In today's fast-paced society, where demanding schedules often lead to the neglect of personal health, a growing number of individuals find it challenging to prioritize well-being. The adoption of regular exercise and yoga, crucial for maintaining a healthy body, remains a practice embraced by only a small fraction of the population. Although health-oriented mobile applications such as Home Workout without any equipment and Google Fit have gained recognition, their consistent utilization remains limited. Existing applications primarily focus on tracking parameters such as heart rate, daily steps, and calories burned, yet they fall short in predicting potential diseases based on these indicators. The proposed model introduces a novel approach to disease prediction by autonomously calculating specific health parameters and determining whether an individual is healthy or may have an underlying disease. In our research, we conducted a comparative analysis of various machine learning algorithms to enhance predictive accuracy. The Random Forest algorithm demonstrated remarkable effectiveness in disease prediction, achieving an impressive accuracy rate of 95%. This suggests that the model, built on an ensemble of decision trees, performed exceptionally well on the evaluated dataset. Nonetheless, it's crucial to consider factors such as dataset size, representativeness, feature selection, and potential biases when interpreting these results. Beyond disease prediction, our innovative application extends its functionality to provide personalized diet recommendations tailored to individual health conditions. This proposed model functions as a virtual health guide, offering early diagnosis comparable to a doctor's role. By alleviating the workload on healthcare professionals, ensuring timely treatment, and potentially saving lives, our application aims to make a substantial contribution to public health.