Preoperative prediction of lymph node metastasis in colorectal cancer using 18F-FDG PET/CT peritumoral radiomics analysis.

BACKGROUND: Radiomics has been used in the diagnosis of tumor lymph node metastasis (LNM). However, to date, most studies have been based on intratumoral radiomics. Few studies have focused on the use of 18F-fluorodeoxyglucose positron emission computed tomography (18F-FDG PET/CT) peritumoral radiomics for the diagnosis of LNM in colorectal cancer (CRC). PURPOSE: Determining the value of radiomics features extracted from 18F-FDG PET/CT images of the peritumoral region in predicting LNM in patients with CRC. METHODS: The clinical data and preoperative 18F-FDG PET/CT images of 244 CRC patients were retrospectively analyzed. Intratumoral and peritumoral radiomics features were screened using the mutual information method, and least absolute shrinkage and selection operator regression. Based on the selected radiomics features, a radiomics score (Rad-score) was calculated, and independent risk factors obtained from univariate and multivariate logistic regression analyses were used to construct clinical and combined (Radiomics + Clinical) models. The performance of these models was evaluated using the DeLong test, while their clinical utility was assessed by decision curve analysis. Finally, a nomogram was constructed to visualize the predictive model. RESULTS: The most optimal set of features retained by the feature filtering process were all peritumoral radiomic features. Carcinoembryonic antigen levels, PET/CT-reported lymph node status and Rad-score were found to be independent risk factors for LNM. All three LNM risk assessment models exhibited good predictive performance, with the combined model showing the best classification results, with areas under the curve of 0.85 and 0.76 in the training and validation groups, respectively. The DeLong test revealed that the performance of the combined model was superior to that of the clinical and radiomics models in both the training and validation groups, although this difference was only statistically significant in the training group. DCA indicated that the combined model displayed better clinical utility. CONCLUSIONS: 18F-FDG PET/CT peritumoral radiomics is uniquely suited to predict the presence of LNM in patients with CRC. In particular, the predictive efficacy of LNM for precision therapy and individualized patient management can be improved by using a combination of clinical risk factors.

Using tumor habitat-derived radiomic analysis during pretreatment 18F-FDG PET for predicting KRAS/NRAS/BRAF mutations in colorectal cancer.

BACKGROUND: To investigate the association between Kirsten rat sarcoma viral oncogene homolog (KRAS) / neuroblastoma rat sarcoma viral oncogene homolog (NRAS) /v-raf murine sarcoma viral oncogene homolog B (BRAF) mutations and the tumor habitat-derived radiomic features obtained during pretreatment 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) in patients with colorectal cancer (CRC). METHODS: We retrospectively enrolled 62 patients with CRC who had undergone 18F-FDG PET/computed tomography from January 2017 to July 2022 before the initiation of therapy. The patients were randomly split into training and validation cohorts with a ratio of 6:4. The whole tumor region radiomic features, habitat-derived radiomic features, and metabolic parameters were extracted from 18F-FDG PET images. After reducing the feature dimension and selecting meaningful features, we constructed a hierarchical model of KRAS/NRAS/BRAF mutations by using the support vector machine. The convergence of the model was evaluated by using learning curve, and its performance was assessed based on the area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve analysis. The SHapley Additive exPlanation was used to interpret the contributions of various features to predictions of the model. RESULTS: The model constructed by using habitat-derived radiomic features had adequate predictive power with respect to KRAS/NRAS/BRAF mutations, with an AUC of 0.759 (95% CI: 0.585-0.909) on the training cohort and that of 0.701 (95% CI: 0.468-0.916) on the validation cohort. The model exhibited good convergence, suitable calibration, and clinical application value. The results of the SHapley Additive explanation showed that the peritumoral habitat and a high_metabolism habitat had the greatest impact on predictions of the model. No meaningful whole tumor region radiomic features or metabolic parameters were retained during feature selection. CONCLUSION: The habitat-derived radiomic features were found to be helpful in stratifying the status of KRAS/NRAS/BRAF in CRC patients. The approach proposed here has significant implications for adjuvant treatment decisions in patients with CRC, and needs to be further validated on a larger prospective cohort.

Non-invasively Discriminating the Pathological Subtypes of Non-small Cell Lung Cancer with Pretreatment 18F-FDG PET/CT Using Deep Learning.

RATIONALE AND OBJECTIVES: To develop an end-to-end deep learning (DL) model for non-invasively predicting non-small cell lung cancer (NSCLC) pathological subtypes based on 18F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) images, and to explore the potential value of DL technology. MATERIALS AND METHODS: Preoperative 18F-FDG PET/CT images of 189 patients with NSCLC were retrospectively collected. The whole cohort was randomly divided into a training cohort, a validation cohort, and an internal/extended test cohort at the ratio of 6:2:2 after preprocessing the images. In the training and validation cohorts, seven DL models-Shufflenet, VGG16, Googlenet, Inception v3, Resnet50, Densenet201, and Mobilenet v2-were trained and optimized. The generalization ability and clinical utility of the optimal model were evaluated in the internal and extended test cohorts. Moreover, Spearman's correlation analysis was used to evaluate the correlation between DL features and traditional radiological features such as tumor size and maximum standardized uptake values (SUVmax). RESULTS: Some DL features were significantly correlated with SUVmax and tumor size (P < 0.05). The Mobilenet v2 model achieved the best performance during the model development and validation phases. In the internal test group (area under the receiver operating characteristic curve [AUC]: 0.744, area under the precision-recall curve [AP]: 0.759) and extended test group (AUC: 0.767, AP: 0.768), the Mobilenet v2 model showed good generalization ability and reproducibility. Meanwhile, the decision curve analysis revealed that patients can benefit from the decisions made based on the Mobilenet v2 model. CONCLUSION: DL models offer great potential for classifying NSCLC pathological subtypes. Specifically, the Mobilenet v2 model performs well at end-to-end non-invasive pathological subtype stratification of NSCLC.

Enhancing the clinical diagnosis of the acute and subacute phases of autoimmune encephalitis and predicting the risk factors: the potential advantages of 18F-FDG PET/CT.

BACKGROUND: 2-deoxy-2-[18F]fluoro-D-glucose positron emission tomography (18F-FDG PET) could help evaluate metabolic abnormalities by semi-quantitative measurement to identify autoimmune encephalitis (AE). Few studies have been conducted to analyze the prognostic factors of AE. The study aimed to explore the values of diagnosis and treatment evaluation by 18F-FDG PET and preliminarily discussed the potential value in predicting the prognosis of AE patients. METHODS: AE patients underwent 18F-FDG PET/CT and magnetic resonance imaging (MRI). There were two steps to analyse 18F-FDG PET imaging data. The first step was visual assessment. The second step was to analyse 18F-FDG PET parameters using Scenium software (Siemens Molecular Imaging Ltd). The mean standardized uptake value (SUVmean) and maximum standardized uptake value (SUVmax) of brain relative regional metabolism (BRRM) were quantified in the case and control groups according to the anatomical automatic labeling (AAL) partition. The main statistical method was the Kruskal-Wallis test. Finally, the simple linear regression method was used to analyse the relationships between 18F-FDG PET parameters and the modified Rankin Scale (mRS) scores before and after treatment. RESULTS: The results on 18F-FDG PET showed that visual assessment abnormalities were in the mesial temporal lobe (MTL) (70.8%), (mainly infringing on the hippocampus and amygdala), basal ganglia (62.5%), frontal lobes (37.5%), occipital lobes (29.2%), and parietal lobes (12.5%). The positive rate of abnormalities on 18F-FDG PET was more sensitive than that on MRI (95.5% vs 32.2%, p = 0.001). The number of lesions on PET was positively correlated with the mRS scores before and after treatment, and the correlation before treatment was more significant. Before treatment, the SUVmean of the left occipital lobe was the most remarkable (SUVmean, R2 = 0.082, p > 0.05) factor associated with the mRS score, and the correlation was negative. With regard to prognosis, the SUVmax of the MTL was the most notable (R2 = 0.1471, p > 0.05) factor associated with the mRS score after treatment, and the correlation was positive. CONCLUSIONS: 18F-FDG PET could be more sensitive and informative than MRI in the early phases of AE. The common pattern of AE was high MTL metabolism on 18F-FDG PET, which was associated with hypometabolism of the occipital lobe, and the number of lesions on PET before treatment may be significant factors in assessing disease severity. The SUVmax of MTL hypermetabolism may serve as a prognostic biomarker in AE.

Clinical application of Al18F-NOTA-FAPI PET/CT in diagnosis and TNM staging of pancreatic adenocarcinoma, compared to 18F-FDG.

PURPOSE: This study aimed to investigate the ability of Al18F-NOTA-FAPI PET/CT to diagnose pancreatic carcinoma and tumor-associated inflammation with the comparison of 18F-FDG PET/CT. METHODS: Prospective analysis of Al18F-NOTA-FAPI PET/CT and 18F-FDG PET/CT scans of 31 patients from 05/2021 to 05/2022 were analyzed. Al18F-NOTA-FAPI imaging was performed in patients who had Ce-CT and FDG PET/CT and the diagnosis was still unclear. Follow-up histopathology or radiographic examination confirmed the findings. Radiotracer uptake, diagnostic performance, and TNM (tumor-node-metastasis) classifications were compared. RESULTS: A total of 31 patients with pancreatic carcinoma (all were adenocarcinoma) underwent Al18F-NOTA-FAPI-04 PET/CT, including 20 male and 11 female patients, with a mean age of 58.2 ± 8.5 years. FAPI-04 PET/CT imaging showed a higher value of SUVmax-15min/30min/60min, SUVmean-15min/30min/60min, TBR1, and TBR2 in pancreatic carcinoma than FDG (all P < 0.01). The mean level of Al18F-NOTA FAPI-04 uptake values of the pancreatic ductal adenocarcinoma was higher than that of pancreatitis in both SUVmax-30min (P < 0.01), SUVmean-30min (P < 0.05), SUVmax-60min (P < 0.01), and SUVmean-60min (P < 0.01). The FAPI △SUVmax-1, △SUVmax-2, and △SUVmean-2 uptake values of pancreatic carcinoma were higher than tumor-associated inflammation (all P < 0.01). TNM staging of 16/31 patients changed after Al18F-NOTA FAPI-04 PET/CT examination with all upstaging changes. CONCLUSION: Al18F-NOTA-FAPI-04 PET/CT at 15 and 30 min also demonstrated an equivalent detection ability of pancreatic lesion to 18F-FDG PET/CT. Delayed-phase Al18F-NOTA-FAPI-04 PET/CT can help differentiate pancreatic carcinoma and tumor-associated inflammation. Al18F-NOTA FAPI-04 PET/CT also performed better than FDG PET/CT in TNM staging. TRIAL REGISTRATION: Chinese Clinical Trial Registry, ChiCTR2100051406. Registered 23 September 2021, https://www.chictr.org.cn/showproj.html?proj=133033.

A clinical study on relationship between visualization of cardiac fibroblast activation protein activity by Al18F-NOTA-FAPI-04 positron emission tomography and cardiovascular disease.

Objective: FAP plays a vital role in myocardial injury and fibrosis. Although initially used to study imaging of primary and metastatic tumors, the use of FAPI tracers has recently been studied in cardiac remodeling after myocardial infarction. The study aimed to investigate the application of FAPI PET/CT imaging in human myocardial fibrosis and its relationship with clinical factors. Materials and methods: Retrospective analysis of FAPI PET/CT scans of twenty-one oncological patients from 05/2021 to 03/2022 with visual uptake of FAPI in the myocardium were applying the American Heart Association 17-segment model of the left ventricle. The patients' general data, echocardiography, and laboratory examination results were collected, and the correlation between PET imaging data and the above data was analyzed. Linear regression models, Kendall's TaU-B test, the Spearman test, and the Mann-Whitney U test were used for the statistical analysis. Results: 21 patients (60.1 ± 9.4 years; 17 men) were evaluated with an overall mean LVEF of 59.3 ± 5.4%. The calcific plaque burden of LAD, LCX, and RCA are 14 (66.7%), 12 (57.1%), and 9 (42.9%). High left ventricular SUVmax correlated with BMI (P < 0.05) and blood glucose level (P < 0.05), and TBR correlated with age (P < 0.05). A strong correlation was demonstrated between SUVmean and CTnImax (r = 0.711, P < 0.01). Negative correlation of SUVmean and LVEF (r = -0.61, P < 0.01), SUVmax and LVEF (r = -0.65, P < 0.01) were found. ROC curve for predicting calcified plaques by myocardial FAPI uptake (SUVmean) in LAD, LCX, and RCA territory showed AUCs were 0.786, 0.759, and 0.769. Conclusion: FAPI PET/CT scans might be used as a new potential method to evaluate cardiac fibrosis to help patients' management further. FAPI PET imaging can reflect the process of myocardial fibrosis. High FAPI uptakes correlate with cardiovascular risk factors and the distribution of coronary plaques.

A nomogram model based on MRI and radiomic features developed and validated for the evaluation of lymph node metastasis in patients with rectal cancer.

PURPOSE: The aim of this study was to develop and validate a nomogram model to evaluate lymph node metastasis (LNM) in patients with rectal cancer (RC). METHODS: A total of 162 patients with RC were included in the study. The MRI reported model, the Radscore model, and the Complex model were constructed using the logistics regression (LR) algorithm. The DeLong test and decision curve analysis (DCA) were used to compare the prediction performance and clinical utility of these models. The nomogram model was constructed to visualize the prediction results of the best model. Model performance was evaluated in the training and validation groups, and the calibration curve and Hosmer-Lemeshow goodness of fit test were used to evaluate the calibration. RESULT: All three models constructed by the LR algorithm were good at identifying LNM. The DeLong test and the DCA results showed that the Complex model outperformed the MRI reported model and the Radscore model in relation to their predictive performance and clinical utility. The nomogram of the Complex model had an area under the curve (AUC) of 0.902 (95% confidence interval (CI) 0.848-0.957) in the training group and an AUC of 0.891 (95% CI 0.799-0.983) in the validation group. Meanwhile, the nomogram showed good calibration. CONCLUSION: The nomogram model constructed based on T2WI radiomics and MRI reported had good diagnostic efficacies for LNM in patients with RC, and provided a new auxiliary method for accurate and individualized clinical management.

The Machine Learning Model for Distinguishing Pathological Subtypes of Non-Small Cell Lung Cancer.

Purpose: Machine learning models were developed and validated to identify lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) using clinical factors, laboratory metrics, and 2-deoxy-2[18F]fluoro-D-glucose ([18F]F-FDG) positron emission tomography (PET)/computed tomography (CT) radiomic features. Methods: One hundred and twenty non-small cell lung cancer (NSCLC) patients (62 LUAD and 58 LUSC) were analyzed retrospectively and randomized into a training group (n = 85) and validation group (n = 35). A total of 99 feature parameters-four clinical factors, four laboratory indicators, and 91 [18F]F-FDG PET/CT radiomic features-were used for data analysis and model construction. The Boruta algorithm was used to screen the features. The retained minimum optimal feature subset was input into ten machine learning to construct a classifier for distinguishing between LUAD and LUSC. Univariate and multivariate analyses were used to identify the independent risk factors of the NSCLC subtype and constructed the Clinical model. Finally, the area under the receiver operating characteristic curve (AUC) values, sensitivity, specificity, and accuracy (ACC) was used to validate the machine learning model with the best performance effect and Clinical model in the validation group, and the DeLong test was used to compare the model performance. Results: Boruta algorithm selected the optimal subset consisting of 13 features, including two clinical features, two laboratory indicators, and nine PEF/CT radiomic features. The Random Forest (RF) model and Support Vector Machine (SVM) model in the training group showed the best performance. Gender (P=0.018) and smoking status (P=0.011) construct the Clinical model. In the validation group, the SVM model (AUC: 0.876, ACC: 0.800) and RF model (AUC: 0.863, ACC: 0.800) performed well, while Clinical model (AUC:0.712, ACC: 0.686) performed moderately. There was no significant difference between the RF and Clinical models, but the SVM model was significantly better than the Clinical model. Conclusions: The proposed SVM and RF models successfully identified LUAD and LUSC. The results indicate that the proposed model is an accurate and noninvasive predictive tool that can assist clinical decision-making, especially for patients who cannot have biopsies or where a biopsy fails.

Imaging analysis of 13 rare cases of renal collecting (Bellini) duct carcinoma in northern China: a case series and literature review.

BACKGROUND: Collecting (Bellini) duct carcinoma (CDC) is a highly malignant and rare kidney tumor. We report our 12-year experience with CDC and the results of a retrospective analysis of patients and tumor characteristics, clinical manifestations, and imaging features by computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET)/CT. METHODS: Retrospective examination of tumors between January 2007 and December 2019 identified 13 cases of CDC from three medical centers in northern China. All 13 patients underwent CT scan, among which eight underwent dynamic enhanced CT scan, two underwent PET/CT scan, and one underwent magnetic resonance cholangiopancreatography (MRCP) examination. The lesions were divided into nephritis type and mass type according to the morphology of the tumors. RESULTS: The study group included ten men and three women with an average age of 64.23 ± 10.74 years. The clinical manifestations were gross hematuria, flank pain, and waist discomfort. The mean tumor size was 8.48 ± 2.48 cm. Of the 13 cases, six (46.2%) were cortical-medullary involved type and seven (53.8%) were cortex-medullary-pelvis involved type. Eleven (84.6%) cases were nephritis type and two (15.4%) were mass type. The lesions appeared solid or complex solid and cystic on CT and MRI. The parenchymal area of the tumors showed isodensity or slightly higher density on unenhanced CT scan in the 13 cases. PET/CT in two cases showed increased radioactivity intake. Evidence of intra-abdominal metastatic disease was present on CT in nine (69.2%) cases. CONCLUSIONS: The imaging characteristics of CDC differ from those of other renal cell carcinomas. In renal tumors located in the junction zone of the renal cortex and medulla that show unclear borders, slight enhancement, and metastases in the early stage, a diagnosis of CDC needs to be considered. PET/CT provides crucial information for the diagnosis of CDC, as well as for designing treatment strategies including surgery.