Diagnostic Accuracy for Breathing Sounds in Using an AI-Assisted Stethoscope Versus a Conventional Stethoscope

Area of Science:

• Medical Devices
• Artificial Intelligence in Healthcare
• Respiratory Medicine

Background:

• Accurate diagnosis of breathing sounds is crucial for patient safety, especially during sedation.
• Traditional stethoscopes have limitations in diagnostic accuracy for dental students.
• Artificial intelligence (AI) offers potential to enhance auscultation skills.

Purpose of the Study:

• To compare the diagnostic accuracy of dental students using traditional versus AI-assisted precordial stethoscopes.
• To evaluate the impact of AI on learning outcomes, confidence, and satisfaction.
• To assess the efficiency of AI tools in identifying specific respiratory sounds.

Main Methods:

• 45 dental students were randomized into groups using conventional stethoscopes or AI-assisted precordial stethoscopes (AIPS) with audio/video.
• Participants completed pre-tests, listened to breathing sound recordings, and took post-tests.
• Statistical analysis included ANOVA and paired t-tests to evaluate diagnostic accuracy, knowledge, confidence, and satisfaction.

Main Results:

• All groups showed improved diagnostic accuracy and knowledge, with AI groups performing better on specific sounds like apnea.
• Student confidence and satisfaction strongly correlated with diagnostic accuracy, with higher satisfaction in AI groups.
• AI groups showed faster response times for apnea, while conventional stethoscopes were faster for normal breathing.

Conclusions:

• AI-assisted precordial stethoscopes enhance diagnostic accuracy and learning for dental students, especially those with lower baseline knowledge.
• Integrating AI tools can improve the detection of respiratory complications during dental procedures.
• AI-assisted stethoscopes show promise for improving patient safety in sedation settings.