RETRACTED ARTICLE: An approach for disease prediction and classification using novel weighting method and multichannel shared functional behaviour

At a worldwide scale, artificial intelligence (AI) is now an intrinsic part of various sectors and a significant strategic element in the business agendas of several industries including health care, finance and retail. Machine learning, an statistical paradigm of AI, is one of the most preferred technologies to achieve AI. Machine learning-oriented approaches exploit massive sized, unstructured and complicated dataset instances to learn from previous experiences and find insightful patterns. A range of statistical, probabilistic and optimization approaches are used to achieve this task. Early detection of chronic diseases is critical in the realm of biomedical research and healthcare communities, where it is pivotal to primarily diagnose the particular disease at probabilistically early stage in order to lower the death rate. Pneumonia mainly affects a huge number of people, particularly children and adults, in the developing and undeveloped nations that are characterized as overcrowding, inadequate sanitation, malnutrition, lack of suitable medical services and other risk factors. It is critical to detect pneumonia at its early stage in order to properly treat the infection. This paper presents an approach, i.e. DPUD (Disease Prediction for Unstructured Data). The proposed framework consists of an statistically novel and fine-tuned algorithmic procedure to address certain pain points of categorization problem such as selection of an optimal set of model hyperparameters and attain an improved statistical performance metrics with multichannel shared activation function and cost function. In experimental evaluation, our approach capitulates state-of-the-art computational achievement on Chest X-Ray Images for the Pneumonia Classification dataset , while being considerably much faster at specifically test time. The methodology presented in this chapter has an empirical scope for societal improvement, modernization and progress.