RESUMO
Patient safety risks associated with the online purchase of medications, especially in case of ophthalmic preparations, are significant. Our study aimed to carry out quality assessment of dorzolamide hydrochloride (DZA) and timolol maleate (TIM) eye drops preserved with benzalkonium chloride (BAC) via online test purchases. Three samples were purchased online, while control preparations were acquired through authorized national drug supply chain. Our method was based on the International Pharmaceutical Federation (FIP) Inspection Checklist and integrated the evaluation of packaging and labelling. Sterility was established according to the European Pharmacopoeia (Ph. Eur.), while qualitative and quantitative quality was assessed with high-performance liquid chromatographic (HPLC) analysis. Several signs of falsification were recognized upon visual inspection of the online samples. All the products were clear, colourless, slightly viscous solutions. They were free from visible contaminants. The samples were sterile as no evidence of microbial growth was found. A quick and inexpensive HPLC analysis, optimized by the authors showed that active ingredients and the preservative deviated significantly (p < 0,05) with more than 10% from the values stated on the labels for at least one component (DZA: 99.3-113.1%, TIM: 112.8-139.2%, BAC: 82.4-97.7%). Development of comprehensive and reliable quality assessment methods are vital to increase public safety of pharmaceutical products sold online. A complex approach, integrating visual inspection, labelling assessment, microbiological analysis coupled with qualitative and quantitative methods provide a most reliable method. Due to its limited feasibility and cost-effectiveness, raising public awareness and limiting illegal online sellers should be the primary approaches to protect patients from substandard and falsified medicinal products sold via the internet. Particularly important for health professionals to understand this market and its public health concern, and to raise patient awareness of the risks associated with uncontrolled online purchase of medication.
RESUMO
Today, the increasing number of illicit internet pharmacies is a global phenomenon, however, the size of the online pharmaceutical market is still relatively unknown and the dubious quality of products is questionable and warrants investigation. Descriptive data from this black market channel are derived from studies analyzing the online availability of different medications procured over the internet and their methodology is quite heterogeneous. Our aim was to develop a comprehensive and specific risk assessment for selecting high patient safety risk medications from the online pharmaceutical market. A rapid tool was developed based upon the two quality and safety standard resolutions in pharmaceutical practice, published by the European Directorate for the Quality of Medicines, and was illustrated on eye drops. We developed five dimensions in support of the risk assessment including intrinsic, extrinsic and potential risks of counterfeiting. The five criteria were integrated in a comprehensively weighted risk-scoring format. The probability of procuring the product from the internet was also assessed based on the number of relevant links within the first twenty search engine results and the cost of the products. With the application of the tool a dorzolamide & timolol combination eye drop represented the highest overall patient safety risk score. In consideration of our literature review of the past 20 years, there is no current, standardized methodology to effectively identify pharmaceutical products associated with high patient safety risks. Notably, the fully comprehensive analysis of the internet pharmaceutical market and the test purchase of all online available medicines is unrealistic. Therefore, we developed a method to aid online surveillance researches and targeted international organizational led joint actions against the uncontrolled sale of falsified and substandard medications (e.g.: Operation Pangea).
RESUMO
BACKGROUND: Preventable patient harm, particularly medication errors, represent significant challenges in healthcare settings. Dispensing the wrong medication is often associated with mix-up of lookalike and soundalike drugs in high workload environments. Replacing manual dispensing with automated unit dose and medication dispensing systems to reduce medication errors is not always feasible in clinical facilities experiencing high patient turn-around or frequent dose changes. Artificial intelligence (AI) based pill recognition tools and smartphone applications could potentially aid healthcare workers in identifying pills in situations where more advanced dispensing systems are not implemented. OBJECTIVE: Most of the published research on pill recognition focuses on theoretical aspects of model development using traditional coding and deep learning methods. The use of code-free deep learning (CFDL) as a practical alternative for accessible model development, and implementation of such models in tools intended to aid decision making in clinical settings, remains largely unexplored. In this study, we sought to address this gap in existing literature by investigating whether CFDL is a viable approach for developing pill recognition models using a custom dataset, followed by a thorough evaluation of the model across various deployment scenarios, and in multicenter clinical settings. Furthermore, we aimed to highlight challenges and propose solutions to achieve optimal performance and real-world applicability of pill recognition models, including when deployed on smartphone applications. METHODS: A pill recognition model was developed utilizing Microsoft Azure Custom Vision platform and a large custom training dataset of 26,880 images captured from the top 30 most dispensed solid oral dosage forms (SODFs) at the three participating hospitals. A comprehensive internal and external testing strategy was devised, model's performance was investigated through the online API, and offline using exported TensorFlow Lite model running on a Windows PC and on Android, using a tailor-made testing smartphone application. Additionally, model's calibration, degree of reliance on color features and device dependency was thoroughly evaluated. Real-world performance was assessed using images captured by hospital pharmacists at three participating clinical centers. RESULTS: The pill recognition model showed high performance in Microsoft Azure Custom Vision platform with 98.7 % precision, 95.1 % recall, and 98.2 % mean average precision (mAP), with thresholds set to 50 %. During internal testing utilizing the online API, the model reached 93.7 % precision, 88.96 % recall, 90.81 % F1-score and 87.35 % mAP. Testing the offline TensorFlow Lite model on Windows PC showed a slight performance reduction, with 91.16 % precision, 83.82 % recall, 86.18 % F1-score and 82.55 % mAP. Performance of the model running offline on the Android application was further reduced to 86.50 % precision, 75.00 % recall, 77.83 % F1-score and 69.24 % mAP. During external clinical testing through the online API an overall precision of 83.10 %, recall of 71.39 %, and F1-score of 75.76 % was achieved. CONCLUSION: Our study demonstrates that using a CFDL approach is a feasible and cost-effective method for developing AI-based pill recognition systems. Despite the limitations encountered, our model performed well, particularly when accessed through the online API. The use of CFDL facilitates interdisciplinary collaboration, resulting in human-centered AI models with enhanced real-world applicability. We suggest that rather than striving to build a universally applicable pill recognition system, models should be tailored to the medications in a regional formulary or needs of a specific clinic, which can in turn lead to improved performance in real-world deployment in these locations. Parallel to focusing on model development, it is crucial to employ a human centered approach by training the end users on how to properly interact with the AI based system to maximize benefits. Future research is needed on refining pill recognition models for broader adaptability. This includes investigating image pre-processing and optimization techniques to enhance offline performance and operation on handheld devices. Moreover, future studies should explore methods to overcome limitations of CFDL development to enhance the robustness of models and reduce overfitting. Collaborative efforts between researchers in this domain and sharing of best practices are vital to improve pill recognition systems, ultimately enhancing patient safety and healthcare outcomes.
Assuntos
Inteligência Artificial , Aprendizado Profundo , Humanos , Reconhecimento Psicológico , Corantes AzurRESUMO
Microbiological preservatives play a great role in the preparation of artificial tears because they protect the eyes from further microorganisms and the preparation from contamination. In this contribution we are summarizing our experimental results given by pharmaceutical and microbiological optimalization of artificial tears. The incidental adaptability of povidone-iodine (PVP-I) as a preservative in artificial tears was examined compared to usually used materials. Some artificial tears (Oculogutta carbomerae and Oculogutta viscosa) were prepared according to the Formulae Normales Edition VII., others were isotonisated and buffered containing 3.0% and 3.5% povidonum as active substance. The analysed samples as a preservative instead of generally used agents contained 0.10%, 0.05% and 0.01% PVP-I. Reference preparations were dispensed using microbiological preservatives (Cetrimidum, Thiomersalum solutum 0.1%, Benzalconium chloratum solutum 10%). Pharmaceutical (pH, viscosity, freezing-point depression, refraction, surface-tension) and microbiological (breeding on aerobe and anaerobe bacteriological culture medium) trials were made to determine the qualitative property and adaptability of analysed preparations in which we also studied the stability and the microbiological changes after opening them. According to our experimental results we can establish that the PVP-I is suitable as microbiological preservative in the examined preparations.