Oxygen Concentration Regulation is considered to be an important task in the medical domain. This paper focuses on an automated method for the Oxygen Concentration Regulation. To execute this task, we have proposed and implemented a technique for the regulated delivery of Liquid Medical Oxygen based on instantaneous body parameters using a proposed Decision Tree Regressor and data proliferation using the CTGAN model with an SDV data synthesizer. We have collected 30 sample records from the centenary hospital, further augmented to 2,00,000 data samples using the proposed CTGAN architecture with SDV data synthesizer. We have generated the oxygen saturation versus oxygen flow graph before and after data proliferation or augmentation. Then the generated dataset is applied through the proposed Decision Tree Regressor architecture after data cleaning, feature engineering, and dataset splitting. The proposed model is evaluated to calculate the actual and predicated error analysis. The proposed model generates an accuracy of 85% and suppresses other existing regressor models such as Linear Regression, Logistic Regression, and Polynomial Regression. The major contributions of the paper are; the proposed CTGAN architecture with SDV data synthesizer model, dealing with real-world dataset, and Nobel methodology in the field of regulated delivery of Liquid Medical Oxygen (LMO) based on instantaneous body parameters. The proposed model may come up as a product applicable to the medical practitioner, patient, and society.