CBS 2019
CBSMD教育中心
中 文

血管内超声指导

Abstract

Recommended Article

Prospective application of pre-defined intravascular ultrasound criteria for assessment of intermediate left main coronary artery lesions results from the multicenter LITRO study Differential prognostic effect of intravascular ultrasound use according to implanted stent length Patterns of calcification in coronary artery disease. A statistical analysis of intravascular ultrasound and coronary angiography in 1155 lesions Serial intravascular ultrasound analysis of the main and side branches in bifurcation lesions treated with the T-stenting technique Intravascular ultrasound-guided percutaneous coronary intervention improves the clinical outcome in patients undergoing multiple overlapping drug-eluting stents implantation Long-term health outcome and mortality evaluation after invasive coronary treatment using drug eluting stents with or without the IVUS guidance. Randomized control trial. HOME DES IVUS Prognostic Value of Intravascular Ultrasound in Patients With Coronary Artery Disease Stent underexpansion and residual reference segment stenosis are related to stent thrombosis after sirolimus-eluting stent implantation: an intravascular ultrasound study

Original Research

JOURNAL:ACC Article Link

Artificial Intelligence in Interventional Cardiology

Bina Ahmed, MD, FACC

Pre-reading

The following are key points to remember from this state-of-the-art review on the impact of artificial intelligence (AI) on interventional cardiology:

  1. 1. AI encompasses a broad application of mathematical algorithms to train machines to mimic human behavior. There is increasing interest in developing AI technology for application in healthcare.

  2. 2. AI operations include machine learning (ML), deep learning (DL), natural language processing (NLP), cognitive computing, computer vision, and robotics.

  3. 3. ML is an automated system that learns to perform a task or make decisions from available data sources. Once an algorithm is programmed, ML has the ability to figure large complex and heterogeneous data sets and make predictions with fewer assumptions compared to conventional statistical methods.

  4. 4. DL is a part of ML, which is based in algorithms called neural networks. DL networks use digitized inputs that work through layers of connected neurons and perform advance pattern recognition to generate an output. DL does not require continued human input. DL is currently best applied to image recognition such as during angiography or echocardiography.

  5. 5. Virtual applications of AI have the potential to enhance image reconstruction, analysis, and interpretation. This is currently being used for coronary anatomic and functional lesion analysis.

  6. 6. Clinical decision support systems apply the use of ML, NLP, and pattern recognition to assist with imitating human thought processing. IBM is currently developing Medical Sieve, an automated cognitive assistant for cardiologists and radiologists to aid in clinical decision making.

  7. 7. Virtual reality platforms are currently being used for periprocedural planning of structural heart interventions.

  8. 8. Robotics are in their initial phase of application in interventional cardiology and not likely to replace a human interventional cardiologist in the near future. Although they can provide physical assistance, they do not perform intelligence assistance at this time.

  9. 9. Challenges to integration of AI in interventional cardiology practice include complexity of its integration, inability to ‘mimic’ human touch and emotions, and how it would impact the workforce.

  10. 10. AI is poised to transform and enhance the practice of interventional cardiology. Whether we can use it intelligently to enhance patient care and outcomes remains to be determined.