Core Concepts
Artificial intelligence (AI) has the potential to revolutionize medical care, but its integration into clinical practice faces significant challenges that require further research and collaboration.
Abstract
The article discusses the current state of AI in the medical field and the potential benefits and challenges of integrating this technology into clinical practice.
Key highlights:
AI has already matched or outperformed human experts in various patient care-related tasks, such as skin cancer classification, sepsis treatment, and medical imaging diagnostics.
However, most of these AI models have been tested retrospectively outside real-world contexts, and there is a lack of randomized controlled medical studies to support the hundreds of AI-enabled medical devices that have been approved by regulatory bodies.
A scoping review analyzed 86 randomized studies that focused on the use of AI in clinical practice, with the majority in the fields of gastroenterology, radiology, surgery, and cardiology.
The review found that AI systems have proved able to optimize insulin dosage, monitor hypotension, reduce acute care and prostate tumor volume, predict the risk of diabetic retinopathy, and facilitate the identification of patients with atrial fibrillation at high risk of stroke.
However, the review also identified areas that require further research, such as the need for increased international collaboration and multicenter trials to ensure the generalizability of AI systems across different populations and healthcare systems.
The prevalence of gastroenterology studies based on videos suggests that clinical AI research is still homogeneous in terms of researchers, study designs, and outcome measures, and further research is needed to evaluate the effect of AI systems that incorporate clinical context or patient history in their decision-making process.
Stats
AI systems have proved able to optimize insulin dosage and monitor hypotension, thus improving the average time patients spend within target ranges for blood glucose and blood pressure.
AI has reduced the rates of acute care and prostate tumor volume in radiation therapy and prostate brachytherapy applications.
AI can immediately predict the risk for diabetic retinopathy, thus increasing patients' adherence to advice compared with those awaiting a doctor's evaluation.
AI can reduce postoperative pain scores through a nociception monitoring system compared with an unassisted medical intervention on the patient.
AI can provide cancer mortality predictions that increase disease discussions between oncologists and patients.
AI facilitates the identification of patients with atrial fibrillation who are at high risk for stroke and allows the doctor to avoid increasing new anticoagulant prescriptions.
Quotes
"Deep learning models have matched human experts' performance in skin cancer classification, timely identification and adjustment of treatment strategies for septic patients, and medical imaging diagnostic procedures."
"However, most of these models have been retrospectively tested outside real-world contexts, and there is a lack of randomized controlled medical studies to support the hundreds of AI-enabled medical devices that have been approved by regulatory bodies."
"The prevalence of gastroenterology studies based on videos indicates that clinical AI research is still homogeneous in terms of researchers, study designs, and outcome measures."