• Korean Journal of Radiology
• /
• v.24 no.5
• /
• pp.395-405
• /
• 2023

Objective: This study aimed to develop and validate models using radiomics features on a native T1 map from cardiac magnetic resonance (CMR) to predict left ventricular reverse remodeling (LVRR) in patients with nonischemic dilated cardiomyopathy (NIDCM). Materials and Methods: Data from 274 patients with NIDCM who underwent CMR imaging with T1 mapping at Severance Hospital between April 2012 and December 2018 were retrospectively reviewed. Radiomic features were extracted from the native T1 maps. LVRR was determined using echocardiography performed ≥ 180 days after the CMR. The radiomics score was generated using the least absolute shrinkage and selection operator logistic regression models. Clinical, clinical + late gadolinium enhancement (LGE), clinical + radiomics, and clinical + LGE + radiomics models were built using a logistic regression method to predict LVRR. For internal validation of the result, bootstrap validation with 1000 resampling iterations was performed, and the optimism-corrected area under the receiver operating characteristic curve (AUC) with 95% confidence interval (CI) was computed. Model performance was compared using AUC with the DeLong test and bootstrap. Results: Among 274 patients, 123 (44.9%) were classified as LVRR-positive and 151 (55.1%) as LVRR-negative. The optimism-corrected AUC of the radiomics model in internal validation with bootstrapping was 0.753 (95% CI, 0.698-0.813). The clinical + radiomics model revealed a higher optimism-corrected AUC than that of the clinical + LGE model (0.794 vs. 0.716; difference, 0.078 [99% CI, 0.003-0.151]). The clinical + LGE + radiomics model significantly improved the prediction of LVRR compared with the clinical + LGE model (optimism-corrected AUC of 0.811 vs. 0.716; difference, 0.095 [99% CI, 0.022-0.139]). Conclusion: The radiomic characteristics extracted from a non-enhanced T1 map may improve the prediction of LVRR and offer added value over traditional LGE in patients with NIDCM. Additional external validation research is required.

• Korean Journal of Radiology
• /
• v.23 no.8
• /
• pp.811-820
• /
• 2022

Objective: To develop a model incorporating radiomic features and clinical factors to accurately predict acute ischemic stroke (AIS) outcomes. Materials and Methods: Data from 522 AIS patients (382 male [73.2%]; mean age ± standard deviation, 58.9 ± 11.5 years) were randomly divided into the training (n = 311) and validation cohorts (n = 211). According to the modified Rankin Scale (mRS) at 6 months after hospital discharge, prognosis was dichotomized into good (mRS ≤ 2) and poor (mRS > 2); 1310 radiomics features were extracted from diffusion-weighted imaging and apparent diffusion coefficient maps. The minimum redundancy maximum relevance algorithm and the least absolute shrinkage and selection operator logistic regression method were implemented to select the features and establish a radiomics model. Univariable and multivariable logistic regression analyses were performed to identify the clinical factors and construct a clinical model. Ultimately, a multivariable logistic regression analysis incorporating independent clinical factors and radiomics score was implemented to establish the final combined prediction model using a backward step-down selection procedure, and a clinical-radiomics nomogram was developed. The models were evaluated using calibration, receiver operating characteristic (ROC), and decision curve analyses. Results: Age, sex, stroke history, diabetes, baseline mRS, baseline National Institutes of Health Stroke Scale score, and radiomics score were independent predictors of AIS outcomes. The area under the ROC curve of the clinical-radiomics model was 0.868 (95% confidence interval, 0.825-0.910) in the training cohort and 0.890 (0.844-0.936) in the validation cohort, which was significantly larger than that of the clinical or radiomics models. The clinical radiomics nomogram was well calibrated (p > 0.05). The decision curve analysis indicated its clinical usefulness. Conclusion: The clinical-radiomics model outperformed individual clinical or radiomics models and achieved satisfactory performance in predicting AIS outcomes.

• Korean Journal of Radiology
• /
• v.24 no.9
• /
• pp.827-837
• /
• 2023

Objective: To investigate the predictive value of radiomics features based on cardiac magnetic resonance (CMR) cine images for left ventricular adverse remodeling (LVAR) after acute ST-segment elevation myocardial infarction (STEMI). Materials and Methods: We conducted a retrospective, single-center, cohort study involving 244 patients (random-split into 170 and 74 for training and testing, respectively) having an acute STEMI (88.5% males, 57.0 ± 10.3 years of age) who underwent CMR examination at one week and six months after percutaneous coronary intervention. LVAR was defined as a 20% increase in left ventricular end-diastolic volume 6 months after acute STEMI. Radiomics features were extracted from the oneweek CMR cine images using the least absolute shrinkage and selection operator regression (LASSO) analysis. The predictive performance of the selected features was evaluated using receiver operating characteristic curve analysis and the area under the curve (AUC). Results: Nine radiomics features with non-zero coefficients were included in the LASSO regression of the radiomics score (RAD score). Infarct size (odds ratio [OR]: 1.04 (1.00-1.07); P = 0.031) and RAD score (OR: 3.43 (2.34-5.28); P < 0.001) were independent predictors of LVAR. The RAD score predicted LVAR, with an AUC (95% confidence interval [CI]) of 0.82 (0.75-0.89) in the training set and 0.75 (0.62-0.89) in the testing set. Combining the RAD score with infarct size yielded favorable performance in predicting LVAR, with an AUC of 0.84 (0.72-0.95). Moreover, the addition of the RAD score to the left ventricular ejection fraction (LVEF) significantly increased the AUC from 0.68 (0.52-0.84) to 0.82 (0.70-0.93) (P = 0.018), which was also comparable to the prediction provided by the combined microvascular obstruction, infarct size, and LVEF with an AUC of 0.79 (0.65-0.94) (P = 0.727). Conclusion: Radiomics analysis using non-contrast cine CMR can predict LVAR after STEMI independently and incrementally to LVEF and may provide an alternative to traditional CMR parameters.

• Korean Journal of Radiology
• /
• v.23 no.11
• /
• pp.1067-1077
• /
• 2022

Objective: To determine whether Sonazoid-enhanced ultrasound (SZUS) was noninferior to SonoVue-enhanced ultrasound (SVUS) in diagnosing hepatocellular carcinoma (HCC) using the same diagnostic criteria. Materials and Methods: This prospective, single-center, noninferiority study (NCT04847726) enrolled 105 at-risk participants (71 male; mean age ± standard deviation, 63 ± 11 years; range, 26-86 years) with treatment-naïve solid hepatic nodules (≥ 1 cm). All participants underwent same-day SZUS (experimental method) and SVUS (control method) for one representative nodule per participant. Images were interpreted by three readers (the operator and two independent readers). All malignancies were diagnosed histopathologically, while the benignity of other lesions was confirmed by follow-up stability or pathology. The primary endpoint was per-lesion diagnostic accuracy for HCC pooled across three readers using the conventional contrast-enhanced ultrasound diagnostic criteria, including arterial phase hyperenhancement followed by mild (assessed within 2 minutes after contrast injection) and late (≥ 60 seconds with a delay of 5 minutes) washout. The noninferiority delta was -10%p. Furthermore, different time delays were compared as washout criteria in SZUS, including delays of 2, 5, and > 10 minutes. Results: A total of 105 lesions (HCCs [n = 61], non-HCC malignancies [n = 19], and benign [n = 25]) were evaluated. Using the 5-minutes washout criterion, per-lesion accuracy of SZUS pooled across the three readers (72.4%; 95% confidence interval [CI], 64.1%-79.3%) was noninferior to that of SVUS (71.4%; 95% CI, 63.1%-78.6%), meeting the statistical criterion for non-inferiority (difference of 0.95%p; 95% CI, -3.8%p-5.7%p). The arterial phase hyperenhancement combined with the 5-minutes washout criterion showed the same sensitivity as that of the > 10-minutes criterion (59.0% vs. 59.0%, p = 0.989), and the specificities were not significantly different (90.9% vs. 86.4%, p = 0.072). Conclusion: SZUS was noninferior to SVUS for diagnosing HCC in at-risk patients using the same diagnostic criteria. No significant improvement in HCC diagnosis was observed by extending the washout time delay from 5 to 10 minutes.

• Korean Journal of Radiology
• /
• v.23 no.11
• /
• pp.1078-1088
• /
• 2022

Objective: To develop and validate a model using radiomics features from apparent diffusion coefficient (ADC) map to diagnose local tumor recurrence in head and neck squamous cell carcinoma (HNSCC). Materials and Methods: This retrospective study included 285 patients (mean age ± standard deviation, 62 ± 12 years; 220 male, 77.2%), including 215 for training (n = 161) and internal validation (n = 54) and 70 others for external validation, with newly developed contrast-enhancing lesions at the primary cancer site on the surveillance MRI following definitive treatment of HNSCC between January 2014 and October 2019. Of the 215 and 70 patients, 127 and 34, respectively, had local tumor recurrence. Radiomics models using radiomics scores were created separately for T2-weighted imaging (T2WI), contrast-enhanced T1-weighted imaging (CE-T1WI), and ADC maps using non-zero coefficients from the least absolute shrinkage and selection operator in the training set. Receiver operating characteristic (ROC) analysis was used to evaluate the diagnostic performance of each radiomics score and known clinical parameter (age, sex, and clinical stage) in the internal and external validation sets. Results: Five radiomics features from T2WI, six from CE-T1WI, and nine from ADC maps were selected and used to develop the respective radiomics models. The area under ROC curve (AUROC) of ADC radiomics score was 0.76 (95% confidence interval [CI], 0.62-0.89) and 0.77 (95% CI, 0.65-0.88) in the internal and external validation sets, respectively. These were significantly higher than the AUROC values of T2WI (0.53 [95% CI, 0.40-0.67], p = 0.006), CE-T1WI (0.53 [95% CI, 0.40-0.67], p = 0.012), and clinical parameters (0.53 [95% CI, 0.39-0.67], p = 0.021) in the external validation set. Conclusion: The radiomics model using ADC maps exhibited higher diagnostic performance than those of the radiomics models using T2WI or CE-T1WI and clinical parameters in the diagnosis of local tumor recurrence in HNSCC following definitive treatment.

• Journal of the Korean Society of Radiology
• /
• v.84 no.5
• /
• pp.1110-1122
• /
• 2023

Purpose This study aimed to assess the variability of transrectal shear wave elastography (SWE) using a designed phantom. Materials and Methods In a phantom, the SWE values were examined by two radiologists using agarose and emulsion silicone of different sizes (1, 2, and 3 cm) and shapes (round, cubic) at three depths (1, 2, and 3 cm), two region of interest (ROI) and locations (central, peripheral) using two ultrasound machines (A, B from different vendors). Variability was evaluated using the coefficient of variation (CV). Results The CVs decreased with increasing phantom size. Significant changes in SWE values included; agarose phantom at 3 cm depth (p < 0.001; machine A), 1 cm depth (p = 0.01; machine B), emulsion silicone at 2 cm depth (p = 0.047, p = 0.020; both machines). The CVs increased with increasing depth. Significant changes in SWE values included; 1 cm agarose (p = 0.037, p = 0.021; both machines) and 2 cm agarose phantom (p = 0.047; machine A). Significant differences in SWE values were observed between the shapes for emulsion silicone phantom (p = 0.032; machines A) and between ROI locations on machine B (p ≤ 0.001). The SWE values differed significantly between the two machines (p < 0.05). The intra-/inter-operator agreements were excellent (intraclass correlation coefficient > 0.9). Conclusion The phantom size, depth, and different machines affected the variability of transrectal SWE.

• Journal of Korea Water Resources Association
• /
• v.57 no.2
• /
• pp.73-85
• /
• 2024

Predicting water quality of rivers and reservoirs is necessary for the management of water resources. Artificial Neural Networks (ANNs) have been used in many studies to predict water quality with high accuracy. Previous studies have used Gradient Descent (GD)-based optimizers as an optimizer, an operator of ANN that searches parameters. However, GD-based optimizers have the disadvantages of the possibility of local optimal convergence and absence of a solution storage and comparison structure. This study developed improved optimizers to overcome the disadvantages of GD-based optimizers. Proposed optimizers are optimizers that combine adaptive moments (Adam) and Nesterov-accelerated adaptive moments (Nadam), which have low learning errors among GD-based optimizers, with Harmony Search (HS) or Novel Self-adaptive Harmony Search (NSHS). To evaluate the performance of Long Short-Term Memory (LSTM) using improved optimizers, the water quality data from the Dasan water quality monitoring station were used for training and prediction. Comparing the learning results, Mean Squared Error (MSE) of LSTM using Nadam combined with NSHS (NadamNSHS) was the lowest at 0.002921. In addition, the prediction rankings according to MSE and R² for the four water quality indices for each optimizer were compared. Comparing the average of ranking for each optimizer, it was confirmed that LSTM using NadamNSHS was the highest at 2.25.

• Journal of Trauma and Injury
• /
• v.37 no.1
• /
• pp.28-36
• /
• 2024

Purpose: Recent advancements in interventional radiology have made angioembolization an invaluable modality in trauma care. Angioembolization is typically performed by interventional radiologists. In this study, we aimed to investigate the safety and efficacy of emergency angioembolization performed by trauma surgeons. Methods: We identified trauma patients who underwent emergency angiography due to significant trauma-related hemorrhage between January 2020 and June 2023 at Jeju Regional Trauma Center. Until May 2022, two dedicated interventional radiologists performed emergency angiography at our center. However, since June 2022, a trauma surgeon with a background and experience in vascular surgery has performed emergency angiography for trauma-related bleeding. The indications for trauma surgeon-performed angiography included significant hemorrhage from liver injury, pelvic injury, splenic injury, or kidney injury. We assessed the angiography results according to the operator of the initial angiographic procedure. The term "failure of the first angioembolization" was defined as rebleeding from any cause, encompassing patients who underwent either re-embolization due to rebleeding or surgery due to rebleeding. Results: No significant differences were found between the interventional radiologists and the trauma surgeon in terms of re-embolization due to rebleeding, surgery due to rebleeding, or the overall failure rate of the first angioembolization. Mortality and morbidity rates were also similar between the two groups. In a multivariable logistic regression analysis evaluating failure after the first angioembolization, pelvic embolization emerged as the sole significant risk factor (adjusted odds ratio, 3.29; 95% confidence interval, 1.05-10.33; P=0.041). Trauma surgeon-performed angioembolization was not deemed a significant risk factor in the multivariable logistic regression model. Conclusions: Trauma surgeons, when equipped with the necessary endovascular skills and experience, can safely perform angioembolization. To further improve quality control, an enhanced training curriculum for trauma surgeons is warranted.

• The Journal of Korean Academy of Prosthodontics
• /
• v.62 no.2
• /
• pp.146-156
• /
• 2024

Implant assisted removable partial denture (IARPD) has been practiced in various forms for a long time, and among them, implant surveyed crown RPD is gaining predictability as well as being considered as a treatment option for patients with anatomical and financial disadvantages. The position of implant could be divided as posterior placement or anterior placement according to the purpose of the treatment and should be planned in consider to the alveolar ridge of patient, anticipated prognosis of remaining teeth, and opposing dentition. This case report describes a treatment for mandibular Kennedy class I partial edentulous patient with two implant-supported surveyed crown and implant assisted removable partial denture. Given the difficulty of posterior placement in this patient and the prognosis of the residual teeth, the plan was to place two implants in close proximity to the residual teeth, which were placed in the planned position, angle, and depth using guided surgery. The process of fabricating the fixed prosthesis was carried out in parallel with the maxillary edentulous tooth arrangement process to increase predictability, and when fabricating the localized tooth, the implant was designed in a form that allows the patient to perform functional movements by preventing excessive loading as the last supporting tooth, and was fabricated through a secondary impression process. Each treatment procedure was proceeded as planned, with aesthetically and functionally satisfactory results for both patient and operator.

• Korean Journal of Radiology
• /
• v.22 no.3
• /
• pp.334-343
• /
• 2021

Objective: We aimed to develop a prediction model for diagnosing severe aortic stenosis (AS) using computed tomography (CT) radiomics features of aortic valve calcium (AVC) and machine learning (ML) algorithms. Materials and Methods: We retrospectively enrolled 408 patients who underwent cardiac CT between March 2010 and August 2017 and had echocardiographic examinations (240 patients with severe AS on echocardiography [the severe AS group] and 168 patients without severe AS [the non-severe AS group]). Data were divided into a training set (312 patients) and a validation set (96 patients). Using non-contrast-enhanced cardiac CT scans, AVC was segmented, and 128 radiomics features for AVC were extracted. After feature selection was performed with three ML algorithms (least absolute shrinkage and selection operator [LASSO], random forests [RFs], and eXtreme Gradient Boosting [XGBoost]), model classifiers for diagnosing severe AS on echocardiography were developed in combination with three different model classifier methods (logistic regression, RF, and XGBoost). The performance (c-index) of each radiomics prediction model was compared with predictions based on AVC volume and score. Results: The radiomics scores derived from LASSO were significantly different between the severe AS and non-severe AS groups in the validation set (median, 1.563 vs. 0.197, respectively, p < 0.001). A radiomics prediction model based on feature selection by LASSO + model classifier by XGBoost showed the highest c-index of 0.921 (95% confidence interval [CI], 0.869-0.973) in the validation set. Compared to prediction models based on AVC volume and score (c-indexes of 0.894 [95% CI, 0.815-0.948] and 0.899 [95% CI, 0.820-0.951], respectively), eight and three of the nine radiomics prediction models showed higher discrimination abilities for severe AS. However, the differences were not statistically significant (p > 0.05 for all). Conclusion: Models based on the radiomics features of AVC and ML algorithms may perform well for diagnosing severe AS, but the added value compared to AVC volume and score should be investigated further.