Machine Learning (ML) is tremendously enhancing the healthcare sector by continuously collaborating in diverse healthcare circumstances. It explores thousands of data beyond human capability, analyzes medical conditions, and suggests outcomes with clinical insights. Still, a large segment of the population does not get quality healthcare due to insufficient medical facilities and socio-economic conditions. Thus, the Portable Health Clinic (PHC) aims to make technology-enabled smart healthcare affordable to the unreached population. The paper reports a comparative analysis of seven supervised learning algorithms to predict the possible health risk in future using the remote healthcare data provided by PHC. The survey and clinical data of PHC have been used in this work to predict the triage condition that a patient may have later. Four categorical variable encoding and two missing value handling techniques have been applied for preprocessing and preparing the healthcare data. The preprocessed PHC data has achieved 90.34% accuracy on the Random Forest Classifier. Thus, this pre-informing health service will reduce health management costs and allow people to take necessary mitigating actions to minimize health risks.