Early diagnosis of chronic medical conditions can lead to enhanced medical care and decreased mortality. Heart disease, diabetes, Parkinson’s, kidney and liver disease impact millions of people around the globe. In the past, diagnosing decreasing deaths from chronic disease was performed through manual examinations of patient records and lab reports which could take a lot of time leading to patients sometimes being misdiagnosed or not receiving the right treatment for their illnesses in a timely manner. This paper presents a multi-disease prediction system developed through machine learning capable of predicting several diseases' chances at once by analyzing patient information such as age, sex/gender, Body Mass Index (BMI), blood pressure, cholesterol level, glucose level, physical habits (lifestyle factors) and genetic markers. The proposed model was developed using XGBoost and MultiOutputClassifier algorithms to improve the accuracy of disease prediction. Publicly available datasets were retrieved from Kaggle for training and assessing the performance of the model's performance measures (i.e., accuracy, precision, recall, F1 score) will be used thus validating the predictive nature of the system.