Yuen Fei Wong1, Hoi Ting Ng2, Kit Yee Leung3, Ka Yan Chan4, Sau Yi Chan4, Chen Change Loy3. 1. Department of Pharmacy, Faculty of Medicine, University of Malaya, Malaysia. Electronic address: yfwong@um.edu.my. 2. Department of Pharmacy, Tuen Moon Hospital, Hong Kong. 3. Department of Information Engineering, The Chinese University of Hong Kong, Hong Kong. 4. Department of Health Sciences, Caritas Bianchi College of Careers, Hong Kong.
Abstract
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.
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.