A new methodology for detection of counterfeit Viagra® and Cialis® tablets by image processing and statistical analysis

Deep learning-based defect detection in film-coated tablets using a convolutional neural network

Film-coating is a critical step in pharmaceutical manufacturing. Traditional visual inspections for film-coated tablet defect assessment are subjective, inefficient, and labor-intensive. We propose a novel approach utilizing machine learning and image analysis to address these limitations. Here, defects of four types- chipping, breaking, color non-uniformity and speckling, were manually induced in red-orange film-coated placebo tablets. Utilizing a 3-D printed tray and a unique segmentation approach, images of good and defective tablets were collected. A convolutional neural network (CNN) was employed to quantitatively analyze the defects. The model was trained on a comprehensive dataset of 25,200 images of tablets, augmented through various transformations to improve robustness. The CNN's performance was evaluated using metrics such as accuracy, precision, recall, and F1-score. The multi-class classification model demonstrated an accuracy of 99.7% in detection of defects in film-coated tablets, clearly outperforming static rule-based method which had 45%, 45% and 70% error in detecting dimensions- major axis, minor axis, and surface area of the tablets, respectively. This work demonstrates a valuable tool for pharmaceutical manufacturers, providing a standardized, objective, and efficient method for defect detection in tablets and presents a promising solution for ensuring product quality and accelerating the development of new pharmaceutical products.

Radio frequency identification technology; A method of analysis of falsified pharmaceutical products: Literature review

Quality has been a concern of the World Health Organization since its inception and is defined as fitness for use. Since our ancestors began trading several millennia ago, Falsified Pharmaceutical Products has been a recurring problem and still threatens economic stability and public health. Its definition various from country to country and according to World Health Organization, 2017, it is 'a product that is deliberately and fraudulently mislabelled with respect to identity and/or source'. The implementation of anti-falsified nanomaterial technologies is the prominent preventive measure to track and/or detect Falsified Pharmaceutical Products. Software and hardware companies had made encouraging progress towards implementing Radio Frequency Identification devices for ensuring the authenticity of pharmaceutical products. The purpose of the review was to critically appraise Radio Frequency Identification devices technology for the purpose of track and trace Falsified Pharmaceutical Products circulating in the market. Different search engines such as Google Scholar, Science Direct and PubMed were applied and mesh terms and keywords were searched. Different guides and related books were investigated in addition to the articles. Radio Frequency Identification devices technology is a compact electronic device that contains a small chip and reader with antenna that enables wireless transmission of identity of pharmaceutical products. The authenticated Radio Frequency Identification devices model is used for pharmaceutical products' authentication from origin of pharmaceutical industry to the pharmacy at any point along the chain of the distribution. Popular pharmaceutical products, such as OxyContin and Sildenafil Citrate, which are particular targets of falsification have mandated the use of Radio Frequency Identification devices technology.

Trends in counterfeit drugs and pharmaceuticals before and during COVID-19 pandemic

Counterfeit, fake, adulterated or falsified drugs and pharmaceuticals, could be branded or generic drugs, excipients and active substances (in drugs and vaccines), medical supplies and devices, etc, intended to pass as the original. Counterfeits are always inferior in terms of quality, safety and efficacy compared to the original pharmaceuticals, and subsequently, they pose an unpredictable risk to public health and lead to loss of confidence in medicines, healthcare providers, and health systems. In the decades before the outbreak of the COVID-19 pandemic, a constant trend of increased trafficking was reported. However, the pandemic created a combination of public health emergency, economic distress, and misinformation-driven panic that made problematic the access and supply of high quality essential medicines and health products, and pushed consumers and vendors even more towards counterfeit pharmaceuticals. This contribution aims to review the trends in counterfeit drugs and pharmaceuticals trafficking, the health impact of their use, as well as, measures and actions implemented to restrict their proliferation, before and during COVID-19 pandemic; the relative recommendations, the expressed perspectives and the existing limitations are thoroughly discussed.

Surface Analysis Techniques in Forensic Science: Successes, Challenges, and Opportunities for Operational Deployment

Surface analysis techniques have rapidly evolved in the last decade. Some of these are already routinely used in forensics, such as for the detection of gunshot residue or for glass analysis. Some surface analysis approaches are attractive for their portability to the crime scene. Others can be very helpful in forensic laboratories owing to their high spatial resolution, analyte coverage, speed, and specificity. Despite this, many proposed applications of the techniques have not yet led to operational deployment. Here, we explore the application of these techniques to the most important traces commonly found in forensic casework. We highlight where there is potential to add value and outline the progress that is needed to achieve operational deployment. We consider within the scope of this review surface mass spectrometry, surface spectroscopy, and surface X-ray spectrometry. We show how these tools show great promise for the analysis of fingerprints, hair, drugs, explosives, and microtraces.

Identifying the most relevant tablet regions in the image detection of counterfeit medicines

This paper proposes a novel image-based approach to detect counterfeit medicines and identify the most relevant regions of the tablet in the task of classification. Images of medicine tablets undergo an initial pre-processing step which (i) removes the background to find the region of interest, (ii) clusters individual pixels into super-pixels, and (iii) extracts features containing color and texture information. The classification relying on Support Vector Machine (SVM) defines the class the respective image will be inserted into. The task of identifying the relevant regions of the tablets for counterfeiting detection is performed using the concept of support vectors, generating a heat map that indicates the regions that contribute the most to the classification purpose. Two datasets containing images of authentic and counterfeit tablets of Cialis and Viagra were used to validate our propositions, achieving correct classification rates of 100% on both datasets. Regarding the task of identifying the most relevant regions, our proposition outperformed the traditional LIME (Local Interpretable Model-agnostic Explanations) method by yielding more robust explanations.

Device-Independent Discrimination of Falsified Amoxicillin Capsules Using Heterogeneous Near-Infrared Spectroscopic Devices for Training and Testing of a Support Vector Machine

The objective of this work is to demonstrate the potential of near-infrared spectroscopy for common screening of falsified medicines in the field by means of a device-independent universal discrimination approach. In order to provide a useful discrimination tool to protect people from low-quality medical products, not only is a low-cost and portable screening device necessary, but a reference library is also essential. The authors believe that a device-dependent reference library inhibits near-infrared spectroscopy from becoming a popular screening tool. In this study, to develop a device-independent method, discrimination performance is evaluated using different devices for training and testing. The training data sets for the reference library were prepared using a bench-top Fourier transform near-infrared spectrophotometer, and predictive discrimination was performed using the spectral data by a low-cost and portable wavelength dispersive near-infrared spectrophotometer. A near-infrared spectrum-based support vector machine was used for these purposes, but the screening resulted in low accuracy thought to be caused by the intrinsically device-dependent features of the spectra data. Thus, principal component analysis was performed to collect the proper components to discriminate low-quality products from standard products. The principal component score-based support vector machine was able to produce highly accurate results, identifying falsified products with no false positive cases.

Forensic analysis of anabolic steroids tablets composition using attenuated total reflection Fourier transform infrared microspectroscopy (µATR-FTIR) mapping

The use of falsified and unregistered drugs is a worldwide public health problem. Because these global market products usually do not follow the Good Manufacturing Practices required by health legislation, its composition may be completely different from the original or may contain relevant concentrations of impurities and toxic contaminants. Since anabolic steroids are among the main irregular therapeutic classes seized in Brazil, here we propose a new methodology for analyzing these products, in tablets form, using Attenuated Total Reflection Fourier Transform Infrared Microspectroscopy (µATR-FTIR) mapping. Spectra were acquired from solid tablets by attenuated total reflection, through point mapping methodology. In data processing, a characteristic absorption band for each Active Pharmaceutical Ingredient (API) was integrated and plotted to create its distribution map. This technique was applied in an unprecedented way for the forensic analysis of anabolic steroids and proved to be effective in distinguishing falsified products based on the detection of their APIs. It was possible to detect APIs in 26 out of 30 samples, five of which were classified as falsified only through µATR-FTIR analysis. We were able to create distribution maps of the detected substances associating the microspectroscopic results with characteristic band integration method, which can be used to detect substances and to study samples' homogeneity. We concluded that this methodology is promising for the analysis of anabolic steroid tablets, and can be used in a complementary way with techniques already consolidated in forensic laboratory routine for a better classification of questioned samples between authentic and falsified ones.

Wavenumber selection based on Singular Value Decomposition for sample classification

The dissemination of falsified medicines is a public health risk. Techniques such as attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy are commonly adopted for fraudulent drug detection. However, the spectrum generated by the ATR-FTIR typically results in hundreds of wavenumbers, reducing the performance of classification methods aimed at discriminating between authentic and falsified medicines. This article proposes a novel method for selecting a reduced size subset of wavenumbers that improves the classifier performance. The singular value decomposition SVD is used to generate a wavenumber importance index. An iterative process creates k-nearest neighbor (KNN) models by adding the wavenumbers in a decreasing order according to the importance index. Wavenumbers that increase classification accuracy are selected. When applied to Cialis® ATR-FTIR data, the proposed approach retained average 0.51% of the original wavenumbers with 100% accurate classifications; as for the Viagra® data set, the method yielded perfect classifications retaining average 0.17% of the original wavenumbers.

Raman spectroscopy and mapping technique for the identification of expired drugs

As expired medical products can be repackaged and sold by unscrupulous counterfeiters, it is essential to find a rapid and convenient method for distinguishing expired and unexpired drugs. Standard detection methods such as high-performance liquid chromatography (HPLC) and thin-layer chromatography are complex, time-consuming, and require organic solvents (that are environmentally unfriendly). Additionally, the Pharmacopoeia publications do not include information about identifying expired drugs. In this study, we proposed a novel method for identifying expired medications based on Raman spectra and verified it using >20 types of expired (Old) and unexpired (New) drugs, each type from the same manufacturer. A portable Raman spectrometer was used to collect Raman spectra of all samples and the similarities between the Old and New drugs (S_N-O) were evaluated. Drugs with S_N-O values <0.9 were classified directly as expired drugs. For drugs with S_N-O values >0.9, the content of active pharmaceutical ingredient (API) might be so low (below or around 10 wt%) that its Raman signal is largely obscured by that of the excipients. In such cases, changes in the API content are undetectable using the portable instrument. Therefore, we adopted Raman mapping technology and established a virtual imaging map to locate areas of high API content. The similarities between the Old or New spectrum and that of the API (S_O-A and S_N-A, respectively) were calculated after removing the signal from the excipients. Our novel methods provide a precise, rapid, convenient, and environmentally friendly way to identify expired drugs that is more effective than the standard HPLC assay.

Instrumental neutron activation analysis (INAA) and liquid chromatography (LC) coupled to high resolution mass spectrometry (HRMS) characterisation of sildenafil based products seized on the Italian illegal market

The commerce of illegal and counterfeit medicinal products on internet is a serious criminal problem. Drugs for erectile dysfunction such as phosphodiesterase type 5 inhibitor are the most commonly counterfeited medicines in Europe. The search of possible toxic chemical substances in seized products is needed. Moreover, the profiling of the material can be the source of relevant forensic information. For the first time a combined approach based on liquid chromatography (LC) coupled to high resolution mass spectrometry (HRMS) and instrumental neutron activation analysis (INAA) is proposed and tested, allowing characterisation of both authentic and illegal pharmaceuticals containing sildenafil seized in Italy. LC-HRMS allowed the detection and identification of unknown impurities not reported on labels in illegal products and the quantitation of the sildenafil. INAA showed to be suitable to provide both qualitative and quantitative information for forensic purposes on 23 elements, allowing discrimination between legal and illegal products.

Decision Support for Breast Cancer Detection: Classification Improvement Through Feature Selection

Several statistical-based approaches have been developed to support medical personnel in early breast cancer detection. This article presents a method for feature selection aimed at classifying cases into categories based on patients' breast tissue measures and protein microarray. The effectiveness of this feature selection strategy was evaluated against the commonly used Wisconsin Breast Cancer Database-WBCD (with several patients and fewer features) and a new protein microarray data set (with several features and fewer patients). Features were ranked according to a feature importance index that combines parameters emerging from the unsupervised method of principal component analysis and the supervised method of Bhattacharyya distance. Observations of a training set were iteratively categorized into malignant and benign cases through 3 classification techniques: k-Nearest Neighbor, linear discriminant analysis, and probabilistic neural network. After each classification, the feature with the smallest importance index was removed, and a new categorization was carried out until there was only one feature left. The subset yielding maximum accuracy was used to classify observations in the testing set. Our method yielded average 99.17% accurate classifications in the testing set while retaining average 4.61 out of 9 features in the WBCD, which is comparable to the best results reported by the literature on that data set, with the advantage of relying on simple and widely available multivariate techniques. When applied to the microarray data, the method yielded average accuracy of 98.30% while retaining average 2.17% of the original features. Our results can aid health-care professionals during early diagnosis of breast cancer.

A multivariate-based conflict prediction model for a Brazilian freeway

Real-time collision risk prediction models relying on traffic data can be useful in dynamic management systems seeking at improving traffic safety. Models have been proposed to predict crash occurrence and collision risk in order to proactively improve safety. This paper presents a multivariate-based framework for selecting variables for a conflict prediction model on the Brazilian BR-290/RS freeway. The Bhattacharyya Distance (BD) and Principal Component Analysis (PCA) are applied to a dataset comprised of variables that potentially help to explain occurrence of traffic conflicts; the parameters yielded by such multivariate techniques give rise to a variable importance index that guides variables removal for later selection. Next, the selected variables are inserted into a Linear Discriminant Analysis (LDA) model to estimate conflict occurrence. A matched control-case technique is applied using traffic data processed from surveillance cameras at a segment of a Brazilian freeway. Results indicate that the variables that significantly impacted on the model are associated to total flow, difference between standard deviation of lanes' occupancy, and the speed's coefficient of variation. The model allowed to asses a characteristic behavior of major Brazilian's freeways, by identifying the Brazilian typical heterogeneity of traffic pattern among lanes, which leads to aggressive maneuvers. Results also indicate that the developed LDA-PCA model outperforms the LDA-BD model. The LDA-PCA model yields average 76% classification accuracy, and average 87% sensitivity (which measures the rate of conflicts correctly predicted).

Chemometrics and the identification of counterfeit medicines-A review

This review article provides readers with a number of actual case studies dealing with verifying the authenticity of selected medicines supported by different chemometric approaches. In particular, a general data processing workflow is discussed with the major emphasis on the most frequently selected instrumental techniques to characterize drug samples and the chemometric methods being used to explore and/or model the analytical data. However, further discussion is limited to a situation in which the collected data describes two groups of drug samples - authentic ones and counterfeits.

Comparison of several chemometric methods of libraries and classifiers for the analysis of expired drugs based on Raman spectra

Some expired drugs are difficult to detect by conventional means. If they are repackaged and sold back into market, they will constitute a new public health challenge. For the detection of repackaged expired drugs within specification, paracetamol tablet from a manufacturer was used as a model drug in this study for comparison of Raman spectra-based library verification and classification methods. Raman spectra of different batches of paracetamol tablets were collected and a library including standard spectra of unexpired batches of tablets was established. The Raman spectrum of each sample was identified by cosine and correlation with the standard spectrum. The average HQI of the suspicious samples and the standard spectrum were calculated. The optimum threshold values were 0.997 and 0.998 respectively as a result of ROC and four evaluations, for which the accuracy was up to 97%. Three supervised classifiers, PLS-DA, SVM and k-NN, were chosen to establish two-class classification models and compared subsequently. They were used to establish a classification of expired batches and an unexpired batch, and predict the suspect samples. The average accuracy was 90.12%, 96.80% and 89.37% respectively. Different pre-processing techniques were tried to find that first derivative was optimal for methods of libraries and max-min normalization was optimal for that of classifiers. The results obtained from these studies indicated both libraries and classifier methods could detect the expired drugs effectively, and they should be used complementarily in the fast-screening.