Image-Based Artificial Intelligence Methods for Product Control of Tablet Coating Quality
Mimicking the human decision-making process is challenging. Especially, many process control situations during the manufacturing of pharmaceuticals are based on visual observations and related experience-based actions. The aim of the present work was to investigate the use of image analysis to class...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-09-01
|
| Series: | Pharmaceutics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4923/12/9/877 |
| _version_ | 1797553739932893184 |
|---|---|
| author | Cosima Hirschberg Magnus Edinger Else Holmfred Jukka Rantanen Johan Boetker |
| author_facet | Cosima Hirschberg Magnus Edinger Else Holmfred Jukka Rantanen Johan Boetker |
| author_sort | Cosima Hirschberg |
| collection | DOAJ |
| description | Mimicking the human decision-making process is challenging. Especially, many process control situations during the manufacturing of pharmaceuticals are based on visual observations and related experience-based actions. The aim of the present work was to investigate the use of image analysis to classify the quality of coated tablets. Tablets with an increasing amount of coating solution were imaged by fast scanning using a conventional office scanner. A segmentation routine was implemented to the images, allowing the extraction of numeric image-based information from individual tablets. The image preprocessing was performed prior to utilization of four different classification techniques for the individual tablet images. The support vector machine (SVM) technique performed superior compared to a convolutional neural network (CNN) in relation to computational time, and this approach was also slightly better at classifying the tablets correctly. The fastest multivariate method was partial least squares (PLS) regression, but this method was hampered by the inferior classification accuracy of the tablets. Finally, it was possible to create a numerical threshold classification model with an accuracy comparable to the SVM approach, so it is evident that there exist multiple valid options for classifying coated tablets. |
| first_indexed | 2024-03-10T16:19:54Z |
| format | Article |
| id | doaj.art-bc5b21301c3e4ba7bdc8242ce52a352a |
| institution | Directory Open Access Journal |
| issn | 1999-4923 |
| language | English |
| last_indexed | 2024-03-10T16:19:54Z |
| publishDate | 2020-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Pharmaceutics |
| spelling | doaj.art-bc5b21301c3e4ba7bdc8242ce52a352a2023-11-20T13:44:49ZengMDPI AGPharmaceutics1999-49232020-09-0112987710.3390/pharmaceutics12090877Image-Based Artificial Intelligence Methods for Product Control of Tablet Coating QualityCosima Hirschberg0Magnus Edinger1Else Holmfred2Jukka Rantanen3Johan Boetker4BASF A/S, Malmparken 5, 2750 Ballerup, DenmarkFaculty of Health and Medical Sciences, University of Copenhagen, 2100 Copenhagen, DenmarkResearch Group for Nano-Bio Science, National Food Institute, Technical University of Denmark, Kemitorvet, 2800 Kgs. Lyngby, DenmarkFaculty of Health and Medical Sciences, University of Copenhagen, 2100 Copenhagen, DenmarkFaculty of Health and Medical Sciences, University of Copenhagen, 2100 Copenhagen, DenmarkMimicking the human decision-making process is challenging. Especially, many process control situations during the manufacturing of pharmaceuticals are based on visual observations and related experience-based actions. The aim of the present work was to investigate the use of image analysis to classify the quality of coated tablets. Tablets with an increasing amount of coating solution were imaged by fast scanning using a conventional office scanner. A segmentation routine was implemented to the images, allowing the extraction of numeric image-based information from individual tablets. The image preprocessing was performed prior to utilization of four different classification techniques for the individual tablet images. The support vector machine (SVM) technique performed superior compared to a convolutional neural network (CNN) in relation to computational time, and this approach was also slightly better at classifying the tablets correctly. The fastest multivariate method was partial least squares (PLS) regression, but this method was hampered by the inferior classification accuracy of the tablets. Finally, it was possible to create a numerical threshold classification model with an accuracy comparable to the SVM approach, so it is evident that there exist multiple valid options for classifying coated tablets.https://www.mdpi.com/1999-4923/12/9/877in silico modellingneural networksimage analysisartificial intelligencemultivariate analysis |
| spellingShingle | Cosima Hirschberg Magnus Edinger Else Holmfred Jukka Rantanen Johan Boetker Image-Based Artificial Intelligence Methods for Product Control of Tablet Coating Quality Pharmaceutics in silico modelling neural networks image analysis artificial intelligence multivariate analysis |
| title | Image-Based Artificial Intelligence Methods for Product Control of Tablet Coating Quality |
| title_full | Image-Based Artificial Intelligence Methods for Product Control of Tablet Coating Quality |
| title_fullStr | Image-Based Artificial Intelligence Methods for Product Control of Tablet Coating Quality |
| title_full_unstemmed | Image-Based Artificial Intelligence Methods for Product Control of Tablet Coating Quality |
| title_short | Image-Based Artificial Intelligence Methods for Product Control of Tablet Coating Quality |
| title_sort | image based artificial intelligence methods for product control of tablet coating quality |
| topic | in silico modelling neural networks image analysis artificial intelligence multivariate analysis |
| url | https://www.mdpi.com/1999-4923/12/9/877 |
| work_keys_str_mv | AT cosimahirschberg imagebasedartificialintelligencemethodsforproductcontroloftabletcoatingquality AT magnusedinger imagebasedartificialintelligencemethodsforproductcontroloftabletcoatingquality AT elseholmfred imagebasedartificialintelligencemethodsforproductcontroloftabletcoatingquality AT jukkarantanen imagebasedartificialintelligencemethodsforproductcontroloftabletcoatingquality AT johanboetker imagebasedartificialintelligencemethodsforproductcontroloftabletcoatingquality |