[Artificial intelligence in ex vivo confocal laser scanning microscopy].

Area of Science:

• Medical imaging analysis
• Machine learning in dermatology
• Digital pathology

Background:

• Visual data from methods like ex vivo confocal laser scanning microscopy (CLSM) are well-suited for machine learning.
• Digital hematoxylin-eosin-like staining enhances accessibility of ex vivo CLSM for digital assessments.

Purpose of the Study:

• To review existing literature on artificial intelligence (AI) applications in ex vivo CLSM.
• To identify potential challenges and limitations of using AI in this context.

Main Methods:

• Literature review and synthesis of current knowledge.
• Incorporation of personal experience with AI in ex vivo CLSM.

Main Results:

• AI algorithms can identify malignant skin lesions from clinical photographs and microscopic images with accuracy comparable to human experts.
• AI has identified novel visual patterns associated with dermatological diseases, some previously overlooked by experts.
• A single study has been published on AI for automated tumor detection using ex vivo CLSM, highlighting the novelty of this application.

Conclusions:

• Digitized ex vivo CLSM is an ideal platform for developing and implementing AI-driven healthcare applications.
• Further research is anticipated to overcome challenges and fully realize the potential of AI in ex vivo CLSM for diagnostics and disease understanding.