We use cookies to understand how you use our site and to improve your experience. This includes personalizing content and advertising. To learn more, click here. By continuing to use our site, you accept our use of cookies. Cookie Policy.

Features Partner Sites Information LinkXpress hp
Sign In
Advertise with Us
ARAB HEALTH - INFORMA

Download Mobile App




Advanced AI Technology Improves Accuracy and Efficiency of Nasal Endoscopy

By HospiMedica International staff writers
Posted on 04 Jul 2024
Print article
Image: The application of convolutional neural networks can improve the accuracy and efficiency of nasal endoscopy (Photo courtesy of Shutterstock)
Image: The application of convolutional neural networks can improve the accuracy and efficiency of nasal endoscopy (Photo courtesy of Shutterstock)

Nasal endoscopy (NE) is a critical diagnostic tool in rhinology, though its effectiveness is often limited by the nasal cavity's complex structure. Now, an insightful article published in the International Forum of Allergy & Rhinology details a study exploring how convolutional neural networks (CNNs) can enhance the precision and efficiency of NE. The research addresses the challenges inherent in navigating the complex anatomy of the nasal cavity for diagnostic purposes.

Conducted by a team from Ochsner Health (New Orleans, LA, USA), the study focused on a CNN-based model tailored to precisely identify and outline key landmarks in NE imagery. The images utilized were collected from NE procedures carried out at a medical center over the period from 2014 to 2023, employing a standard digital endoscope. A total of 2,111 images were manually segmented by three physicians. The researchers adapted the YOLOv8 object detection model to classify whether a turbinate was present, identify its location, and apply a segmentation mask to outline its boundaries. The model was refined through transfer learning techniques involving backpropagation and stochastic gradient descent. By carefully adjusting hyperparameters and halting training after a 15-epoch lack of improvement in validation performance, the model demonstrated notable success.

The model was able to detect the inferior turbinate (IT) and middle turbinate (MT) with an average accuracy of 91.5%, precision of 92.5%, and recall rate of 93.8%. With a confidence threshold set at 60%, the model achieved an average F1-score of 93.1%. The effective application of the YOLOv8 model marks a significant progression in the field of rhinology. Its capacity to accurately locate and delineate the IT and MT can significantly support clinicians in the diagnosis and treatment of sinonasal conditions. This advancement is especially beneficial for trainees and non-specialists, who may struggle with the intricate anatomy of the nasal cavity.

"This study showcases the potential of CNNs to enhance nasal endoscopy's accuracy and efficiency," said senior otolaryngologist Dr. Edward D. McCoul, who guided the research team. "By leveraging advanced AI technologies, we can markedly improve our diagnostic capabilities and provide superior patient care for those with sinonasal conditions."

Related Links:
Ochsner Health

Gold Member
12-Channel ECG
CM1200B
New
Gold Member
X-Ray QA Meter
T3 AD Pro
New
Doppler String Phantom
CIRS Model 043A
New
Vertebral Body Replacement System
Hydrolift

Print article

Channels

Critical Care

view channel
Image: The AI-powered algorithm offers quick, no-contact screenings for high blood pressure and diabetes (Photo courtesy of 123RF)

AI-Powered Algorithm Offers Quick, Contactless Blood Pressure and Diabetes Screening

A newly developed system that combines high-speed video with an artificial intelligence (AI)-powered algorithm may provide a quick, non-contact method for screening high blood pressure and Type 1 or Type... Read more

Patient Care

view channel
Image: The portable biosensor platform uses printed electrochemical sensors for the rapid, selective detection of Staphylococcus aureus (Photo courtesy of AIMPLAS)

Portable Biosensor Platform to Reduce Hospital-Acquired Infections

Approximately 4 million patients in the European Union acquire healthcare-associated infections (HAIs) or nosocomial infections each year, with around 37,000 deaths directly resulting from these infections,... Read more

Health IT

view channel
Image: First ever institution-specific model provides significant performance advantage over current population-derived models (Photo courtesy of Mount Sinai)

Machine Learning Model Improves Mortality Risk Prediction for Cardiac Surgery Patients

Machine learning algorithms have been deployed to create predictive models in various medical fields, with some demonstrating improved outcomes compared to their standard-of-care counterparts.... Read more

Point of Care

view channel
Image: The acoustic pipette uses sound waves to test for biomarkers in blood (Photo courtesy of Patrick Campbell/CU Boulder)

Handheld, Sound-Based Diagnostic System Delivers Bedside Blood Test Results in An Hour

Patients who go to a doctor for a blood test often have to contend with a needle and syringe, followed by a long wait—sometimes hours or even days—for lab results. Scientists have been working hard to... Read more
Copyright © 2000-2024 Globetech Media. All rights reserved.