Health of bones is essential for overall physical well-being. Especially during conditions like osteoporosis or osteoarthritis, the health of bone must be maintained as these are the diseases which can be controlled by avoiding the progression. The avoidance of progression can be achieved by early detection of such diseases. The research focused on using ensemble learning techniques to develop an enhanced method for assessing bone health and also integrating lifestyle management techniques for prevention and care. The proposed work used an ensemble of Xception model and DenseNet201 model for detecting the presence of osteoarthritis or osteoporosis at various stages. The dataset consists of 1889 images. The ensemble model has achieved 90.64% testing accuracy. It used an explainable artificial intelligence technique called LIME to interpret the results produced by the model. The explanations were also evaluated using various metrics. The users are provided with calcium content of the food they consume using a method called case-based reasoning (4R model) so that they can plan their diet accordingly. Finally all these functionalities were integrated into an application which would assist healthcare professionals and patients for effectively maintaining the health of the bone.