Comparison of Machine Learning Models Using Diffusion-Weighted Images for Pathological Grade of Intrahepatic Mass-Forming Cholangiocarcinoma.

Is the radiomic approach, utilizing diffusion-weighted imaging (DWI), capable of predicting the various pathological grades of intrahepatic mass-forming cholangiocarcinoma (IMCC)? Furthermore, which model demonstrates superior performance among the diverse algorithms currently available? The objective of our study is to develop DWI radiomic models based on different machine learning algorithms and identify the optimal prediction model. We undertook a retrospective analysis of the DWI data of 77 patients with IMCC confirmed by pathological testing. Fifty-seven patients initially included in the study were randomly assigned to either the training set or the validation set in a ratio of 7:3. We established four different classifier models, namely random forest (RF), support vector machines (SVM), logistic regression (LR), and gradient boosting decision tree (GBDT), by manually contouring the region of interest and extracting prominent radiomic features. An external validation of the model was performed with the DWI data of 20 patients with IMCC who were subsequently included in the study. The area under the receiver operating curve (AUC), accuracy (ACC), precision (PRE), sensitivity (REC), and F1 score were used to evaluate the diagnostic performance of the model. Following the process of feature selection, a total of nine features were retained, with skewness being the most crucial radiomic feature demonstrating the highest diagnostic performance, followed by Gray Level Co-occurrence Matrix lmc1 (glcm-lmc1) and kurtosis, whose diagnostic performances were slightly inferior to skewness. Skewness and kurtosis showed a negative correlation with the pathological grading of IMCC, while glcm-lmc1 exhibited a positive correlation with the IMCC pathological grade. Compared with the other three models, the SVM radiomic model had the best diagnostic performance with an AUC of 0.957, an accuracy of 88.2%, a sensitivity of 85.7%, a precision of 85.7%, and an F1 score of 85.7% in the training set, as well as an AUC of 0.829, an accuracy of 76.5%, a sensitivity of 71.4%, a precision of 71.4%, and an F1 score of 71.4% in the external validation set. The DWI-based radiomic model proved to be efficacious in predicting the pathological grade of IMCC. The model with the SVM classifier algorithm had the best prediction efficiency and robustness. Consequently, this SVM-based model can be further explored as an option for a non-invasive preoperative prediction method in clinical practice.

Using machine learning-based radiomics to differentiate between glioma and solitary brain metastasis from lung cancer and its subtypes.

OBJECTIVE: To establish a machine learning-based radiomics model to differentiate between glioma and solitary brain metastasis from lung cancer and its subtypes, thereby achieving accurate preoperative classification. MATERIALS AND METHODS: A retrospective analysis was conducted on MRI T1WI-enhanced images of 105 patients with glioma and 172 patients with solitary brain metastasis from lung cancer, which were confirmed pathologically. The patients were divided into the training group and validation group in an 8:2 ratio for image segmentation, extraction, and filtering; multiple layer perceptron (MLP), support vector machine (SVM), random forest (RF), and logistic regression (LR) were used for modeling; fivefold cross-validation was used to train the model; the validation group was used to evaluate and assess the predictive performance of the model, ROC curve was used to calculate the accuracy, sensitivity, and specificity of the model, and the area under curve (AUC) was used to assess the predictive performance of the model. RESULTS: The accuracy and AUC of the MLP differentiation model for high-grade glioma and solitary brain metastasis in the validation group was 0.992, 1.000, respectively, while the sensitivity and specificity were 1.000, 0.968, respectively. The accuracy and AUC for the MLP and SVM differentiation model for high-grade glioma and small cell lung cancer brain metastasis in the validation group was 0.966, 1.000, respectively, while the sensitivity and specificity were 1.000, 0.929, respectively. The accuracy and AUC for the MLP differentiation model for high-grade glioma and non-small cell lung cancer brain metastasis in the validation group was 0.982, 0.999, respectively, while the sensitivity and specificity were 0.958, 1.000, respectively. CONCLUSION: The application of machine learning-based radiomics has a certain clinical value in differentiating glioma from solitary brain metastasis from lung cancer and its subtypes. In the HGG/SBM and HGG/NSCLC SBM validation groups, the MLP model had the best diagnostic performance, while in the HGG/SCLC SBM validation group, the MLP and SVM models had the best diagnostic performance.

Corrigendum: DWI Combined With Hepatobiliary Phase Enhanced Imaging Can Better Differentiate Cholangiocarcinoma From Atypical Liver Abscesses.

[This corrects the article DOI: 10.3389/fonc.2022.723089.].

DWI Combined With Hepatobiliary-Phase Enhanced Imaging Can Better Differentiate Cholangiocarcinoma From Atypical Liver Abscesses.

Objective: To investigate the value of diffusion-weighted imaging (DWI) combined with the hepatobiliary phase (HBP) Gd-BOPTA enhancement in differentiating intrahepatic mass-forming cholangiocarcinoma (IMCC) from atypical liver abscess. Materials and Methods: A retrospective analysis was performed on 43 patients with IMCCs (IMCC group) and 25 patients with atypical liver abscesses (liver abscess group). The DWI signal, the absolute value of the contrast noise ratio (│CNR│) at the HBP, and visibility were analyzed. Results: A relatively high DWI signal and a relatively high peripheral signal were presented in 29 patients (67.5%) in the IMCC group, and a relatively high DWI signal was displayed in 15 patients (60.0%) in the atypical abscess group with a relatively high peripheral signal in only one (6.7%) patient and a relatively high central signal in 14 (93.3%, 14/15). A significant (P<0.001) difference existed in the pattern of signal between the two groups of patients. On T2WI, IMCC was mainly manifested by homogeneous signal (53.5%), whereas atypical liver abscesses were mainly manifested by heterogeneous signal and relatively high central signal (32%, and 64%), with a significant difference (P<0.001) in T2WI imaging presentation between the two groups. On the HBP imaging, there was a statistically significant difference in peripheral │CNR│ (P< 0.001) and visibility between two groups. The sensitivity of the HBP imaging was significantly (P=0.002) higher than that of DWI. The sensitivity and accuracy of DWI combined with enhanced HBP imaging were significantly (P=0.002 and P<0.001) higher than those of either HBP imaging or DWI alone. Conclusion: Intrahepatic mass-forming cholangiocarcinoma and atypical liver abscesses exhibit different imaging signals, and combination of DWI and hepatobiliary-phase enhanced imaging has higher sensitivity and accuracy than either technique in differentiating intrahepatic mass-forming cholangiocarcinoma from atypical liver abscesses.

Values of MRI Imaging Presentations in the Hepatobiliary Phase, DWI and T2WI Sequences in Predicting Pathological Grades of Intrahepatic Mass-Forming Cholangiocarcinoma.

Objective: To retrospectively investigate the value of various MRI image menifestations in the hepatobiliary phase (HBP), DWI and T2WI sequences in predicting the pathological grades of intrahepatic mass-forming cholangiocarcinoma (IMCC). Materials and Methods: Forty-three patients of IMCCs confirmed by pathology were enrolled including 25 cases in well- or moderately-differentiated group and 18 cases in poorly-differentiated group. All patients underwent DWI, T2WI and HBP scan. The Chi square test was used to compare the differences in the general information. Logistic regression analysis was used to analyze the risk factors in predicting the pathological grade of IMCCs. Results: The maximal diameter of the IMCC lesion was < 3 cm in 11 patients, between 3 cm and 6 cm in 15, and > 6 cm in 17. Sixteen cases had intrahepatic metastasis, including 5 in the well- or moderately-differentiated group and 11 in the poorly-differentiated group. Seventeen (39.5%) patients presented with target signs in the DWI sequence, including 9 in the well- or moderately-differentiated group and 8 in the poorly-differentiated group. Twenty (46.5%) patients presented with target signs in the T2WI sequence, including 8 in the well- or moderately-differentiated group and 12 in the poorly-differentiated group. Nineteen cases (54.3%) had a complete hypointense signal ring, including 13 in the well- or moderately-differentiated group and 6 in the poorly-differentiated group. Sixteen (45.7%) cases had an incomplete hypointense signal ring, including 5 in the well- or moderately-differentiated group and 11 in the poorly-differentiated group. The lesion size, intrahepatic metastasis, T2WI signal, and integrity of a hypointense signal ring in HBP were statistically significantly different between two gourps. T2WI signal, presence or non-presence of intrahepatic metastasis, and integrity of hypointense signal ring were the independent influencing factors for pathological grade of IMCC. Conclusion: Target sign in T2WI sequence, presence of intrahepatic metastasis and an incomplete hypointense-signal ring in HBP are more likely to be present in poorly-differentiated IMCCs.

Retrospective Analysis of the Value of Enhanced CT Radiomics Analysis in the Differential Diagnosis Between Pancreatic Cancer and Chronic Pancreatitis.

PURPOSE: To investigate the feasibility of enhanced computed tomography (CT) radiomics analysis to differentiate between pancreatic cancer (PC) and chronic pancreatitis. METHODS AND MATERIALS: The CT images of 151 PCs and 24 chronic pancreatitis were retrospectively analyzed in the three-dimensional regions of interest on arterial phase (AP) and venous phase (VP) and segmented by MITK software. A multivariable logistic regression model was established based on the selected radiomics features. The radiomics score was calculated, and the nomogram was established. The discrimination of each model was analyzed by the receiver operating characteristic curve (ROC). Decision curve analysis (DCA) was used to evaluate clinical utility. The precision recall curve (PRC) was used to evaluate whether the model is affected by data imbalance. The Delong test was adopted to compare the diagnostic efficiency of each model. RESULTS: Significant differences were observed in the distribution of gender (P = 0.034), carbohydrate antigen 19-9 (P < 0.001), and carcinoembryonic antigen (P < 0.001) in patients with PC and chronic pancreatitis. The area under the ROC curve (AUC) value of AP multivariate regression model, VP multivariate regression model, AP combined with VP features model (Radiomics), clinical feature model, and radiomics combined with clinical feature model (COMB) was 0.905, 0.941, 0.941, 0.822, and 0.980, respectively. The sensitivity and specificity of the COMB model were 0.947 and 0.917, respectively. The results of DCA showed that the COMB model exhibited net clinical benefits and PRC shows that COMB model have good precision and recall (sensitivity). CONCLUSION: The COMB model could be a potential tool to distinguish PC from chronic pancreatitis and aid in clinical decisions.

Dynamic Imaging Changes of COVID-19 Pneumonia at Different Stages.

INTRODUCTION: To investigate the Computed Tomography (CT) imaging characteristics and dynamic changes of COVID-19 pneumonia at different stages. METHODS: Forty-six patients infected with COVID-19 who had chest CT scans were enrolled, and CT scans were performed 4-6 times with an interval of 2-5 days. RESULTS: At the early stage (n=25), ground glass opacity was presented in 11 patients (11/25 or 44.0 %) and ground glass opacity mixed with consolidation in 13 (13/25 or 52.0 %) in the lung CT images. At the progressive stage (n=38), ground glass opacity was presented in only one patient (1/38 or 2.6 %) and ground glass opacity mixed with consolidation in 33 (33/38 or 86.8 %). In the early improvement stage (n=38), the imaging presentation was ground glass opacity alone in three patients (3/38 or 7.9 %) and ground glass opacity mixed with consolidation in 34 (34/38 or 89.5 %). In the late improvement (absorption) stage (n=33), the primary imaging presentation was ground glass presentation in eight patients (8/33 or 24.2 %) and ground glass opacity mixed with consolidation in 23 (23/33 or 69.7 %). The lesion reached the peak at 4-16 days after disease onset, and 26 (26/38 or 68.4 %) patients reached the disease peak within ten days. Starting from 6 to 20 days after onset, the disease began to be improved, with 30 (30/38 or 78.9 %) patients being improved within 15 days. CONCLUSION: COVID-19 pneumonia will progress to the peak stage at a mediate time of seven days and enter the improvement stage at twelve days. Computed tomography imaging of the pulmonary lesion has a common pattern from disease onset to improvement and recovery and provides important information for evaluation of the disease course and treatment effect.

Differences in Clinical and Imaging Features between Asymptomatic and Symptomatic COVID-19 Patients.

Objectives: The clinical and imaging features of asymptomatic carriers of severe acute respiratory syndrome coronavirus 2 and symptomatic COVID-19 patients. Methods: The clinical and chest computed tomography imaging data of 47 asymptomatic carriers and 36 symptomatic COVID-19 patients were derived. All patients underwent 4-6 CT scans over a period of 2-5 days. Results: The bulk of asymptomatic carriers who developed symptoms and most of the COVID-19 patients were older than 18 years of age with a decreased lymphocyte count, abnormal hepatic and renal function, and increased D-dimer and C-reactive protein. In the early stage, the pulmonary lesion involved mostly 1-2 lobes at the peripheral area in asymptomatic carriers but more than three lobes at both the central and peripheral areas in COVID-19 patients. In the progression stage, the lesion of asymptomatic carriers extended from the peripheral to the central area, and no significant difference was found in the lesion range compared with the symptomatic control group. In early improvement stage, the lesion was rapidly absorbed, and lesions were located primarily at the peripheral area in asymptomatic carriers; contrastingly, lesions were primarily located at both the central and peripheral areas in symptomatic patients. Asymptomatic carriers reflected a significantly shorter duration from disease onset to peak progression stage compared with the symptomatic. Conclusions: Asymptomatic carriers are a potential source of transmission and may become symptomatic COVID-19 patients despite indicating less severe pulmonary damage, earlier improvement, and better prognosis. Early isolation and intervention can eliminate such carriers as potential sources of transmission and improve their prognosis.

Differentiating Between an Atypical Hepatic Abscess and Tumor Metastasis Using Magnetic Resonance Imaging and Hepatobiliary Phase Imaging.

OBJECTIVE: To identify diffusion-weighted imaging (DWI) patterns and conspicuity discrepancies on hepatobiliary phase imaging (HBPI) to distinguish atypical hepatic abscesses from hepatic metastases. MATERIALS AND METHODS: This retrospective study recruited 31 patients with 43 atypical hepatic abscesses and 32 patients with 35 hepatic metastases who underwent gadobenate dimeglumine-enhanced magnetic resonance imaging. All lesions were confirmed by pathological or clinical diagnosis. For the qualitative and quantitative analyses, the signal intensity, DWI pattern, apparent diffusion coefficient, degree of perilesional edema, perilesional hyperemia, perilesional signal on HBPI, conspicuity, size discrepancy between sequences, contrast-to-noise ratio, signal-to-noise ratio, and relative enhancement ratio on dynamic phases were independently assessed by two radiologists. Significant findings for differentiating the two groups were identified via univariate and multivariate analyses with a nomogram for predicting atypical hepatic abscesses. The interobserver agreement was also analyzed for each variable. RESULTS: The multivariate analysis revealed that the conspicuity discrepancy (odds ratio [OR] 34.78, 95% confidence interval [CI] 2.09-579.47, p = 0.013) and non-peripheral high signal intensity (SI) rim on DWI (OR 67.46, 95% CI 2.64, 1723.20, p = 0.011) were significant independent factors for predicting atypical hepatic abscesses. They were also shown to be high predictor points on the nomogram. When any of the set criteria were satisfied, 97.7% of atypical hepatic abscesses were correctly identified, with a specificity of 65.7%. When both criteria were combined, the specificity was up to 100%, with a sensitivity of 44.9%. CONCLUSION: Conspicuity discrepancy and a non-peripheral high SI rim on DWI are reliable and meaningful features that can distinguish atypical hepatic abscesses from hepatic metastases.

Clinical and imaging features of pediatric COVID-19.

BACKGROUND: Pediatric COVID-19 is relatively mild and may vary from that in adults. This study was to investigate the epidemic, clinical, and imaging features of pediatric COVID-19 pneumonia for early diagnosis and treatment. METHODS: Forty-one children infected with COVID-19 were analyzed in the epidemic, clinical and imaging data. RESULTS: Among 30 children with mild COVID-19, seven had no symptoms, fifteen had low or mediate fever, and eight presented with cough, nasal congestion, diarrhea, headache, or fatigue. Among eleven children with moderate COVID-19, nine presented with low or mediate fever, accompanied with cough and runny nose, and two had no symptoms. Significantly (P < 0.05) more children had a greater rate of cough in moderate than in mild COVID-19. Thirty children with mild COVID-19 were negative in pulmonary CT imaging, whereas eleven children with moderate COVID-19 had pulmonary lesions, including ground glass opacity in ten (90.9%), patches of high density in six (54.5%), consolidation in three (27.3%), and enlarged bronchovascular bundles in seven (63.6%). The lesions were distributed along the bronchus in five patients (45.5%). The lymph nodes were enlarged in the pulmonary hilum in two patients (18.2%). The lesions were presented in the right upper lobe in two patients (18.1%), right middle lobe in one (9.1%), right lower lobe in six (54.5%), left upper lobe in five (45.5%), and left lower lobe in eight (72.7%). CONCLUSIONS: Children with COVID-19 have mild or moderate clinical and imaging presentations. A better understanding of the clinical and CT imaging helps ascertaining those with negative nucleic acid and reducing misdiagnosis rate for those with atypical and concealed symptoms.

Effects of Drying Methods on the Content, Structural Isomers, and Composition of Astaxanthin in Antarctic Krill.

Antarctic krill (Euphausia superba) is one of the important bioresources in Antarctic waters, containing many bioactives (e.g., astaxanthin), which have a highly potential value for commercial exploitation. In this study, the effects of processing methods on the content, structural isomers, and composition of astaxanthins (free astaxanthin and astaxanthin esters) were studied. Three drying methods, comprising freeze-drying, microwave drying, and hot-air drying, were used. Free astaxanthin (Ast), astaxanthin monoesters (AM), and astaxanthin diesters (AD) in boiled krill (control) and dried krill were extracted and analyzed using high-resolution mass spectrometry with ultraviolet detection. After the three processes, total astaxanthin loss ranged from 8.6 to 64.9%, and the AM and AD contents ranged from 78.3 to 16.6 and 168.7 to 90.5 µg/g, respectively. Compared to other kinds of astaxanthin esters, astaxanthin esters, which linked to eicosapentaenoic acid and docosahexaenoic acid, as well as the Ast, were more easily degraded, and AM was more susceptible to degradation than AD. All-E-astaxanthin easily transformed to the 13Z-astaxanthin than to the 9Z-astaxanthin during the drying process, but the proportions of optical isomers changed due to drying by no more than 5%. The results suggested that freeze-drying, low-power microwave drying (≤1 kW), and low-temperature hot-air drying (≤60 °C) are optimal drying methods for ensuring the quality of krill products.

[An authentication method of bear bile powder based on the near infrared spectroscopy].

A method based on the near infrared (NIR) spectroscopy was established for the authentication of bear bile powder and the quantitative analysis of the proportion in the counterfeits. The NIR spectra of 30 bear bile powder samples from 4 different origins, 6 pig gall powder samples, and 65 mixed samples of bear bile powder and pig gall powder with different proportions range from 0 to 100% were collected. The principal component analysis (PCA) was done and well-marked boundaries were observed between the bear bile and the counterfeits, also the bear bile samples from different origins. Partial least squares discriminant analysis (PLS-DA) was used to differentiate the bear bile and the counterfeits, and the rate of accuracy is 95%. Finally, the PLS calibration model was established for the quantitative analysis of the bear bile proportion in the counterfeits, the coefficient of determination of the prediction set is 0.9759, and the root-mean-square prediction error is 4.25%, which indicated that the result was satisfied. This research contributes to the rapid and undamaged identification of bear bile powder, and also offers reference for the quality identification of other expensive medicinal materials.