Hepatocellular carcinoma (HCC) is a leading cause of cancer-related mortality, with Transarterial Chemoembolization (TACE) commonly used as a treatment for intermediate-stage patients. Accurate treatment response assessment is critical for guiding clinical decisions, where RECIST 1.1 provides standardized criteria based on changes in tumor diameters. However, manual diameter measurement is time-intensive and prone to inter-observer variability. We propose an automated pipeline for tumor segmentation, diameter extraction, and RECIST 1.1 classification using pre- and post-TACE CT scans from the HCC-TACE-SEG dataset. Nine deep learning-based segmentation models were trained from scratch to delineate tumors. Longest diameters were extracted from the axial plane using a 26-way connectivity algorithm to ensure coherent tumor representation. Based on these measurements, automated RECIST classifications, Complete Response (CR), Partial Response (PR), Stable Disease (SD), and Progressive Disease (PD), were derived. Experimental results demonstrated robust segmentation performance across models. Bland-Altman analysis confirmed high agreement between automated and expert diameter measurements for both pre- and post-treatment scans, while automated RECIST classifications achieved substantial agreement with expert labels, as quantified by Cohen’s Kappa coefficient. This study demonstrates the feasibility of an objective, fully automated system for tumor response evaluation in HCC. By standardizing RECIST 1.1 assessment, the proposed framework reduces subjectivity, minimizes inter-observer variability, and supports more consistent clinical decision-making in TACE-based treatment.