1.Radiomics and Deep Learning in Brain Metastases: Current Trends and Roadmap to Future Applications
Yae Won PARK ; Narae LEE ; Sung Soo AHN ; Jong Hee CHANG ; Seung-Koo LEE
Investigative Magnetic Resonance Imaging 2021;25(4):266-280
Advances in radiomics and deep learning (DL) hold great potential to be at the forefront of precision medicine for the treatment of patients with brain metastases.Radiomics and DL can aid clinical decision-making by enabling accurate diagnosis, facilitating the identification of molecular markers, providing accurate prognoses, and monitoring treatment response. In this review, we summarize the clinical background, unmet needs, and current state of research of radiomics and DL for the treatment of brain metastases. The promises, pitfalls, and future roadmap of radiomics and DL in brain metastases are addressed as well.
2.Imaging Findings of a Nonenhancing Intradural Paramedian Chordoma Mimicking an Epidermoid Cyst
Min Jeong KIM ; Yae Won PARK ; Soo Mee LIM
Journal of the Korean Radiological Society 2018;79(3):175-180
Intracranial chordoma is a rare tumor, originating from embryonic remnants of the primitive notochord. It typically appears as an enhancing extradural midline tumor with bone involvement. We introduce a rare case of a 27-year-old male who had a nonenhancing intradural chordoma showing paramedian location, involving the left cavernous sinus, Meckel's cave, and prepontine cistern. The pathologic diagnosis was confirmed as an intradural chordoma. The imaging findings of this unusual case of a nonenhancing intradural paramedian chordoma will be presented with the differential diagnosis focused on the epidermoid cyst.
3.Influence of Clinical and Demographical Variables on Depression, Instrumental Activities of Daily Living, and Cognitive Function in Patients with Parkinson's Disease.
Kyeong Yae SOHNG ; Dong Won CHOI ; Hea Ja PARK
Journal of Korean Academy of Fundamental Nursing 2006;13(2):249-256
PURPOSE: To investigate the influence of clinical and demographical variables on depression, instrumental activities of daily living (IADL), and cognitive function in patients with Parkinson's disease. METHOD: Using a structured questionnaire data were collected from 100 participants registered in the neurology department of C university hospital. Duration and stage of disease, fall history, vision impairment, duration and quality of sleep, orthostatic hypotension, ambulation impairment, and use of walking aid were included in clinical variables. Depression, IADL, and cognitive function were assessed using Kee's GDSSF-K, Cho's scale, and K-MMSE. Collected data were analyzed using the SAS program. RESULTS: The depression score for the participants was 7.78, higher than cut-off score(5). Participants who were male, living with spouse, above high school education, high economic status, with no vision impairment, and no ambulation impairment revealed high cognitive scores. Average IADL score were significantly higher for participants who were male, who had high economic status, low stage of disease, and no ambulation impairment. Depression IADL & cognitive function scores were significantly different according to ambulation impairment. Scores for ADL and cognitive function were positively correlated. CONCLUSION: It is recommended that make programs for patients with Parkinson's disease, clinical and demographic variables should be considered according to their individual needs.
Activities of Daily Living*
;
Cognition
;
Depression*
;
Education
;
Humans
;
Hypotension, Orthostatic
;
Male
;
Neurology
;
Parkinson Disease*
;
Surveys and Questionnaires
;
Spouses
;
Walking
4.Complete Recovery of Perfusion Abnormalities in a Cardiac Arrest Patient Treated with Hypothermia: Results of Cerebral Perfusion MR Imaging
Min Jeong KIM ; Yae Won PARK ; Soo Mee LIM
Investigative Magnetic Resonance Imaging 2018;22(1):56-60
Therapeutic hypothermia in cardiac arrest patients is associated with favorable outcomes mediated via neuroprotective mechanisms. We report a rare case of a 32-year-old male who demonstrated complete recovery of signal changes on perfusion-weighted imaging after therapeutic hypothermia due to cardiac arrest. Brain MRI with perfusion-weighted imaging, performed three days after ending the hypothermia therapy, showed a marked decrease in relative cerebral blood flow (rCBF) and delay in mean transit time (MTT) in the bilateral basal ganglia, thalami, brain stem, cerebellum, occipitoparietal cortex, and frontotemporal cortex. However, no cerebral ischemia was not noted on diffusion-weighted imaging (DWI) or fluid-attenuated inversion recovery (FLAIR) sequences. A follow-up brain MRI after one week showed complete resolution of the perfusion deficit and the patient was discharged without any neurologic sequelae. The mechanism and interpretation of the perfusion changes in cardiac arrest patients treated with therapeutic hypothermia are discussed.
Adult
;
Basal Ganglia
;
Brain
;
Brain Ischemia
;
Brain Stem
;
Cerebellum
;
Cerebrovascular Circulation
;
Follow-Up Studies
;
Heart Arrest
;
Humans
;
Hypothermia
;
Hypothermia, Induced
;
Magnetic Resonance Imaging
;
Male
;
Perfusion
5.Comparison of Contrast-Enhanced T2 FLAIR and 3D T1 Black-Blood Fast Spin-Echo for Detection of Leptomeningeal Metastases
Investigative Magnetic Resonance Imaging 2018;22(2):86-93
PURPOSE: Imaging plays a significant role in diagnosing leptomeningeal metastases. However, the most appropriate sequence for the detection of leptomeningeal metastases has yet to be determined. This study compares the efficacies of contrast-enhanced T2 fluid attenuated inversion recovery (FLAIR) and contrast-enhanced 3D T1 black-blood fast spin echo (FSE) imaging for the detection of leptomeningeal metastases. MATERIALS AND METHODS: Tube phantoms containing varying concentrations of gadobutrol solution were scanned using T2 FLAIR and 3D T1 black-blood FSE. Additionally, 30 patients with leptomeningeal metastases were retrospectively evaluated to compare conspicuous lesions and the extent of leptomeningeal metastases detected by T2 FLAIR and 3D T1 black-blood FSE. RESULTS: The signal intensities of low-concentration gadobutrol solutions (< 0.5 mmol/L) on T2 FLAIR images were higher than in 3D T1 black-blood FSE. The T2 FLAIR sequences exhibited significantly greater visual conspicuity scores than the 3D T1 black-blood sequence in leptomeningeal metastases of the pial membrane of cistern (P = 0.014). T2 FLAIR images exhibited a greater or equal extent (96.7%) of leptomeningeal metastases than 3D T1 black-blood FSE images. CONCLUSION: Because of its high sensitivity even at low gadolinium concentrations, contrast-enhanced T2 FLAIR images delineated leptomeningeal metastases in a wider territory than 3D T1 black-blood FSE.
Gadolinium
;
Humans
;
Membranes
;
Neoplasm Metastasis
;
Retrospective Studies
6.A case of intestinal lymphangiectasia.
Yae Kyung SUH ; Kyung Hee PARK ; Chul Ho JANG ; Bum Soo PARK ; Jeong Kee SEO ; Sung Hae PARK ; Je Geun CHI ; Kyung Mo YEON ; Kwi Won PARK
Journal of the Korean Pediatric Society 1992;35(12):1737-1743
No abstract available.
Protein-Losing Enteropathies
7.The prognostic factors for neonatal outcomes of pregnancies with preterm premature rupture of membranes.
Hee Young CHO ; Yong Won PARK ; Young Han KIM ; Jae Hak LIM ; Yae Jung MOON
Korean Journal of Obstetrics and Gynecology 2006;49(12):2528-2534
OBJECTIVE: To study the neonatal outcomes of pregnancies complicated with preterm premature rupture of the membranes (PPROM) and to assess the prognostic factors for poor neonatal outcomes. METHODS: We performed a retrospective study of pregnancies registered and managed in Korea from Jan 1st, 2000 to Dec 31st, 2005. The 217 women with preterm premature rupture of membranes in singleton pregnancies from 16 to 36 gestational weeks were enrolled in this study. We investigated respiratory distress syndrome (RDS), intraventricular hemorrhage (IVH), necrotizing enterocolitis (NEC), and neonatal infection for poor neonatal outcomes. Prognostic variables included for analysis were age, BMI, parity, history of PPROM, history of preterm labor, preterm labor, amniotic fluid index, tocolysis, antibiotics, antenatal steroid, hemoglobin, WBC, CRP, gestational age at PPROM, interval from PPROM to delivery, gestational age at delivery, placental findings, and etc. We compared the characteristics of those who developed neonatal morbidity (n=144) with those who did not (n=65) using the Student t test for continuous variables, and the Pearson's chi-square or the Fisher exact test for categorical variables. RESULTS: Total of 217 cases of PPROM were registered and treated expectantly during the study period. The majority of cases were delivered within 48 hours of PPROM. There were 64 cases diagnosed with RDS, 21 with IVH, 3 with NEC, 117 with neonatal infection and 8 neonatal mortalities. The gestational age at PPROM (p<0.000), interval from PPROM to delivery (p<0.036), gestational age at delivery (p<0.000), decreased amniotic fluid index (p<0.048), and high level of WBC, CRP (p<0.010, p<0.049) were significantly correlated with poor neonatal outcomes. Birth weight (p<0.000) and APGAR (p<0.000) score at 5 minute were significantly lower in cases resulting in neonatal morbidity. CONCLUSION: Neonatal outcomes of pregnancies with preterm premature rupture of membranes are related to gestational age at PPROM, interval from PPROM to delivery, and gestational age at delivery. The decreased amniotic fluid index, and high level of WBC, CRP presented before delivery are significantly correlated with poor neonatal outcomes. Birth weight and APGAR scores at 5 minute are significantly lower in cases resulting in neonatal morbidity.
Amniotic Fluid
;
Anti-Bacterial Agents
;
Birth Weight
;
Enterocolitis, Necrotizing
;
Female
;
Gestational Age
;
Hemorrhage
;
Humans
;
Infant
;
Infant Mortality
;
Korea
;
Membranes*
;
Obstetric Labor, Premature
;
Parity
;
Pregnancy*
;
Retrospective Studies
;
Rupture*
;
Tocolysis
8.Assessment of Validity and Reliability of Plantar Pressure in Smart Insole
Ho Won KANG ; Yae Lynn AN ; Dae-Yoo KIM ; Dong-Oh LEE ; Gil Young PARK ; Dong Yeon LEE
Journal of Korean Foot and Ankle Society 2022;26(3):130-135
Purpose:
Smart insoles are wearable devices that are inserted into shoes. Smart insoles with built-in pressure and acceleration sensors can measure the plantar pressure, stride length, and walking speed. This study evaluated the validity and reliability of the plantar pressure measurements of smart insoles during walking on flat ground.
Materials and Methods:
Twenty one subjects were included in this study. After wearing smart insoles, I-SOL® (Gilon, Seongnam, Korea), the subjects walked a 10 m corridor six times at a rate of 100 steps/min, and the middle three steps, free from direction changes, were chosen for data analysis. The same protocol was repeated after wearing Pedar-X (Novel Corporation, Munich, Germany), an insoletype plantar pressure measurement equipment with proven validity. The average maximum pressure (Ppeak , kPa) and the time at which Ppeak appeared (Ptime , %stride) were calculated for each device. The validity of smart insoles was evaluated by using the interclass correlation coefficient (ICC) of Ppeak and Ptime between the two instruments, and Cronbach’s alpha was obtained from the Ppeak values to evaluate the reliability.
Results:
The ICC of Ppeak was 0.651 (good) in the hallux, 0.744 (good) in the medial forefoot, 0.839 (excellent) in the lateral forefoot, and 0.854 (excellent) in the hindfoot. The ICC of Ptime showed 0.868 (excellent) in the hallux, 0.892 (excellent) in the medial forefoot, 0.721 (good) in the lateral forefoot, and 0.832 (excellent) in the hindfoot. All ICC values showed good or excellent results. The Cronbach’s alpha of Ppeak measured in the smart insoles was 0.990 in the hallux, 0.961 in the medial forefoot, 0.973 in the lateral forefoot, and 0.995 in the hindfoot; all indicated excellent reliability in all areas.
Conclusion
The plantar pressure measurements of smart insoles during walking on a flat ground showed validity compared to Pedar-X, and high reliability after repeated measurements.
9.Production and Provision of ContextBased Patient-Specific Educational Resources
Ye-Eun PARK ; Jeong Hyun HAN ; Yae Won TAK ; Yura LEE
Journal of Korean Medical Science 2022;37(43):e316-
Background:
Patient education is generally accompanied by instructive materials. The Korean government has recommended the provision of patient-specific educational materials (PEMs) via an electronic medical record (EMR) certification system. However, there are currently no clear standards or guidelines for including PEMs in current domestic educational materials. We investigated the benefits of integrating PEMs with the EMR certification system and the methods by which this integration can be achieved.
Methods:
We developed and administered three structured Delphi surveys to 26 healthcare providers in clinical settings based on data collected from separate semi-structured advisory interviews with five experts. The surveys included the following topics: 1) expected effects of patient-specific education and health-related notifications/alarms, 2) desirable methods for providing PEMs, and 3) appropriate fee-setting and government support. We distributed the Delphi surveys via e-mail and calculated the average and standard deviation of the survey responses.
Results:
PEMs are expected to have significant educational effects, such as the provision of surgery/intervention-related information, and will improve the understanding of various treatment processes/procedures. The preferred method for providing PEMs was via automatic request after receiving confirmation from healthcare providers. The provision of these materials was based on set fees and government support. The average fee per session was set at approximately USD 23 (as of October 2021, USD 1 = KRW 1,196).
Conclusion
In this study, we investigated the prerequisites, contents, methods, and fees related to the provision of effective and efficient PEMs. The study findings can facilitate the production and provision of PEMs.
10.Radiomics MRI Phenotyping with Machine Learning to Predict the Grade of Lower-Grade Gliomas: A Study Focused on Nonenhancing Tumors
Yae Won PARK ; Yoon Seong CHOI ; Sung Soo AHN ; Jong Hee CHANG ; Se Hoon KIM ; Seung Koo LEE
Korean Journal of Radiology 2019;20(9):1381-1389
OBJECTIVE: To assess whether radiomics features derived from multiparametric MRI can predict the tumor grade of lower-grade gliomas (LGGs; World Health Organization grade II and grade III) and the nonenhancing LGG subgroup. MATERIALS AND METHODS: Two-hundred four patients with LGGs from our institutional cohort were allocated to training (n = 136) and test (n = 68) sets. Postcontrast T1-weighted images, T2-weighted images, and fluid-attenuated inversion recovery images were analyzed to extract 250 radiomics features. Various machine learning classifiers were trained using the radiomics features to predict the glioma grade. The trained classifiers were internally validated on the institutional test set and externally validated on a separate cohort (n = 99) from The Cancer Genome Atlas (TCGA). Classifier performance was assessed by determining the area under the curve (AUC) from receiver operating characteristic curve analysis. An identical process was performed in the nonenhancing LGG subgroup (institutional training set, n = 73; institutional test set, n = 37; and TCGA cohort, n = 37) to predict the glioma grade. RESULTS: The performance of the best classifier was good in the internal validation set (AUC, 0.85) and fair in the external validation set (AUC, 0.72) to predict the LGG grade. For the nonenhancing LGG subgroup, the performance of the best classifier was good in the internal validation set (AUC, 0.82), but poor in the external validation set (AUC, 0.68). CONCLUSION: Radiomics feature-based classifiers may be useful to predict LGG grades. However, radiomics classifiers may have a limited value when applied to the nonenhancing LGG subgroup in a TCGA cohort.
Cohort Studies
;
Genome
;
Glioma
;
Humans
;
Machine Learning
;
Magnetic Resonance Imaging
;
ROC Curve
;
World Health Organization