The utilization of artificial intelligence is rapidly increasing across various metrics. A recent article by StudyFinds examined the perceptions of potential patients regarding AI’s impact on medical diagnosis. There is a notable sentiment longing for the era when the patient-doctor relationship was a cornerstone of effective healthcare. This prompts the question of AI’s current position within the medical field.
Uncertainties surround AI implementation in clinical settings
For instance, the Epic Sepsis Model, utilized in numerous U.S. hospitals, lacks sufficient testing despite its use for predicting sepsis, showing poor accuracy. Similarly, in radiology, limited studies explore the attitudes of medical professionals toward adopting AI innovations, reflecting uncertainties around AI implementation in clinical settings.
A comprehensive review published in The Lancet Digital Health in May 2024 examined the impact of AI in medicine prior to 2023. It assessed 86 randomized studies, predominantly in gastroenterology (43%), radiology (13%), surgery (6%), and cardiology (6%).
Interestingly, only 18 studies evaluated AI’s influence on quality-of-care management in clinical practice. The review aimed to understand AI’s potential to enhance care, incorporate patient behavior, and improve clinical decision-making efficiency while identifying areas requiring further research.
AI system positively impacted disease management
AI systems were found to have positive impacts on disease management. They optimized insulin dosage, leading to improved blood glucose levels within target ranges. Additionally, they enhanced blood pressure control, reduced prostate tumor volume during radiation therapy, and predicted diabetic retinopathy risk, prompting better adherence to medical advice. Moreover, AI reduced postoperative pain scores, increased discussions about cancer mortality between oncologists and patients, and helped identify atrial fibrillation patients at high risk for stroke.
AI research in healthcare has predominantly focused on gastroenterology, radiology, surgery, and cardiology, with limited attention given to primary care. This gap underscores the necessity for further investigation in primary care, which forms the basis of all medical specialties. The majority of studies have been conducted within individual countries, particularly the United States and China, indicating a need for increased international collaboration and multicenter trials. Such endeavors could enhance the applicability of AI systems across diverse populations and healthcare settings.
