Artificial intelligence and real-world data for drug and food safety – A regulatory science perspective

Thakkar, Shraddha; Slikker, William; Yiannas, Frank; Silva, Primal; Blais, Burton W.; Chng, Kern Rei; Liu, Zhichao; Adholeya, Alok; Pappalardo, Francesco; Soares, Mônica da Luz Carvalho; Beeler, Patrick E.; Whelan, Maurice; Roberts, Ruth; Borlak, Jürgen; Hugas, M; Torrecilla-Salinas, Carlos J.; Girard, Philippe; Diamond, Matthew C; Verloo, Didier; Panda, Binay; Rose, Miquella C; Jornet, Joaquim Berenguer; Furuhama, Ayako; Fang, Hong; Kwegyir-Afful, Ernest E.; Heintz, Kasey; Arvidson, Kirk; Burgos, Juan García; Horst, A. J. van der; Tong, Weida

doi:10.1016/j.yrtph.2023.105388

Cited by 11 publications

(6 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We then assessed RxBERT on two common NLP tasks: NER and Text Classification, both of which have already been widely applied in the regulatory research of drug labeling. [20][21][22] RxBERT showed competitive results compared to previous approaches in both tasks. In particular, for the drug labeling sentence classification task, RxBERT outperforms a collection of BERT-based models, including the biomedical domain-specific model BioBERT, which was pretrained on PubMed abstracts and used to initialize RxBERT.…”

Section: Discussionmentioning

confidence: 89%

RxBERT: Enhancing drug labeling text mining and analysis with AI language modeling

Wu,

Gray,

Dang

et al. 2023

Exp Biol Med (Maywood)

View full text Add to dashboard Cite

The US drug labeling document contains essential information on drug efficacy and safety, making it a crucial regulatory resource for Food and Drug Administration (FDA) drug reviewers. Due to its extensive volume and the presence of free-text, conventional text mining analysis have encountered challenges in processing these data. Recent advances in artificial intelligence (AI) for natural language processing (NLP) have provided an unprecedented opportunity to identify key information from drug labeling, thereby enhancing safety reviews and support for regulatory decisions. We developed RxBERT, a Bidirectional Encoder Representations from Transformers (BERT) model pretrained on FDA human prescription drug labeling documents for an enhanced application of drug labeling documents in both research and drug review. RxBERT was derived from BioBERT with further training on human prescription drug labeling documents. RxBERT was demonstrated in several tasks using regulatory datasets, including those involved in the National Institutes of Technology Text Analysis Challenge Dataset (NIST TAC dataset), the FDA Adverse Drug Event Evaluation Dataset (ADE Eval dataset), and the classification of texts from submission packages into labeling sections (US Drug Labeling dataset). For all these tasks, RxBERT reached 86.5 F1-scores in both TAC and ADE Eval classification, respectively, and prediction accuracy of 87% for the US Drug Labeling dataset. Overall, RxBERT was shown to be as competitive or have better performance compared to other NLP approaches such as BERT, BioBERT, etc. In summary, we developed RxBERT, a transformer-based model specific for drug labeling that outperformed the original BERT model. RxBERT has the potential to be used to assist research scientists and FDA reviewers to better process and utilize drug labeling information toward the advancement of drug effectiveness and safety for public health. This proof-of-concept study also demonstrated a potential pathway to customized large language models (LLMs) tailored to the sensitive regulatory documents for internal application.

show abstract

Section: Discussionmentioning

confidence: 89%

RxBERT: Enhancing drug labeling text mining and analysis with AI language modeling

Wu,

Gray,

Dang

et al. 2023

Exp Biol Med (Maywood)

View full text Add to dashboard Cite

show abstract

“…Real‐world data have already found uses in daily practice, whereas AI could help agencies to improve their operation. Data standardization will be key in this regard (Thakkar et al., 2023).…”

Section: Recommendationsmentioning

confidence: 99%

Making food systems more resilient to food safety risks by including artificial intelligence, big data, and internet of things into food safety early warning and emerging risk identification tools

Mu,

Kleter,

Bouzembrak

et al. 2024

Comp Rev Food Sci Food Safe

View full text Add to dashboard Cite

To enhance the resilience of food systems to food safety risks, it is vitally important for national authorities and international organizations to be able to identify emerging food safety risks and to provide early warning signals in a timely manner. This review provides an overview of existing and experimental applications of artificial intelligence (AI), big data, and internet of things as part of early warning and emerging risk identification tools and methods in the food safety domain. There is an ongoing rapid development of systems fed by numerous, real‐time, and diverse data with the aim of early warning and identification of emerging food safety risks. The suitability of big data and AI to support such systems is illustrated by two cases in which climate change drives the emergence of risks, namely, harmful algal blooms affecting seafood and fungal growth and mycotoxin formation in crops. Automation and machine learning are crucial for the development of future real‐time food safety risk early warning systems. Although these developments increase the feasibility and effectiveness of prospective early warning and emerging risk identification tools, their implementation may prove challenging, particularly for low‐ and middle‐income countries due to low connectivity and data availability. It is advocated to overcome these challenges by improving the capability and capacity of national authorities, as well as by enhancing their collaboration with the private sector and international organizations.

show abstract

“…Moreover, collaboration can facilitate communication and knowledge sharing among stakeholders, leading to improved decision-making and more effective risk assessments. Collaboration can also enhance transparency and accountability in the development and implementation of digital technologies, ensuring that the interests of all stakeholders are taken into account …”

Section: Challenges and Limitations Associated With Implementing Digi...mentioning

confidence: 99%

“…Collaboration can also enhance transparency and accountability in the development and implementation of digital technologies, ensuring that the interests of all stakeholders are taken into account. 83 The implementation of digital technologies in toxicology requires essential collaboration among interdisciplinary researchers and stakeholders since there are technical barriers such as the complexity of integrating different systems and platforms. Digital technologies rely on a range of hardware and software solutions, and compatibility issues can arise when integrating these technologies with existing systems.…”

Section: With Implementing Digital Technologies In Toxicologymentioning

confidence: 99%

Digital Transformation in Toxicology: Improving Communication and Efficiency in Risk Assessment

Singh,

Bansod,

Mahajan

et al. 2023

ACS Omega

View full text Add to dashboard Cite

Toxicology is undergoing a digital revolution, with mobile apps, sensors, artificial intelligence (AI), and machine learning enabling better recordkeeping, data analysis, and risk assessment. Additionally, computational toxicology and digital risk assessment have led to more accurate predictions of chemical hazards, reducing the burden of laboratory studies. Blockchain technology is emerging as a promising approach to increase transparency, particularly in the management and processing of genomic data related with food safety. Robotics, smart agriculture, and smart food and feedstock offer new opportunities for collecting, analyzing, and evaluating data, while wearable devices can predict toxicity and monitor health-related issues. The review article focuses on the potential of digital technologies to improve risk assessment and public health in the field of toxicology. By examining key topics such as blockchain technology, smoking toxicology, wearable sensors, and food security, this article provides an overview of how digitalization is influencing toxicology. As well as highlighting future directions for research, this article demonstrates how emerging technologies can enhance risk assessment communication and efficiency. The integration of digital technologies has revolutionized toxicology and has great potential for improving risk assessment and promoting public health.

show abstract

Artificial intelligence and real-world data for drug and food safety – A regulatory science perspective

Cited by 11 publications

References 39 publications

RxBERT: Enhancing drug labeling text mining and analysis with AI language modeling

RxBERT: Enhancing drug labeling text mining and analysis with AI language modeling

Making food systems more resilient to food safety risks by including artificial intelligence, big data, and internet of things into food safety early warning and emerging risk identification tools

Digital Transformation in Toxicology: Improving Communication and Efficiency in Risk Assessment

Contact Info

Product

Resources

About