Currently, Medical errors are a serious problem when examining patients. Creating information systems that use the capabilities of evidence-based medicine and artificial intelligence methods will allow the doctor to make an informed and proven decision. In this article, the authors offer a description of an information system that solves the problem of supporting medical decision making based on evidence-based medicine. This is achieved by using artificial intelligence methods. This work was supported by a grant from the Ministry of Education and Science of the Russian Federation, a unique project identifier RFMEFI60819X0278.
The article describes a method for assessing the malignancy potential of thyroid nodules and their stratification according to the European Thyroid Imaging And Reporting Data System (Eu-TIRADS) scale based on ultrasound diagnostic images using an artificial intelligence system. The method is based on the use of transfer learning technology for multi-parameter models of convolutional neural networks and their subsequent fine tuning. It was shown that even on a small dataset consisting of 1129 thyroid ultrasound images classified by 5 Eu-TIRADS categories, the application of the method provides high training accuracy (Accuracy: 0.8, AUC: 0.92). This makes it possible to introduce and use this technology in clinical practice as an additional tool (‘second opinion’) for an objective assessment of the risk of malignancy in thyroid nodules for the purpose of their further selection for fine needle biopsy.
Recent developments in Digital Medicine approaches concern pharmaceutical product optimization. Artificial Intelligence (AI) has multiple applications for pharmaceutical products’ lifecycle, increasing development speed, quality of the products, and efficiency of the therapy. Here, we systematically review the overall approach for AI implementation in pharmaceutical products’ lifecycle. The published studies in PubMed and IEEE Xplore were searched from inception to March 2022. The papers were screened for relevant outcomes, publication types, and data sufficiency, and a total of 73 (1.2%) out of 6131 studies were retrieved after the selection. We extracted the data according to the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) statement. All Artificial Intelligence systems could be divided into multiple overlapping categories by implementation. For the 177 projects found, the most popular areas of AI implementation are clinical trials and pre-clinical tests (34%). In second place are novel small molecule design systems, with 33% of the total. The third most popular scope for AI implementation is target identification for novel medicines. More than 25% of the systems provide this functionality. It is interesting that most of the systems specialize in only one area (102 systems—57%). None of the systems provide functionality for full coverage of the lifecycle and function in all categories of the tasks. This meta-analysis demonstrated that Artificial Intelligence solutions in pharmaceutical products’ lifecycle could find numerous implementations, and none of the available market solutions covers them all.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.