Hepatocellular carcinoma (HCC) remains not only a cause of a considerable part of oncologic mortality, but also a diagnostic and therapeutic challenge for healthcare systems worldwide. Early detection of the disease and consequential adequate therapy are imperative to increase patients’ quality of life and survival. Imaging plays, therefore, a crucial role in the surveillance of patients at risk, the detection and diagnosis of HCC nodules, as well as in the follow-up post-treatment. The unique imaging characteristics of HCC lesions, deriving mainly from the assessment of their vascularity on contrast-enhanced computed tomography (CT), magnetic resonance (MR) or contrast-enhanced ultrasound (CEUS), allow for a more accurate, noninvasive diagnosis and staging. The role of imaging in the management of HCC has further expanded beyond the plain confirmation of a suspected diagnosis due to the introduction of ultrasound and hepatobiliary MRI contrast agents, which allow for the detection of hepatocarcinogenesis even at an early stage. Moreover, the recent technological advancements in artificial intelligence (AI) in radiology contribute an important tool for the diagnostic prediction, prognosis and evaluation of treatment response in the clinical course of the disease. This review presents current imaging modalities and their central role in the management of patients at risk and with HCC.

Background: Response to neoadjuvant therapy in locally advanced rectal cancer (LARC) is heterogeneous, and early identification of non-responders may help optimize treatment strategies and reduce unnecessary toxicity. This study aimed to develop and internally validate a machine learning model based on radiomic features extracted from baseline magnetic resonance imaging (MRI) to predict treatment response defined according to MRI tumor regression grade (mrTRG) at restaging MRI. Methods: In this retrospective single-center study, 86 patients with histologically confirmed LARC who underwent baseline and restaging MRI, neoadjuvant therapy, and surgery were included. Primary tumors were manually segmented on oblique axial T2-weighted images. A total of 107 radiomic features were extracted using PyRadiomics (vrs 3.0.1), with and without N4 bias field correction. Feature selection was performed using LASSO, followed by elastic net–regularized logistic regression. Model performance was evaluated using repeated stratified 5-fold cross-validation (20 repetitions). Treatment response was defined according to MRI tumor regression grade (mrTRG) at restaging, dichotomized into responders (mrTRG ≤ 2) and non-responders (mrTRG ≥ 3). Results: The model achieved a mean area under the receiver operating characteristic curve (AUC-ROC) of 0.73, with an accuracy of 72.5%, sensitivity of 79.2%, and specificity of 50%. Conclusions: Baseline MRI-based radiomics shows potential for identifying patients at higher risk of non-response to neoadjuvant therapy in LARC. However, limited specificity and the absence of external validation restrict immediate clinical applicability. Further validation in larger multicenter cohorts and integration with clinical variables are warranted to improve model robustness and generalizability.

Background: Response to neoadjuvant therapy in locally advanced rectal cancer (LARC) is heterogeneous and early identification of non-responders may help optimize treatment strategies and reduce unnecessary toxicity. This study aimed to develop and internally validate a machine learning model based on radiomic features extracted from baseline magnetic resonance imaging (MRI) to predict treatment response assessed at restaging MRI. Methods: In this retrospective single-center study 86 patients with histologically confirmed LARC who underwent baseline and restaging MRI, neoadjuvant therapy, and surgery, were included. Primary tumors were manually segmented on oblique axial T2-weighted images. A total of 107 radiomic features were extracted using PyRadiomics, with and without N4 bias field correction. Feature selection was performed using LASSO, followed by elasticnet–regularized logistic regression. Model performance was assessed using repeated stratified 5-fold cross-validation. Response was defined according to MRI tumor regression grade (mrTRG) at restaging, dichotomized into responders (mrTRG ≤ 2) and non-responders (mrTRG ≥ 3). Results: The model achieved a mean area under the receiver operating characteristic curve (AUC-ROC) of 0.73, accuracy of 72.5%, sensitivity of 79.2%, and specificity of 50%. Conclusions: Baseline MRI-based radiomics demonstrated to potentially identify patients at higher risk of non-response to neoadjuvant therapy in LARC.