Palmer, G, Schnieders, B, Savani, R 
ORCID: 0000-0003-1262-7831, Tuyls, K, Fossel, J and Flore, H
(2020)
The Automated Inspection of Opaque Liquid Vaccines.
In: EUROPEAN CONFERENCE ON ARTIFICIAL INTELLIGENCE.
|
Text
2002.09406v1.pdf - Submitted version Download (3MB) | Preview |
Abstract
In the pharmaceutical industry the screening of opaque vaccines containing suspensions is currently a manual task carried out by trained human visual inspectors. We show that deep learning can be used to effectively automate this process. A moving contrast is required to distinguish anomalies from other particles, reflections and dust resting on a vial's surface. We train 3D-ConvNets to predict the likelihood of 20-frame video samples containing anomalies. Our unaugmented dataset consists of hand-labelled samples, recorded using vials provided by the HAL Allergy Group, a pharmaceutical company. We trained ten randomly initialized 3D-ConvNets to provide a benchmark, observing mean AUROC scores of 0.94 and 0.93 for positive samples (containing anomalies) and negative (anomaly-free) samples, respectively. Using Frame-Completion Generative Adversarial Networks we: (i) introduce an algorithm for computing saliency maps, which we use to verify that the 3D-ConvNets are indeed identifying anomalies; (ii) propose a novel self-training approach using the saliency maps to determine if multiple networks agree on the location of anomalies. Our self-training approach allows us to augment our data set by labelling 217,888 additional samples. 3D-ConvNets trained with our augmented dataset improve on the results we get when we train only on the unaugmented dataset.
| Item Type: | Conference or Workshop Item (Unspecified) |
|---|---|
| Additional Information: | 8 pages, 5 Figures, 3 Tables, ECAI 2020 Conference Proceedings |
| Uncontrolled Keywords: | cs.CV, cs.CV |
| Divisions: | Faculty of Science and Engineering > School of Electrical Engineering, Electronics and Computer Science |
| Depositing User: | Symplectic Admin |
| Date Deposited: | 16 Aug 2021 14:58 |
| Last Modified: | 18 Jan 2023 21:33 |
| DOI: | 10.3233/FAIA200307 |
| Open Access URL: | https://ebooks.iospress.nl/publication/55102 |
| Related URLs: | |
| URI: | https://livrepository.liverpool.ac.uk/id/eprint/3133709 |
Dimensions
Dimensions
