Development of fine-grained pill identification algorithm using deep convolutional network
Objective Oral pills, including tablets and capsules, are one of the most popular pharmaceutical dosage forms available. Compared to other dosage forms, such as liquid and injections, oral pills are very stable and are easy to be administered. However, it is not uncommon for pills to be misidentifie...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Published: |
Elsevier
2017
|
| Subjects: |
| _version_ | 1825721938319769600 |
|---|---|
| author | Wong, Yuen Fei Ng, Hoi Ting Leung, Kit Yee Chan, Ka Yan Chan, Sau Yi Loy, Chen Change |
| author_facet | Wong, Yuen Fei Ng, Hoi Ting Leung, Kit Yee Chan, Ka Yan Chan, Sau Yi Loy, Chen Change |
| author_sort | Wong, Yuen Fei |
| collection | UM |
| description | Objective Oral pills, including tablets and capsules, are one of the most popular pharmaceutical dosage forms available. Compared to other dosage forms, such as liquid and injections, oral pills are very stable and are easy to be administered. However, it is not uncommon for pills to be misidentified, be it within the healthcare institutes or after the pills were dispensed to the patients. Our objective is to develop groundwork for automatic pill identification and verification using Deep Convolutional Network (DCN) that surpasses the existing methods. Materials and methods A DCN model was developed using pill images captured with mobile phones under unconstraint environments. The performance of the DCN model was compared to two baseline methods of hand-crafted features. Results The DCN model outperforms the baseline methods. The mean accuracy rate of DCN at Top-1 return was 95.35%, whereas the mean accuracy rates of the two baseline methods were 89.00% and 70.65%, respectively. The mean accuracy rates of DCN for Top-5 and Top-10 returns, i.e., 98.75% and 99.55%, were also consistently higher than those of the baseline methods. Discussion The images used in this study were captured at various angles and under different level of illumination. DCN model achieved high accuracy despite the suboptimal image quality. Conclusion The superior performance of DCN underscores the potential of Deep Learning model in the application of pill identification and verification. |
| first_indexed | 2024-03-06T05:57:59Z |
| format | Article |
| id | um.eprints-22782 |
| institution | Universiti Malaya |
| last_indexed | 2024-03-06T05:57:59Z |
| publishDate | 2017 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | um.eprints-227822019-10-21T07:59:44Z http://eprints.um.edu.my/22782/ Development of fine-grained pill identification algorithm using deep convolutional network Wong, Yuen Fei Ng, Hoi Ting Leung, Kit Yee Chan, Ka Yan Chan, Sau Yi Loy, Chen Change R Medicine RS Pharmacy and materia medica Objective Oral pills, including tablets and capsules, are one of the most popular pharmaceutical dosage forms available. Compared to other dosage forms, such as liquid and injections, oral pills are very stable and are easy to be administered. However, it is not uncommon for pills to be misidentified, be it within the healthcare institutes or after the pills were dispensed to the patients. Our objective is to develop groundwork for automatic pill identification and verification using Deep Convolutional Network (DCN) that surpasses the existing methods. Materials and methods A DCN model was developed using pill images captured with mobile phones under unconstraint environments. The performance of the DCN model was compared to two baseline methods of hand-crafted features. Results The DCN model outperforms the baseline methods. The mean accuracy rate of DCN at Top-1 return was 95.35%, whereas the mean accuracy rates of the two baseline methods were 89.00% and 70.65%, respectively. The mean accuracy rates of DCN for Top-5 and Top-10 returns, i.e., 98.75% and 99.55%, were also consistently higher than those of the baseline methods. Discussion The images used in this study were captured at various angles and under different level of illumination. DCN model achieved high accuracy despite the suboptimal image quality. Conclusion The superior performance of DCN underscores the potential of Deep Learning model in the application of pill identification and verification. Elsevier 2017 Article PeerReviewed Wong, Yuen Fei and Ng, Hoi Ting and Leung, Kit Yee and Chan, Ka Yan and Chan, Sau Yi and Loy, Chen Change (2017) Development of fine-grained pill identification algorithm using deep convolutional network. Journal of Biomedical Informatics, 74. pp. 130-136. ISSN 1532-0464, DOI https://doi.org/10.1016/j.jbi.2017.09.005 <https://doi.org/10.1016/j.jbi.2017.09.005>. https://doi.org/10.1016/j.jbi.2017.09.005 doi:10.1016/j.jbi.2017.09.005 |
| spellingShingle | R Medicine RS Pharmacy and materia medica Wong, Yuen Fei Ng, Hoi Ting Leung, Kit Yee Chan, Ka Yan Chan, Sau Yi Loy, Chen Change Development of fine-grained pill identification algorithm using deep convolutional network |
| title | Development of fine-grained pill identification algorithm using deep convolutional network |
| title_full | Development of fine-grained pill identification algorithm using deep convolutional network |
| title_fullStr | Development of fine-grained pill identification algorithm using deep convolutional network |
| title_full_unstemmed | Development of fine-grained pill identification algorithm using deep convolutional network |
| title_short | Development of fine-grained pill identification algorithm using deep convolutional network |
| title_sort | development of fine grained pill identification algorithm using deep convolutional network |
| topic | R Medicine RS Pharmacy and materia medica |
| work_keys_str_mv | AT wongyuenfei developmentoffinegrainedpillidentificationalgorithmusingdeepconvolutionalnetwork AT nghoiting developmentoffinegrainedpillidentificationalgorithmusingdeepconvolutionalnetwork AT leungkityee developmentoffinegrainedpillidentificationalgorithmusingdeepconvolutionalnetwork AT chankayan developmentoffinegrainedpillidentificationalgorithmusingdeepconvolutionalnetwork AT chansauyi developmentoffinegrainedpillidentificationalgorithmusingdeepconvolutionalnetwork AT loychenchange developmentoffinegrainedpillidentificationalgorithmusingdeepconvolutionalnetwork |