Clinical Trials (CTs) are vital for validating the safety and effectiveness of new medical intervention before they are approved for the public use. However, finding eligible participants for a trial is a complex and time-consuming task, as eligibility criteria are often expressed in lengthy and unstructured documents which require expert interpretation. This manual processing leads to delays, inconsistencies and increased costs in clinical research. To address this challenge, proposed SmartCTR an automated framework that extracts and evaluates the trial eligibility criteria and perform patient trial matching. System employs the Optical character recognition (OCR) to extract raw text, followed by Large Language Model (LLM) specifically Llama-3(8B-instruct) to transform free text criteria into structured JSON representation. Synthetic python rule-based module verifies the clinical details such as age, gender, vaccination and pregnancy status, while the BioBERT model compares the condition criteria with patient information to measure the accuracy of disease. The framework is evaluated using Retrieval-Augmented Generation (RAG) method that compares LLM generated outputs with human expert interpretations, ensuring the accuracy and consistency. Results show that SmartCTR provides a balanced and explainable approach to eligibility matching, combining the interpretability of rule-based logic with the flexibility of LLM driven reasoning. Overall, SmartCTR demonstrated a scalable and interpretable method for automating clinical trial eligibility extraction, with the potential to accelerate recruitment and enhance the readability of clinical research workflows.