Voice Over Phonetic Medical Prescriptions for Diagnosing Diseases Using Random Forest Classifier

The proposed software is a novel solution for diagnosing diseases using audio input. Based on machine learning algorithms, it extracts symptoms from voice characteristics such as pitch, tone, and breathing sounds, and correlates them to specific medical conditions. The technology has the potential to transform healthcare by providing faster, more accurate diagnoses remotely and at an early stage, improving patient outcomes and reducing healthcare costs. The software’s ability to analyze audio input remotely can significantly enhance medicine and remote consultations, making healthcare more accessible and efficient for patients and healthcare professionals. Furthermore, it eliminates the need for additional diagnostic tests, saving costs and resources. This technology has potential applications in various medical specialties, including respiratory disorders, cardiovascular diseases, and mental health. It could revolutionize the way patients receive medical care and improve health outcomes, especially in underserved communities. The proposed algorithm utilizes Random Forest Classifier, with a test accuracy of 0.93, to predict the patient’s disease based on their symptoms, which in turn is recorded by speech recognizer, with an accuracy of 0.98. Combining the models results in a highly accurate software which is able to successfully predict majority of the diseases within the Columbia University Dataset.