Search

Purpose: To establish an explainable 18F-FDG PET/CT-derived prediction model to identify EGFR mutation status and subtypes (EGFR wild, EGFR-E19, and EGFR-E21) in lung adenocarcinoma (LUAD). Methods: Baseline 18F-FDG PET/CT images of 478 patients with LUAD from 2 hospitals were collected. Data from hospital A (n = 390) was randomly split into a training group (n = 312) and an internal test group (n = 78), with data from hospital B (n = 88) utilized for external test. Further, a total of 4,760 handcrafted radiomics features (HRFs) were extracted from PET/CT scans. Candidates for the prediction model were constructed by cross-combinations of 11 feature selection methods and 7 classifiers. The optimal model was determined by combining the results of cross-center data validation and model visualization (Yellowbrick). The predictive performance was assessed via receiver operating characteristic curve, confusion matrix and classification report. Four explainable artificial intelligence technologies were used for optimal model interpretation. Results: Sex and SUVmax were selected as clinical risk factors, which were then combined with 8 robust PET/CT HRFs to establish the models. The optimal performance was obtained by combining a light gradient boosting machine classifier with random forest feature selection method achieving an optimal performance with a macro-average AUC of 0.75 in the internal test group and 0.81 in the external test group. Conclusion: The explainable EGFR mutation status prediction model have certain clinical practicability and good generalization performance, which may help in the timely selection of treatment options and prognosis prediction in patients with LUAD. [ABSTRACT FROM AUTHOR]

Background: Vessels encapsulating tumor clusters (VETC) pattern is a novel microvascular pattern associated with poor outcomes of hepatocellular carcinoma (HCC). Preoperative estimation of VETC has potential to improve treatment decisions. Purpose: To develop and validate a nomogram based on gadoxetate disodium‐enhanced MRI for estimating VETC in HCC and to evaluate whether the estimations are associated with recurrence after hepatic resection. Study Type: Retrospective. Population: A total of 320 patients with HCC and histopathologic VETC pattern assessment from three centers (development cohort:validation cohort = 173:147). Field Strength/Sequence: A3.0 T/turbo spin‐echo T2‐weighted, spin‐echo echo‐planar diffusion‐weighted, and 3D T1‐weighted gradient‐echo sequences. Assessment: A set of previously reported VETC‐ and/or prognosis‐correlated qualitative and quantitative imaging features were assessed. Clinical and imaging variables were compared based on histopathologic VETC status to investigate factors indicating VETC pattern. A regression‐based nomogram was then constructed using the significant factors for VETC pattern. The nomogram‐estimated VETC stratification was assessed for its association with recurrence. Statistical Tests: Fisher exact test, t‐test or Mann–Whitney test, logistic regression analyses, Harrell's concordance index (C‐index), nomogram, Kaplan–Meier curves and log‐rank tests. P value < 0.05 was considered statistically significant. Results: Pathological VETC pattern presence was identified in 156 patients (development cohort:validation cohort = 83:73). Tumor size, presence of heterogeneous enhancement with septations or with irregular ring‐like structures, and necrosis were significant factors for estimating VETC pattern. The nomogram incorporating these indicators showed good discrimination with a C‐index of 0.870 (development cohort) and 0.862 (validation cohort). Significant differences in recurrence rates between the nomogram‐estimated high‐risk VETC group and low‐risk VETC group were found (2‐year recurrence rates, 50.7% vs. 30.3% and 49.6% vs. 31.8% in the development and validation cohorts, respectively). Data Conclusion: The nomogram integrating gadoxetate disodium‐enhanced MRI features was associated with VETC pattern preoperatively and with postoperative recurrence in patients with HCC. Evidence Level: 4 Technical Efficacy: Stage 2 [ABSTRACT FROM AUTHOR]