Aim: To evaluate the implementation of a clinical pathway (CP) and identify clinical factors affecting the CP for cleft lip and palate (CLP) patients. Methods: A specific CP for CLP patients was developed at CLP Medical Center of Stomatological Hospital affiliated to Nanjing Medical University in 2008. The authors reviewed the collected data of 1810 consecutive patients using the CP for repairing cleft lip, cleft palatal, and alveolar cleft. The patients were treated between January 2008 and December 2019. The rate of completion and risk factors affecting dropout from the CP were analyzed. Results: The completion rates of the CP in cleft lip, cleft palate and alveolar cleft patients were 68.3% (n = 345), 82.4% (n = 785) and 76.1% (n = 268), respectively. The overall completion rate was 77.2% (n = 1398). The main reasons for dropping out were pre-operation events (n = 212, 11.7%) and post-operation events (n = 188, 10.4%). Among the factors of dropout of CP, laboratory test abnormalities accounted for the majority of pre- and post-operation events (n = 179, 9.9%). In statistical analysis, the combined abnormities and events associated with operations were significant risk factors affecting the dropout rate from CP. Conclusion: The use of CP for CLP patients was reliable but the completion rate was relatively low because of perioperative events. These results provided some evidence of risk factors which should be considered when modifying the protocol of CP for CLP patients in order to achieve higher completion rate.

Background Dispensing Error occurs frequently in outpatient pharmacies. The aim of this article is to explore the value of the YOLO-V5 deep learning algorithm for object detection in packaging drugs from outpatient pharmacies and to develop an artificial intelligence assist pharmacist drug dispensation system (AI-APDDS) capable of completely reducing or eliminating dispensing errors. Methods A total of 1784 images from 136 different packaging drugs were collected and labeled to form a deep learning dataset. The dataset was split into training and validation sets at a ratio of 3:1. The YOLO-V5 deep-learning algorithm was trained using images from our dataset (training epochs:1000, batch size:4, learning rate:0.01). The values of precision (Pr) and mean average precision(mAP) were used as measures for model performance evaluation. Results Pr in the training set was, for all models, equal to 1.00. The mAP_0.5 of YOLO-V5x was 0.992, which was higher than that of YOLO-V5n (0.975), YOLO-V5s (0.976), YOLO-V5m (0.978), and YOLO-V5l (0.983). The mAP_0.5:0.95 of YOLO-V5x was 0.955, which was higher than those of YOLO-V5n (0.916), YOLO-V5s (0.920), YOLO-V5m (0.921) and YOLO-V5l (0.935). The training time and model size were 92.67 hours and 465MB, respectively, for YOLO-V5x, which were the highest among the four models. The speed of detection for one image was 6.0ms for YOLO-V5n, which was the fastest among the four models. Conclusions It can identify the packaging drugs accurately for five submodels of YOLO-V5, it is feasible to implement the artificial-intelligence assisted drug dispensation system for pharmacist to achieve “zero” dispensing error.