Key result
Natural language processing remains disproportionately underutilized in vascular surgery AI, featuring in ~10% of studies.
Why the study?
AI applications are expanding in vascular surgery, but data sources for algorithm development—especially natural language processing—and their clinical applications and target journals needed scoping.
Systematic Review (n=342)
Despite the abundance of textual data in electronic health records, natural language processing remains disproportionately underutilized in vascular surgery AI research compared to image-based deep learning.
Answers are grounded in this paper and cite their sources.
NLP underuse misses EHR text for vascular insights; challenges DL/ML dominance and extends need for balanced AI modalities.
Background Applications of Artificial Intelligence (AI) are gaining traction in healthcare, including vascular surgery. While most healthcare data is in the form of narrative text or audio recordings, natural language processing (NLP) offers the ability to extract structured information from unstructured medical records. This can be used to develop more accurate risk prediction models, and to identify patients for trials to improve recruitment. The goals of this scoping review were to determine the source of data used to develop AI-based algorithms with emphasis on natural language processing, establish their application in different areas of vascular surgery and identify a target audience of published journals. Methods A literature search was carried out using PubMed, EMBASE, and Google Scholar database from January 1996 to March 2023. Following screening, 342 peer-reviewed articles met the eligibility criteria. Findings Among the 342 articles meeting eligibility criteria, the majority (191) were published after 2020. NLP algorithms were described in 34 papers, while 115 and 193 papers focused on machine learning (ML) and deep learning (DL), respectively. Two-thirds (67.25%) were published in non-clinical journals. The AI-based algorithms found widest application in research related to aorta and its branches (126 articles), followed by carotid disease (85), and peripheral arterial disease (65). Image-based data were utilised in 216 articles, while 153 and 85 papers relied on medical records, and clinical parameters. The AI algorithms were used for predictive modelling (123 papers), medical image segmentation (118), and to aid identification, detection, and diagnosis (103). Only 18 publications aimed to enhance the understanding of vascular disease pathophysiology. Interpretation Utilisation of different data sources and AI technologies depends on the purpose of the undertaken research. Despite the abundance of available of textual data, the NLP is disproportionally underutilised AI sub-domain.
No takes yet. Share an insight, caveat, or question.
Powezka et al. (2023) conducted a systematic review in Vascular surgery conditions (aortic, carotid, peripheral arterial disease, etc.) (n=342). Artificial Intelligence (NLP, ML, DL) was evaluated on Source of data and type of AI technology utilised. Despite the abundance of available textual data, natural language processing is a disproportionally underutilised AI sub-domain in vascular surgery, featuring in only 34 of 342 reviewed articles compared to 115 for machine learning and 193 for deep learning.
Synapse has enriched 10 related papers. Consider them for comparative context: