Skip directly to search Skip directly to A to Z list Skip directly to navigation Skip directly to page options Skip directly to site content

Main|Search|PHGKB
Search PHGKB:

Last Posted: Sep 23, 2022
spot light Highlights

Integrated multimodal artificial intelligence framework for healthcare applications
LR Soenksen et al, NPJ Digital Medicine, September 20, 2022

AI systems leveraging multiple data sources and input modalities are poised to become a viable method to deliver more accurate results and deployable pipelines across a wide range of applications. In this work, we propose and evaluate a unified Holistic AI in Medicine (HAIM) framework to facilitate the generation and testing of AI systems that leverage multimodal inputs. Our approach uses generalizable data pre-processing and machine learning modeling stages that can be readily adapted for research and deployment in healthcare environments.

Multimodal biomedical AI.
Acosta Julián N et al. Nature medicine 2022 9

The increasing availability of biomedical data from large biobanks, electronic health records, medical imaging, wearable and ambient biosensors, and the lower cost of genome and microbiome sequencing have set the stage for the development of multimodal artificial intelligence solutions that capture the complexity of human health and disease. In this Review, we outline the key applications enabled, along with the technical and analytical challenges.

NIH launches Bridge2AI program to expand the use of artificial intelligence in biomedical and behavioral research
NIH, September 13, 2022 Brand

AI is both a field of science and a set of technologies that enable computers to mimic how humans sense, learn, reason, and take action. Although AI is already used in biomedical research and healthcare, its widespread adoption has been limited in part due to challenges of applying AI technologies to diverse data types. This is because routinely collected biomedical and behavioral data sets are often insufficient, meaning they lack important contextual information about the data type, collection conditions, or other parameters. Without this information, AI technologies cannot accurately analyze and interpret data. AI technologies may also inadvertently incorporate bias or inequities unless careful attention is paid to the social and ethical contexts in which the data is collected.

Application of Artificial Intelligence Techniques to Predict Risk of Recurrence of Breast Cancer: A Systematic Review
C Mazo et al, J Per Med, September 13, 2022

This review found three areas that require further work. First, there is no agreement on artificial intelligence methodologies, feature predictors, or assessment metrics. Second, issues such as sampling strategies, missing data, and class imbalance problems are rarely addressed or discussed. Third, representative datasets for breast cancer recurrence are scarce, which hinders model validation and deployment. We conclude that predicting breast cancer recurrence remains an open problem despite the use of artificial intelligence.


Disclaimer: Articles listed in the Public Health Genomics and Precision Health Knowledge Base are selected by the CDC Office of Public Health Genomics to provide current awareness of the literature and news. Inclusion in the update does not necessarily represent the views of the Centers for Disease Control and Prevention nor does it imply endorsement of the article's methods or findings. CDC and DHHS assume no responsibility for the factual accuracy of the items presented. The selection, omission, or content of items does not imply any endorsement or other position taken by CDC or DHHS. Opinion, findings and conclusions expressed by the original authors of items included in the update, or persons quoted therein, are strictly their own and are in no way meant to represent the opinion or views of CDC or DHHS. References to publications, news sources, and non-CDC Websites are provided solely for informational purposes and do not imply endorsement by CDC or DHHS.

TOP