Suspected Acute Pulmonary Embolism: Gestalt, Scoring Systems, and Artificial Intelligence

Area of Science:

• Pulmonology
• Diagnostic Imaging
• Medical Informatics

Background:

• Pulmonary embolism (PE) presents with nonspecific signs and symptoms, necessitating further investigation.
• Clinical probability assessment is crucial for ruling out PE, with methods like Wells score and revised Geneva score being validated.
• Increased use of computed tomography pulmonary angiography (CTPA) has led to more testing without significant improvement in patient outcomes, highlighting a need for diagnostic strategy refinement.

Purpose of the Study:

• To review recent strategies for optimizing the diagnostic approach to pulmonary embolism.
• To address the challenge of limiting testing for suspected PE while maintaining diagnostic accuracy.
• To explore the role of digital tools and artificial intelligence in the clinical practice of PE diagnosis.

Main Methods:

• Review of current diagnostic strategies for pulmonary embolism.
• Analysis of clinical probability assessment methods (gestalt, Wells score, revised Geneva score).
• Discussion of emerging approaches including simplification of strategies and digital support tools.

Main Results:

• Clinical probability assessment guides the interpretation of diagnostic tests for PE.
• Widespread CTPA use has increased investigations but not patient outcomes, emphasizing the need to limit testing.
• New strategies focus on simplifying diagnostic pathways or employing digital tools for more sophisticated decision-making.

Conclusions:

• Effective PE diagnosis relies on accurate clinical probability assessment and judicious use of investigations.
• Limiting the number of tests for suspected PE is a key challenge, with AI and machine learning showing future promise.
• Implementation of refined diagnostic strategies, potentially aided by technology, is essential for improving PE care.