Applied Artificial Intelligence in Healthcare: A Review of Computer Vision Technology Application in Hospital...

Area of Science:

• Artificial Intelligence
• Computer Vision
• Healthcare Technology

Background:

• Computer vision (CV), a subset of artificial intelligence (AI), extensively used in various industries, faces limited adoption in healthcare due to privacy, safety, and ethical concerns.
• Despite challenges, CV holds potential to revolutionize patient monitoring, enhance system efficiencies, and reduce healthcare professional workload.

Purpose of the Study:

• To review end-user applications of computer vision in healthcare, contrasting with previous reviews focusing on technology.
• To categorize CV applications across different industries and explore its developments in hospital, outpatient, and community settings.
• To identify future opportunities for CV integration in healthcare through journey mapping and discuss associated challenges.

Main Methods:

• Categorization of CV applications in non-healthcare industries (job enhancement, surveillance, automation, augmented reality).
• Review of CV advancements in diverse healthcare settings: hospital, outpatient, and community.
• Exploration of specific monitoring applications: delirium, pain, sedation, patient deterioration, mechanical ventilation, mobility, patient safety, and surgical procedures.
• Journey mapping to identify future application opportunities.
• Discussion of privacy, safety, and ethical considerations, and algorithm development processes.

Main Results:

• CV applications are categorized across various industries, providing a comparative framework.
• Recent advances in CV for monitoring patient conditions (delirium, deterioration, pain, sedation), interventions (mechanical ventilation), mobility, and safety are highlighted.
• CV is being explored for surgical applications, workload quantification, and remote patient monitoring outside traditional settings.
• Journey mapping identified potential areas for expanded CV use.
• Privacy, safety, ethical concerns, and algorithm development processes were discussed as limitations to CV expansion.

Conclusions:

• Computer vision presents substantial opportunities for improving patient care, operational efficiency, and safety in healthcare settings.
• Overcoming privacy, safety, and ethical hurdles through robust algorithm development and testing is crucial for wider CV adoption.
• This review provides a comprehensive overview of current CV applications and a roadmap for its expanded integration into healthcare systems.