Search

Objectives: To predict KRAS mutation in rectal cancer (RC) through computer vision of [18F]fluorodeoxyglucose (18F‐FDG) positron emission tomography (PET)/computed tomography (CT) by using metric learning (ML). Methods: This study included 160 patients with RC who had undergone preoperative PET/CT. KRAS mutation was identified through polymerase chain reaction analysis. This model combined ML with the deep‐learning framework to analyze PET data with or without CT images. The Batch Balance Wrapper framework and K‐fold cross‐validation were employed during the learning process. A receiver operating characteristic (ROC) curve analysis was performed to assess the model's predictive performance. Results: Genetic alterations in KRAS were identified in 82 (51%) tumors. Both PET and CT images were used, and the proposed model had an area under the ROC curve of 0.836 for its ability to predict a mutation status. The sensitivity, specificity, and accuracy were 75.3%, 79.3%, and 77.5%, respectively. When PET images alone were used, the area under the curve was 0.817, whereas the sensitivity, specificity, and accuracy were 73.2%, 79.6%, and 76.2%, respectively. Conclusions: The ML model presented herein revealed that baseline 18F‐FDG PET/CT images could provide supplemental information to determine KRAS mutation in RC. Additional studies are required to maximize the predictive accuracy. Advances in knowledge: The results of the ML model presented herein indicate that baseline 18F‐FDG PET/CT images could provide supplemental information for determining KRAS mutation in RC. The predictive accuracy of the model was 77.5% when both image types were used and 76.2% when PET images alone were used. Additional studies are required to maximize the predictive accuracy. [ABSTRACT FROM AUTHOR]

Background: We designed a deep learning model for assessing 18F-FDG PET/CT for early prediction of local and distant failures for patients with locally advanced cervical cancer.Methods: All 142 patients with cervical cancer underwent 18F-FDG PET/CT for pretreatment staging and received allocated treatment. To augment the amount of image data, each tumor was represented as 11 slice sets each of which contains 3 2D orthogonal slices to acquire a total of 1562 slice sets. In each round of k-fold cross-validation, a well-trained proposed model and a slice-based optimal threshold were derived from a training set and used to classify each slice set in the test set into the categories of with or without local or distant failure. The classification results of each tumor were aggregated to summarize a tumor-based prediction result.Results: In total, 21 and 26 patients experienced local and distant failures, respectively. Regarding local recurrence, the tumor-based prediction result summarized from all test sets demonstrated that the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 71%, 93%, 63%, 95%, and 89%, respectively. The corresponding values for distant metastasis were 77%, 90%, 63%, 95%, and 87%, respectively.Conclusion: This is the first study to use deep learning model for assessing 18F-FDG PET/CT images which is capable of predicting treatment outcomes in cervical cancer patients.Key Points: • This is the first study to use deep learning model for assessing 18 F-FDG PET/CT images which is capable of predicting treatment outcomes in cervical cancer patients. • All 142 patients with cervical cancer underwent 18 F-FDG PET/CT for pretreatment staging and received allocated treatment. To augment the amount of image data, each tumor was represented as 11 slice sets each of which contains 3 2D orthogonal slices to acquire a total of 1562 slice sets. • For local recurrence, all test sets demonstrated that the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 71%, 93%, 63%, 95%, and 89%, respectively. The corresponding values for distant metastasis were 77%, 90%, 63%, 95%, and 87%, respectively. [ABSTRACT FROM AUTHOR]

Background: In this study, we investigated the correlation between the lymph node (LN) status or histological types and textural features of cervical cancers on F-fluorodeoxyglucose positron emission tomography/computed tomography. Methods: We retrospectively reviewed the imaging records of 170 patients with International Federation of Gynecology and Obstetrics stage IB-IVA cervical cancer. Four groups of textural features were studied in addition to the maximum standardized uptake value (SUV), metabolic tumor volume, and total lesion glycolysis (TLG). Moreover, we studied the associations between the indices and clinical parameters, including the LN status, clinical stage, and histology. Receiver operating characteristic curves were constructed to evaluate the optimal predictive performance among the various textural indices. Quantitative differences were determined using the Mann-Whitney U test. Multivariate logistic regression analysis was performed to determine the independent factors, among all the variables, for predicting LN metastasis. Results: Among all the significant indices related to pelvic LN metastasis, homogeneity derived from the gray-level co-occurrence matrix (GLCM) was the sole independent predictor. By combining SUV, the risk of pelvic LN metastasis can be scored accordingly. The TLG was the independent feature of positive para-aortic LNs. Quantitative differences between squamous and nonsquamous histology can be determined using short-zone emphasis (SZE) from the gray-level size zone matrix (GLSZM). Conclusion: This study revealed that in patients with cervical cancer, pelvic or para-aortic LN metastases can be predicted by using textural feature of homogeneity from the GLCM and TLG respectively. SZE from the GLSZM is the sole feature associated with quantitative differences between squamous and nonsquamous histology. [ABSTRACT FROM AUTHOR]

Purpose: This study investigated the correlation of the matrix heterogeneity of tumors on F-fluorodeoxyglucose positron emission tomography-computed tomography (PET-CT) with gene-expression profiling in patients with pharyngeal cancer and determined the prognostic factors for radiotherapy-based treatment outcomes. Methods: We retrospectively reviewed the records of 57 patients with stage III-IV oropharyngeal or hypopharyngeal cancer who had completed definitive therapy. Four groups of the textural features as well as 31 indices were studied in addition to maximum standard uptake value, metastatic tumor volume, and total lesion glycolysis. Immunohistochemical data from pretreatment biopsy specimens ( Glut1, CAIX, VEGF, HIF-1α, EGFR, Ki-67, Bcl-2, CLAUDIN-4, YAP-1, c-Met, and p16) were analyzed. The relationships between the indices and genomic expression were studied, and the robustness of various textural features relative to cause-specific survival and primary relapse-free survival was analyzed. Results: The overexpression of VEGF was positively associated with the increased values of the matrix heterogeneity obtained using gray-level nonuniformity for zone (GLNUz) and run-length nonuniformity (RLNU). Advanced T stage ( p = 0.01, hazard ratio [HR] = 3.38), a VEGF immunoreactive score of >2 ( p = 0.03, HR = 2.79), and a higher GLNUz value ( p = 0.04, HR = 2.51) were prognostic factors for low cause-specific survival, whereas advanced T stage, a HIF-1α staining percentage of ≥80%, and a higher GLNUz value were prognostic factors for low primary-relapse free survival. Conclusions: The overexpression of VEGF was associated with the increased matrix index of GLNUz and RLNU. For patients with pharyngeal cancer requiring radiotherapy, the treatment outcome can be stratified according to the textural features, T stage, and biomarkers. [ABSTRACT FROM AUTHOR]