Abstract: In agriculture, the early detection of diseases is important for an efficient crop yield. Healthy or unhealthy, the bacterial spot, late blight, sesame leaf spot, seedling blight, and yellow curved leaf diseases affect the quality of the crop. To avoid errors due to manual detection of diseases, deep learning methods are used for image processing. Using a deep convolution neural network that has been trained and fine tuned to adequately match a database of a plant leaves that was compiled independently for various plant illnesses, the technique paper presented here may represent a novel method for identifying plant diseases. So, we propose using deep learning techniques that are involved in the detection of diseases in their very early stages, using defined techniques to classify image databases. Finally, a classification technique is used to specify the type of plant leaf disease. The predicted technique successfully diagnoses the illness and accurately categorizes the plants leaf illness when compared to CNN models.

Abstract: The rapid growth in the medical image analysis domain is encouraging the researchers to look forward for the improvement in disease detection. Numerous research attempts are being made to enhance the detection of diseases based on the manual data accumulation from the magnetic resonance (MR) images and processing the data further to generate subsequent analysable reports for prediction and detection of syndromes related to human brain. Nevertheless, the approaches are prone to errors as the first level analysis is made manually and then the gathered data is analysed again, finally resulting into a questionable outcome. Thus, the demand from the modern research is to make the analysis and detection of brain tumour automatic, which is accurate and timely. Henceforth, this work proposes a novel framework to analyse and detect brain tumours from MR Images and subsequently deploy a predictive sub framework to detect the probable diseases may cause to the subject (Patient).