Self-medication and self-prescription are common practices that can be observed in many countries around the world, from the most advanced in terms of medical services in Europe to the less ones as in South America or Africa. Self-medication is defined as the consumption of one or more drugs without the advice of a physician. Many studies on several countries reveal the type of medi- cations that are consumed as well the social groups that normally use this practice. The consequences for this can range from a mild allergic reaction to death. On the other hand, it is easy to buy drugs without a prescription in pharmacies or super- markets, but consumers do not always know which one to choose, neither the ingredients nor side effects they can cause. Here we present a Web App which uses a classifier model for counter medication based on computer vision and machine learning techniques such as Bag-of-visual words, K-Means, and support vector machines. We collected 150 images from 11 different counter medications. The classifier was tested with 43 new images, and obtained 90.7% of accuracy, 93% of precision, 91% of recall, and 91% of F1-score.
CITATION STYLE
Ceh-Varela, E., Hernández-Chan, G., Villanueva-Escalante, M., & Sánchez-Cervantes, J. L. (2018). MED-IS-IN, an Intelligent Web App for Recognizing Non-prescription Drugs (pp. 273–292). https://doi.org/10.1007/978-3-319-56871-3_14
Mendeley helps you to discover research relevant for your work.