Pill Recognition Using Minimal Labeled Data
MetadataShow full item record
© 2017 IEEE. Inappropriate medication use such as wrong drug or wrong dose intake can be harmful to patients. In this work we present a method to automatically identify a pill from a single image using Convolutional Neural Network (CNN). We first localize the pill in the image by detecting the region with the highest concentration of edges. To overcome the challenge of minimal labeled training data and domain shift from the training images taken under the controlled lab environment to the consumer images taken under natural living conditions, several data augmentation techniques are applied on the Region of Interest to generate synthetic pill images for training the CNN. We adopted GoogLeNet Inception Network as our main classifier. Three GoogLeNet models with different specialties on color, shape and feature are trained on the augmented dataset. We evaluate our proposed method with a publicly available dataset provided by National Institute of Health that contains 1000 different pill classes.
Showing items related by title, author, creator and subject.
Truelove, M.; Patel, A.; Bompoint, S.; Brown, A.; Cass, A.; Hillis, G.; Peiris, D.; Rafter, N.; Reid, Christopher; Rodgers, A.; Tonkin, A.; Usherwood, T.; Webster, R.; Kanyini GAP Collaboration (2015)AIMS: Recent trials of cardiovascular polypills in high-risk populations show improvements in use of cardiovascular preventive treatments, compared to usual care. We describe patterns of pill burden in Australian practice, ...
Requests for emergency contraception in community pharmacy: an evaluation of services provided to mystery patientsHiggins, S.; Hattingh, Laetitia (2013)Background: Requests for supply of the emergency contraceptive pill (ECP) through community pharmacies require consideration of a range of factors and the application of professional judgment. Pharmacists should therefore ...
The effect of approach/avoidance training on alcohol consumption is mediated by change in alcohol action tendencySharbanee, Jason; Hu, L.; Stritzke, W.; Wiers, R.; Rinck, M.; MacLeod, C. (2014)Training people to respond to alcohol images by making avoidance joystick movements can affect subsequent alcohol consumption, and has shown initial efficacy as a treatment adjunct. However, the mechanisms that underlie ...