Exploring AI-Assisted Ultrasound for Carotid Plaque Detection by GPs in China
In recent advancements within the field of primary care, a groundbreaking study published in the Annals of Family Medicine explores the use of AI-assisted portable ultrasound technology by general practitioners (GPs) in China to detect carotid artery plaque, a key indicator of cardiovascular disease. The research, conducted by a team of experts led by Dr. Xiaochuan Liu, indicates promising results but also raises questions regarding the consistency of the diagnostic performance among the participating GPs.
Carotid artery plaque refers to the accumulation of fatty deposits within the arterial walls in the neck, which can lead to serious cardiovascular complications. To address this health concern, the emerging technology of point-of-care ultrasound (POCUS) has been adopted in primary care settings, particularly enhanced by artificial intelligence (AI) which improves both the quality of imaging and diagnostic accuracy. The objective of this study was to assess whether AI-assisted POCUS could enable GPs to effectively screen for carotid artery plaque within their local clinics after undergoing a systematic training program.
The study took place in Shanghai, where seven GPs were trained through a blended approach of theory and practical sessions focused on utilizing AI-assisted POCUS. This program enabled the GPs to conduct screenings on 169 high-risk patients, proactively recruiting them from four community health centers during regular outpatient visits. Patients were offered complimentary screening for carotid plaque, which was performed both by the GPs and a senior ultrasound specialist to provide a benchmark for accuracy.
The AI software integrated into the ultrasound device analyzes the imaging data in real time, signaling the operator with markers on the screen to indicate potential plaque locations. Following the scans, two senior specialists reviewed all recordings and ultrasound images to confirm the presence of carotid plaque, facilitating a robust metric against which the GPs' diagnostic findings were measured.
The results of the study revealed that the GPs successfully identified approximately 87% of patients with carotid plaque while accurately excluding about 91% of those without it. These outcomes demonstrated a high degree of agreement with the benchmark set by the ultrasound specialists. However, the study did identify that a significant percentage of missed cases were located at the bifurcation, where the carotid artery branches, indicating that the limitations of the AI system and the GPs' integration of AI feedback could be contributing factors to this discrepancy.
Despite the overall accuracy, the researchers noted that each screening session took nearly 8 minutes, which may present a challenge in standard GP visits where time is often limited. They propose that the methodology introduced in this study may be better suited for specialized environments like stroke screening clinics or community health initiatives rather than typical outpatient settings.
This innovative study marks the first of its kind in China to evaluate a structured POCUS training program for GPs within a clinical context, coupled with AI capabilities. The researchers emphasize that for the widespread implementation of AI-assisted POCUS protocols, there is an urgent need for standardized, tailored training modules to be developed as well as further validation across diverse cohorts of general practitioners to secure consistent and reliable diagnostic performance.
As the healthcare landscape continues to evolve with technological advancements, the integration of AI in primary care practices holds significant potential for improving patient outcomes, particularly in cardiovascular health. The full study can be accessed in the Annals of Family Medicine, which continues to serve as a vital resource for evidence-based information affecting the family medicine disciplines.
Carotid artery plaque refers to the accumulation of fatty deposits within the arterial walls in the neck, which can lead to serious cardiovascular complications. To address this health concern, the emerging technology of point-of-care ultrasound (POCUS) has been adopted in primary care settings, particularly enhanced by artificial intelligence (AI) which improves both the quality of imaging and diagnostic accuracy. The objective of this study was to assess whether AI-assisted POCUS could enable GPs to effectively screen for carotid artery plaque within their local clinics after undergoing a systematic training program.
The study took place in Shanghai, where seven GPs were trained through a blended approach of theory and practical sessions focused on utilizing AI-assisted POCUS. This program enabled the GPs to conduct screenings on 169 high-risk patients, proactively recruiting them from four community health centers during regular outpatient visits. Patients were offered complimentary screening for carotid plaque, which was performed both by the GPs and a senior ultrasound specialist to provide a benchmark for accuracy.
The AI software integrated into the ultrasound device analyzes the imaging data in real time, signaling the operator with markers on the screen to indicate potential plaque locations. Following the scans, two senior specialists reviewed all recordings and ultrasound images to confirm the presence of carotid plaque, facilitating a robust metric against which the GPs' diagnostic findings were measured.
The results of the study revealed that the GPs successfully identified approximately 87% of patients with carotid plaque while accurately excluding about 91% of those without it. These outcomes demonstrated a high degree of agreement with the benchmark set by the ultrasound specialists. However, the study did identify that a significant percentage of missed cases were located at the bifurcation, where the carotid artery branches, indicating that the limitations of the AI system and the GPs' integration of AI feedback could be contributing factors to this discrepancy.
Despite the overall accuracy, the researchers noted that each screening session took nearly 8 minutes, which may present a challenge in standard GP visits where time is often limited. They propose that the methodology introduced in this study may be better suited for specialized environments like stroke screening clinics or community health initiatives rather than typical outpatient settings.
This innovative study marks the first of its kind in China to evaluate a structured POCUS training program for GPs within a clinical context, coupled with AI capabilities. The researchers emphasize that for the widespread implementation of AI-assisted POCUS protocols, there is an urgent need for standardized, tailored training modules to be developed as well as further validation across diverse cohorts of general practitioners to secure consistent and reliable diagnostic performance.
As the healthcare landscape continues to evolve with technological advancements, the integration of AI in primary care practices holds significant potential for improving patient outcomes, particularly in cardiovascular health. The full study can be accessed in the Annals of Family Medicine, which continues to serve as a vital resource for evidence-based information affecting the family medicine disciplines.
Topics Health）
【About Using Articles】
You can freely use the title and article content by linking to the page where the article is posted.
※ Images cannot be used.
【About Links】
Links are free to use.