A Deep Learning Approach for Medicine Preparation Machines With SVM and CNN Integration

Abstract

This chapter provides an innovative way to medical medication teaching by the integration of a robotic dispensing arm with better machine learning models, especially support vector machines (SVM) and convolutional neural networks (CNN). The machine employs QR code-scanning to become aware of drug packing containers, with each medicine holding a unique QR code for improved traceability. Feature extraction from the QR codes, along with the teaching of SVM and CNN models, allows the device to accurately classify medications. The SVM version shows a decent accuracy of 92.22%, but the CNN version exceeds with an amazing accuracy of 99.21%. These models demonstrate good accuracy, recall, and F1 score values, offering a full assessment of their effectiveness in drug identification. The confusion matrices give a detailed understanding of real high quality, fake bad, false fantastic, and true dreadful periods, thus proving the styles' prediction skills. The suggested machine includes a twin-conveyor mechanism, speeding the sorting method based on SVM and CNN model predictions. This study improves the area of pharmaceutical automation, and it provides the framework for shrewd and adaptable healthcare structures. The integration of robotics and device mastery in medication contains potential implications for higher performance, accuracy, and protection in pharmaceutical strategies, with capability packages extending beyond medicinal drug practices into various healthcare domains.