Combining radiomics with ultrasound-based risk stratification systems for thyroid nodules: an approach for improving performance.

OBJECTIVES: To develop a radiomics score using ultrasound images to predict thyroid malignancy and to investigate its potential as a complementary tool to improve the performance of risk stratification systems. METHODS: We retrospectively included consecutive patients who underwent fine-needle aspiration (FNA) for thyroid nodules that were cytopathologically diagnosed as benign or malignant. Nodules were randomly assigned to a training and test set (8:2 ratio). A radiomics score was developed from the training set, and cutoff values based on the maximum Youden index (Rad_maxY) and for 5%, 10%, and 20% predicted malignancy risk (Rad_5%, Rad_10%, Rad_20%, respectively) were applied to the test set. The performances of the American College of Radiology (ACR) and the American Thyroid Association (ATA) guidelines were compared with the combined performances of the guidelines and radiomics score with interpretations from expert and nonexpert readers. RESULTS: A total of 1624 thyroid nodules from 1609 patients (mean age, 50.1 years [range, 18-90 years]) were included. The radiomics score yielded an AUC of 0.85 (95% CI: 0.83, 0.87) in the training set and 0.75 (95% CI: 0.69, 0.81) in the test set (Rad_maxY). When the radiomics score was combined with the ACR or ATA guidelines (Rad_5%), all readers showed increased specificity, accuracy, and PPV and decreased unnecessary FNA rates (all p < .05), with no difference in sensitivity (p > .05). CONCLUSION: Radiomics help predict thyroid malignancy and improve specificity, accuracy, PPV, and unnecessary FNA rate while maintaining the sensitivity of the ACR and ATA guidelines for both expert and nonexpert readers. KEY POINTS: • The radiomics score yielded an AUC of 0.85 and 0.75 in the training and test set, respectively. • For all readers, combining a 5% predicted malignancy risk cutoff for the radiomics score with the ACR and ATA guidelines significantly increased specificity, accuracy, and PPV and decreased unnecessary FNA rates, with no decrease in sensitivity. • Radiomics can help predict malignancy in thyroid nodules in combination with risk stratification systems, by improving specificity, accuracy, and PPV and unnecessary FNA rates while maintaining sensitivity for both expert and nonexpert readers.

Electric fields in a counterflow nonpremixed flame: measurement and simulation.

In electric field modified flames, the electric body force on fluid elements can play a role in modifying the flow field, affecting flame characteristics by this modified flow motion. Numerical studies have developed ion kinetic mechanisms and appropriate transport models for charged species, validating them with a voltage-current trend in 1D premixed flames. Recent experimental approaches have measured the electric field by adopting the Electric Field Induced Second Harmonic generation (EFISH) technique. However, the quantification has turned out very challenging due to the inherent distortion in the EFISH signal, as well as inhomogeneous temperature and concentration fields in the combustion field. Here, we propose measurement and calibration schemes to quantify the EFISH signal in a laminar counterflow nonpremixed flame and present comparison with numerical results using an in-house multi-physics CFD (Computational Fluid Dynamics) code. Overall, the quantified electric fields agreed well with those from numerical simulation, specifically capturing null electric fields near the flame in the sub-saturated regime due to the electric field screening effect. In the saturated regime, notable discrepancy was found in a fuel stream when electrons moved through it: experiment indicated a significant number of negative ions in the fuel stream, whereas numerical results predicted negligible negative ions, due to the implemented ion-mechanism. This suggested that the experimentally obtained electric fields may serve as validation data for modeling studies to improve transport models and ion-mechanism. In-situ measurement of charged species in the presence of external electric fields should be a future work.

Learnability of Thyroid Nodule Assessment on Ultrasonography: Using a Big Data Set.

OBJECTIVE: The aims of the work described here were to evaluate the learnability of thyroid nodule assessment on ultrasonography (US) using a big data set of US images and to evaluate the diagnostic utilities of artificial intelligence computer-aided diagnosis (AI-CAD) used by readers with varying experience to differentiate benign and malignant thyroid nodules. METHODS: Six college freshmen independently studied the "learning set" composed of images of 13,560 thyroid nodules, and their diagnostic performance was evaluated after their daily learning sessions using the "test set" composed of images of 282 thyroid nodules. The diagnostic performance of two residents and an experienced radiologist was evaluated using the same "test set." After an initial diagnosis, all readers once again evaluated the "test set" with the assistance of AI-CAD. RESULTS: Diagnostic performance of almost all students increased after the learning program. Although the mean areas under the receiver operating characteristic curves (AUROCs) of residents and the experienced radiologist were significantly higher than those of students, the AUROCs of five of the six students did not differ significantly compared with that of the one resident. With the assistance of AI-CAD, sensitivity significantly increased in three students, specificity in one student, accuracy in four students and AUROC in four students. Diagnostic performance of the two residents and the experienced radiologist was better with the assistance of AI-CAD. CONCLUSION: A self-learning method using a big data set of US images has potential as an ancillary tool alongside traditional training methods. With the assistance of AI-CAD, the diagnostic performance of readers with varying experience in thyroid imaging could be further improved.

Radiomics Feature Analysis Using Native T1 Mapping for Discriminating Between Cardiac Tumors and Thrombi.

RATIONALE AND OBJECTIVES: Accurate differential diagnosis is essential because cardiac tumors and thrombi have different prognoses and therapeutic approaches. Native T1 map provides an objective T1 time quantifications of cardiac mass without the need for a contrast agent. We examined the diagnostic performance of radiomics features for differentiating cardiac tumors from thrombi using cardiac magnetic resonance imaging T1 mapping technique compared to that of late gadolinium enhancement (LGE) imaging. MATERIALS AND METHODS: This retrospective study included 22 cardiac tumors and 21 thrombi of 41 patients who underwent cardiac magnetic resonance imaging from December 2013 to May 2018. Fifty-six radiomics features were extracted from native T1 images. The least absolute shrinkage and selection operator method was used for feature selection and rad score extraction. The diagnostic performance of the rad score was compared to that of the native T1 value (mean T1) and LGE ratio. RESULTS: The area under the receiver operating characteristic curve of the rad score was higher than that of the mean T1 and LGE ratio (0.98 vs. 0.86 vs. 0.82, p = 0.001). With the optimal cut-off value, the rad score showed sensitivity, specificity, and accuracy of 95.4%, 95.2%, and 95.2%, respectively. Combination of the rad score and mean T1 showed a significantly higher diagnostic performance than mean T1 (p = 0.019) or LGE ratio (p = 0.022). CONCLUSION: The rad score derived from native T1 maps can differentiate thrombi from tumors better than the mean T1 or LGE ratio. This is valuable for determining a treatment strategy for cardiac lesions in patients who cannot tolerate contrast agents.

Treatment Response Evaluation of Cardiac Amyloidosis Using Serial T1- and T2-Mapping Cardiovascular Magnetic Resonance Imaging.

Amyloidosis is a multisystemic disease characterized by the accumulation of abnormal proteins in extracellular spaces in various organs, with frequent involvement of the myocardium. We report a case of a patient who had cardiac amyloidosis with a trend of reduction in native T1 and T2 values and extracellular volume fraction on serial cardiac magnetic resonance imaging after chemotherapy and stem cell transplantation. The native T1 value and the extracellular volume fraction are closely associated with tissue amyloid burden in amyloidosis patients. This case demonstrated that cardiac magnetic resonance imaging may be used as a non-invasive and quantitative biomarker in the treatment monitoring of amyloidosis.

Prediction of breast cancer molecular subtypes using radiomics signatures of synthetic mammography from digital breast tomosynthesis.

We aimed to predict molecular subtypes of breast cancer using radiomics signatures extracted from synthetic mammography reconstructed from digital breast tomosynthesis (DBT). A total of 365 patients with invasive breast cancer with three different molecular subtypes (luminal A + B, luminal; HER2-positive, HER2; triple-negative, TN) were assigned to the training set and temporally independent validation cohort. A total of 129 radiomics features were extracted from synthetic mammograms. The radiomics signature was built using the elastic-net approach. Clinical features included patient age, lesion size and image features assessed by radiologists. In the validation cohort, the radiomics signature yielded an AUC of 0.838, 0.556, and 0.645 for the TN, HER2 and luminal subtypes, respectively. In a multivariate analysis, the radiomics signature was the only independent predictor of the molecular subtype. The combination of the radiomics signature and clinical features showed significantly higher AUC values than clinical features only for distinguishing the TN subtype. In conclusion, the radiomics signature showed high performance for distinguishing TN breast cancer. Radiomics signatures may serve as biomarkers for TN breast cancer and may help to determine the direction of treatment for these patients.

Differentiation of thyroid nodules on US using features learned and extracted from various convolutional neural networks.

Thyroid nodules are a common clinical problem. Ultrasonography (US) is the main tool used to sensitively diagnose thyroid cancer. Although US is non-invasive and can accurately differentiate benign and malignant thyroid nodules, it is subjective and its results inevitably lack reproducibility. Therefore, to provide objective and reliable information for US assessment, we developed a CADx system that utilizes convolutional neural networks and the machine learning technique. The diagnostic performances of 6 radiologists and 3 representative results obtained from the proposed CADx system were compared and analyzed.

Imaging Features of Hepatocellular Carcinoma: Quantitative and Qualitative Comparison Between MRI-Enhanced With Gd-EOB-DTPA and Gd-DTPA.

OBJECTIVES: The aim of this study was to compare the major imaging features of hepatocellular carcinoma (HCC) on magnetic resonance imaging (MRI) scans with Gd-EOB-DTPA (EOB) and extracellular agent (ECA; Gd-DTPA) contrast media. MATERIALS AND METHODS: Among 184 surgically proven HCCs in 169 patients who underwent a liver MRI with either EOB (n = 120) or ECA (n = 49), 55 HCCs were matched according to tumor size, Edmonson grade (major and worst), and gross type for each of the 2 contrast media. For the qualitative analysis, 2 board-certified radiologists independently reviewed arterial phase hyperenhancement, hypointensity on portal venous phase, hypointensity on delayed or transitional phase (DP/TP, 120-150 seconds), and capsule appearance. For the quantitative analysis, a third radiologist measured the signal intensity at each phase by placing the region of interest for tumor and normal liver parenchyma. The lesion-to-liver contrast (LLC) and lesion-to-liver contrast enhancement ratio (LLCER) were calculated. RESULTS: On qualitative analysis, hypointensity on DP/TP was seen more frequently with EOB (91% in reader 1, 89% in reader 2) than with ECA (73% in reader 1, 75% in reader 2; P = 0.026). Capsule appearance was seen less frequently with EOB (31% in reader 1, 44% in reader 2) than with ECA (73% in reader 1, 78% in reader 2; P < 0.001). On quantitative analysis, the LLC on arterial phase (AP) was better with ECA (P = 0.003), whereas LLC on DP was better with EOB (P < 0.001). The LLCER from precontrast to AP was higher with ECA (P = 0.022), whereas the LLCER from portal venous phase to DP was higher with EOB (P < 0.001). CONCLUSIONS: ECA-MRI revealed better LLC on AP and detection rate of capsule appearance than EOB-MRI. EOB-MRI showed superior LLC on TP.