The goal of this work is to anticipate obstructions related to Totally Implantable Venous Access Ports (TIVAPs) by applying different boosting algorithms, such as AdaBoost, Gradient Boosting Machines (GBM), XGBoost, and LightGBM. The work shows how each boosting approach performs in terms of accuracy, precision, recall, and F1-Score using a dataset of 1084 individuals that are characterized by factors including age, gender, tumor type, WHO performance status, and medical history. It is clear from the data that LightGBM is the most successful model overall, with the highest accuracy (88%) and the best balance between precision and recall. While recall is a little less accurate than precision, XGBoost performs quite well, suggesting that it is good at detecting positive examples. It appears that AdaBoost and GBM are more conservative in their obstruction prediction since they exhibit high precision but low recall. The findings highlight the potential of these machine learning techniques in enhancing the prediction and management of TIVAP-related complications, ultimately contributing to better patient care and outcomes.