The backbone of global food security is crops rich in carbohydrates, such as wheat, maize and rice, combined with protein sources like quinoa and lentils and oil crops such as canola and sunflower. However, these crops are increasingly vulnerable to genetic diseases that affect both yield and nutrient quality. To avoid losses in crops and there will be the sustainable production of wheat, maize, rice, quinoa, lentils, canola and sunflower in a manner that will have the disease resistance along with the high macronutrient yield, thereby ensuring food requirements in the future and the sustainability of agriculture. To propose a machine learning-driven Genetic Disease Prediction and Optimization Model (GDPM), which support selection, prediction and control genetic diseases on essential crops. GDPM concentrate on varieties that are resistant to rust or smut and also have a high starch or fiber content. To predict early detection of disease in the crop and help the breeders develop varieties with nutrient integrity even under stressed conditions like drought or pest attacks. Plant breeders, agricultural scientists and AI experts will have to collaborate for developing accurate models and ensuring practical adoption. The GDPM also enables farmers to know the data that is available currently, thereby allowing them to choose varieties that are resistant to such diseases.