Key result
A novel deep learning-based system for ultrasound carotid stenosis measurement accurately stratified stroke risk, achieving area-under-curve values of 0.90, 0.94, and 0.86 compared to manual radiologist tracings.
Why the study?
Current conventional image-based methods for non-invasive carotid stenosis measurement are not accurate or reliable, lacking shape and intelligent learning components.
Does a deep learning-based methodology accurately measure carotid stenosis and stratify risk in ultrasound scans of diabetic patients?
Cross-Sectional (n=204)
No
Does a deep learning-based methodology accurately measure carotid stenosis and stratify risk in ultrasound scans of diabetic patients?
A deep learning-based system demonstrated high accuracy (AUC up to 0.94) for measuring carotid stenosis and stratifying risk in ultrasound scans of diabetic patients.
DL-based CCA stenosis measurement may improve accuracy over conventional methods; leaves open prospective validation before clinical adoption.
BACKGROUND: Stroke is in the top three leading causes of death worldwide. Non-invasive monitoring of stroke can be accomplished via stenosis measurements. The current conventional image-based methods for these measurements are not accurate and reliable. They do not incorporate shape and intelligent learning component in their design. METHODS: In this study, we propose a deep learning (DL)-based methodology for accurate measurement of stenosis in common carotid artery (CCA) ultrasound (US) scans using a class of AtheroEdge system from AtheroPoint, USA. Three radiologists manually traced the lumen-intima (LI) for the near and the far walls, respectively, which served as a gold standard (GS) for training the DL-based model. Three DL-based systems were developed based on three types of GS. RESULTS: IRB approved (Toho University, Japan) 407 US scans from 204 patients were collected. The risk was characterized into three classes: low, moderate, and high-risk. The area-under-curve (AUC) corresponding to three DL systems using receiver operating characteristic (ROC) analysis computed were: 0.90, 0.94 and 0.86, respectively. CONCLUSIONS: Novel DL-based strategy showed reliable, accurate and stable stenosis severity index (SSI) measurements.
No takes yet. Share an insight, caveat, or question.
Saba et al. (2019) conducted a cross-sectional in Subclinical atherosclerosis and moderate carotid stenosis (n=204). Deep learning (DL)-based methodology for carotid stenosis measurement vs. Manual lumen border delineation by radiologists (Gold Standard) was evaluated on Area-under-curve (AUC) for risk stratification (AUC 0.90, 0.94, and 0.86). A novel deep learning-based system for ultrasound carotid stenosis measurement accurately stratified stroke risk, achieving area-under-curve values of 0.90, 0.94, and 0.86 compared to manual radiologist tracings.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: