Hepatocellular carcinoma (HCC) remains a leading cause of cancer-related mortality worldwide, largely due to high recurrence rates after curative treatment and poor prognosis in advanced-stage disease. Multiphasic computed tomography (CT) and magnetic resonance imaging (MRI) play central roles in the noninvasive diagnosis and management of HCC; however, diagnostic variability and limited sensitivity for early-stage and atypical lesions continue to pose major clinical challenges. In recent years, artificial intelligence (AI) has emerged as a powerful tool for enhancing medical image analysis and improving clinical decision-making in liver oncology. This review summarizes recent advances in AI-based applications for HCC imaging using CT and MRI. We discuss the roles of AI in lesion detection and segmentation, differential diagnosis and tumor classification, imaging-based biological characterization, and prediction of treatment response and prognosis. In particular, radiomics and deep learning approaches have enabled automated extraction of high-dimensional imaging features, facilitating noninvasive assessment of tumor aggressiveness and heterogeneity. The integration of multimodal data, including imaging, clinical, and molecular information, further enhances predictive performance and personalized risk stratification. Despite encouraging results, most existing studies are limited by retrospective designs, heterogeneous methodologies, and insufficient external validation. Challenges related to model interpretability, data standardization, and clinical integration also remain unresolved. We highlight current limitations and future research directions, emphasizing the need for multi-center collaboration, prospective validation, and development of transparent and regulatory-compliant systems. With continued technological innovation and interdisciplinary collaboration, AI-driven imaging analysis is expected to play an increasingly important role in precision medicine and outcome optimization for patients with HCC.
Background: Major depressive disorder (MDD) is frequently accompanied by cognitive deficits and appetite hormone dysregulation. The retinoid X receptor-α (RXR-α)–peroxisome proliferator–activated receptor-γ (PPAR-γ) axis and its coactivator PPAR-γ coactivator 1α (PGC-1α) regulate metabolic, inflammatory, and neuroprotective processes, yet their role in first-episode adolescent MDD remains largely unexplored. Methods: Adolescents between the ages of 12 and 17 with first-episode MDD and age-matched healthy controls were enrolled according to DSM-5 diagnostic criteria. Serum levels of RXR-α, PPAR-γ, PGC-1α, and appetite hormones (insulin, leptin, adiponectin) were measured. Executive function was assessed using the Wisconsin Card Sorting Test (WCST). Group differences and biomarker associations with cognitive and metabolic measures were analyzed using generalized linear models adjusted for age, sex, BMI, depression severity, and medication use. The association analyses were exploratory and not corrected for multiple comparisons. Results: A total of 62 adolescents with first-episode MDD (mean age 15.92 years, 74.2% female and 25.8% male) and 64 healthy controls were recruited. Compared with controls, adolescents with MDD exhibited significantly lower PPAR-γ ( p = 0.033) and PGC-1α levels ( p = 0.024), although these differences did not remain significant after correction for multiple comparisons. Higher fasting insulin ( p = 0.002) and leptin ( p = 0.017) were also observed. After adjusting for covariates, PGC-1α—but not PPAR-γ or RXR-α—showed exploratory associations with WCST performance (fewer non-perseverative errors, p = 0.035) and was negatively correlated with insulin ( p = 0.019) and leptin ( p = 0.027) levels. Conclusion: In adolescents with first-episode MDD, lower PGC-1α levels were observed and were associated with both poorer WCST performance and disturbances in appetite hormone regulation. These preliminary findings suggest that PGC-1α may represent a candidate biomarker linking metabolic changes and cognitive deficits in early-stage MDD, warranting further investigation.
Background: Idiopathic pulmonary fibrosis (IPF) carries a poor prognosis despite antifibrotic therapy. The Gender–Age–Physiology (GAP) index enables baseline risk assessment but does not capture longitudinal disease behavior. To determine whether combining baseline GAP score with early change in forced vital capacity (Δ1YFVC%) improves prediction of 2-year mortality in IPF. Methods: A prospective, multicenter observational cohort study conducted across three hospitals in Taiwan between 2018 and 2023. Patients with IPF receiving reimbursed antifibrotic therapy were enrolled. Individuals who died within six months, were lost to follow-up, or lacked one-year pulmonary function data were excluded. The primary outcome was 2-year mortality. Baseline GAP score and Δ1YFVC% were evaluated using a survival-tree model to determine prognostic cutoffs. Survival was analyzed using Kaplan–Meier curves and Cox proportional-hazards regression. Results: Among 156 patients included (median age 77 years; 76.3% male; median baseline FVC 73.5% predicted). 23 (14.7%) died within 2 years. The survival-tree model identified a key prognostic split by GAP (<5 vs ≥5) and Δ1YFVC%. A decline in predicted FVC% greater than 10.5 percentage points (Δ1YFVC% declined > 10.5 percentage points) was associated with increased mortality, even among patients with lower GAP scores (adjusted HR 3.30, p =0.010). Among patients with GAP ≥5, relatively stable lung function (Δ1YFVC% improved > 1.5 percentage points) was associated with lower mortality risk. Stratification based on GAP score and Δ1YFVC% significantly discriminated mortality risk groups (log-rank p <0.0001, concordance index 0.761) Conclusion: Integrating Δ1YFVC% with the GAP score provides a clinically applicable approach for risk stratification and may support the identification of high-risk patients during antifibrotic therapy.
