Review
Medicine, General & Internal
Phoebe Clark, Jayne Kim, Yindalon Aphinyanaphongs
Summary: A systematic review of FDA-approved AI- or ML-enabled medical devices found that one-fifth of the devices surveyed had discrepancies between their clearance documentation and marketing material. The inconsistencies were more prominent in radiological and cardiovascular devices.
Article
Biochemistry & Molecular Biology
Qi Lv, Feilong Zhou, Xinhua Liu, Liping Zhi
Summary: This review provides an overview of the utilization of artificial intelligence in drug design, focusing on protein structure prediction, molecular virtual screening, molecular design, and ADMET prediction. The role and limitations of AI in drug development are discussed, along with its impact on decision-making processes.
BIOORGANIC CHEMISTRY
(2023)
Article
Multidisciplinary Sciences
Mitsuru Yuba, Kiyotaka Iwasaki
Summary: This study compares the requirements for AI/ML-based CAD devices approved by the FDA in the US and the PMDA in Japan, highlighting the differences in evaluation methods based on the countries' viewpoints. The findings of this study may be useful for defining a unified global development and approval standard for AI/ML-based CAD.
SCIENTIFIC REPORTS
(2022)
Review
Food Science & Technology
Zhe Liu, Shuzhe Wang, Yudong Zhang, Yichen Feng, Jiajia Liu, Hengde Zhu
Summary: This article presents a quantitative and systematic review of the development of AI technologies in food safety. It identifies China and the United States as the countries with the most published literature in this field, and the Chinese Academy of Sciences as the institution with the highest number of relevant articles. The article also highlights the current hotspots and future research trends in AI technologies in food safety, providing a comprehensive overview for researchers, practitioners, and policymakers through 28 enlightening articles.
Review
Environmental Sciences
Rohan Mark Bennett, Mila Koeva, Kwabena Asiama
Summary: This paper challenges the notion that photogrammetry and remote sensing are recent additions to land administration, showcasing their successful application dating back much earlier and often complementing ground-based methods. Through the study of historical works, the paper provides a more enriched and complete synthesis of the development of these technologies in the context of land administration.
Review
Biochemistry & Molecular Biology
Kevser Kubra Kirboga, Sumra Abbasi, Ecir Ugur Kucuksille
Summary: Recently, the use of AI techniques in drug discovery has increased. However, the traditional AI techniques may lack explanations for decision processes. Explainable AI (XAI) techniques provide a better understanding of the causes and consequences of decision processes, improving drug discovery and decision-making.
CHEMICAL BIOLOGY & DRUG DESIGN
(2023)
Review
Pharmacology & Pharmacy
Zachary F. Greenberg, Kiley S. Graim, Mei He
Summary: Extracellular Vesicles (EVs), especially exosomes, have recently emerged as a promising drug delivery approach in nanomedicine due to their high biocompatibility, stability, and bioavailability in vivo. However, the heterogeneity of EVs poses a challenge in achieving molecular targeting precision. The integration of large-scale EV data with artificial intelligence (AI) provides a powerful strategy for the rational design of engineered EVs in precision drug delivery.
ADVANCED DRUG DELIVERY REVIEWS
(2023)
Article
Toxicology
Onat Kadioglu, Sabine M. Klauck, Edmond Fleischer, Letian Shan, Thomas Efferth
Summary: The development of a toxicity prediction platform using machine-learning strategies successfully identified safe artemisinin derivatives and achieved high AUC scores for cardiotoxicity indications. Experimental validation supported the in silico cardiotoxicity model predictions.
ARCHIVES OF TOXICOLOGY
(2021)
Review
Food Science & Technology
Arzoo Thapa, Shivani Nishad, Deblina Biswas, Swarup Roy
Summary: With the increasing global population and growing demand for food, the application of artificial intelligence and sensors in the food industry has become more widespread. AI-assisted systems can improve food quality, provide control mechanisms, categorize foods, and predict demand, while sensors can be used for freshness sensing, pathogen detection, and other functions. This technological fusion has the potential to revolutionize the entire food industry, enabling efficient food manufacturing, optimal food processing, and high-quality food production and preservation.
Article
Food Science & Technology
Poonam Sharma, Archana Vimal, Reena Vishvakarma, Pradeep Kumar, Luciana Porto de Souza Vandenberghe, Vivek Kumar Gaur, Sunita Varjani
Summary: It is crucial to stop food wastage in order to address the challenges of starvation, hunger, and malnutrition worldwide. The adoption of modern techniques and policies formulated by governments can lead to sustainable solutions for reducing food loss, ensuring food security, and saving the lives of thousands of individuals who die from starvation every day.
INTERNATIONAL JOURNAL OF FOOD MICROBIOLOGY
(2022)
Article
Food Science & Technology
Sungmin Jeong, Dayeon Lee, Geunhyuk Yang, Hyukjin Kwon, Minhyo Kim, Suyong Lee
Summary: The physicochemical features of US wheat flours harvested from 1999 to 2020 were analyzed using machine learning to classify wheat flour varieties and predict bread loaf volumes. Protein content and water absorption were the main factors influencing the classification. The multilayer perceptron neural network with tuned hyperparameters showed the highest accuracy and F1-score in classification, while the use of Adam optimizer with adjusted hyperparameters greatly improved the accuracy in predicting bread loaf volume.
LWT-FOOD SCIENCE AND TECHNOLOGY
(2022)
Editorial Material
Cell Biology
Natalia Moskal, G. Angus McQuibban
Summary: We used artificial intelligence to simplify the small molecule drug screening process and discovered the cholesterol-reducing compound probucol. Probucol enhanced mitophagy and protected dopaminergic neurons from mitochondrial toxins in flies and zebrafish. Further investigation revealed that probucol modulates mitophagy through its target protein ABCA1, which is required for the regulation of lipid droplet dynamics during mitophagy. In this article, we will summarize the combination of in silico and cell-based screening that led us to identify and characterize probucol as a compound that enhances mitophagy, and discuss future directions for research in this area.
Review
Chemistry, Medicinal
Prafulla C. Tiwari, Rishi Pal, Manju J. Chaudhary, Rajendra Nath
Summary: By using AI algorithms and machine learning techniques, the drug discovery process can be transformed, offering numerous advantages. AI can quickly and efficiently screen large compound libraries, greatly increasing the identification of potential drug candidates. AI algorithms can also predict the efficacy and safety profiles of candidate compounds, providing valuable insights and reducing reliance on extensive testing.
DRUG DEVELOPMENT RESEARCH
(2023)
Article
Computer Science, Artificial Intelligence
Kerstin N. Vokinger, Urs Gasser
Summary: Regulatory frameworks for artificial intelligence are being developed on both sides of the Atlantic, eagerly anticipated by the scientific and industrial community. Commonalities and differences in approaches to AI in medicine are beginning to emerge.
NATURE MACHINE INTELLIGENCE
(2021)
Editorial Material
Biochemistry & Molecular Biology
Eric Wu, Kevin Wu, Roxana Daneshjou, David Ouyang, Daniel E. Ho, James Zou
Summary: A comprehensive overview of medical AI devices approved by the US Food and Drug Administration sheds light on limitations of the evaluation process that may mask vulnerabilities of devices when deployed on patients.
