The International Classification of Diseases (ICD) is an International standard for epidemiological investigation, health management and clinical diagnosis that play a critical role in intelligent healthcare. The task of assigning medical codes to clinical free-text documents have been recognized as advantageous for purposes like insurance claim processes, E-health record management and research. Healthcare professionals and professional medical coders assign such codes manually to the health records like discharge summaries, radiology reports, death certificates, pathology reports, etc. Using an automated approach for this task helps in reducing expenses, alleviating administrative burden and prevents errors. The main objective of this literature review is to provide detailed overview about automated ICD coding, deep learning algorithms and state-of-the art models used for ICD coding. Different NLP techniques were utilized for effective text processing and assignment of ICD codes to Electronic Health Records (EHR). The performance of several approaches on publicly available EHR datasets like MIMIC was compared using appropriate performance metrics.