of small vessel disease may be present in patients with TGA. ABBREVIATIONS: TGA=transient global amnesia; TIA= transient ischemic attack; PSMD= peak width of skeletonized mean diffusivity; DTI= diffusion tensor imaging.