Background: The entire pharynx should be observed endoscopically to avoid missing pharyngeal lesions. An artificial intelligence (AI) model recognizing anatomical locations can help identify blind spots. We developed and evaluated an AI model classifying pharyngeal and laryngeal endoscopic locations.
Methods: The AI model was trained using 5382 endoscopic images, categorized into 15 anatomical locations, and evaluated using an independent dataset of 1110 images. The main outcomes were model accuracy, precision, recall, and F1-score. Moreover, we investigated focused regions in the input images contributing to the model predictions using gradient-weighted class activation mapping (Grad-CAM) and Guided Grad-CAM.
Results: Our AI model correctly classified pharyngeal and laryngeal images into 15 anatomical locations, with an accuracy of 93.3%. The weighted averages of precision, recall, and F1-score were 0.934, 0.933, and 0.933, respectively.
Conclusion: Our AI model has an excellent performance determining pharyngeal and laryngeal anatomical locations, helping endoscopists notify of blind spots.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1002/hed.27370 | DOI Listing |
World J Gastroenterol
December 2024
School of Medicine, National Yang Ming Chiao Tung University, Taipei 112, Taiwan.
Background: Diagnosing laryngopharyngeal reflux (LPR) is challenging due to overlapping symptoms. While proton pump inhibitors (PPIs) are commonly prescribed, reliable predictors of their responsiveness are unclear. Reflux monitoring technologies like dual potential of hydrogen (pH) sensors and multichannel intraluminal impedance-pH (MII-pH) could improve diagnosis.
View Article and Find Full Text PDFInt J Lang Commun Disord
December 2024
Hearing, Speech & Language Center, Sheba Medical Center, Tel Hashomer, Israel.
Background: Head and neck cancer (HNC) is amongst the 10 most common cancers worldwide and has a major effect on patients' quality of life. Given the complexity of this unique group of patients, a multidisciplinary team approach is preferable. Amongst the debilitating sequels of HNC and/or its treatment, swallowing, speech and voice impairments are prevalent and require the involvement of speech-language pathologists (SLPs).
View Article and Find Full Text PDFBMC Surg
December 2024
Department of Otolaryngology Head and Neck Surgery, The Affiliated Hospital of Southwest Medical University, No.25, Taiping Street, Jiangyang District, Luzhou, 646000, China.
Objective: To evaluate the efficacy of pedicled supraclavicular flaps in hypopharyngectomy reconstruction, with a focus on preserving laryngeal function.
Methods: From August 2019 to June 2022, 14 patients with primary hypopharyngeal carcinoma who met the inclusion and exclusion criteria and underwent the repair of hypopharyngeal defects using pedicled supraclavicular flaps were included retrospectively. Relevant clinical evaluation indicators include patient characteristics, defect sizes, flap sizes, flap harvesting time, postoperative hospital stay, postoperative complications, recurrence, and survival outcomes.
Laryngoscope
December 2024
Division of Otolaryngology - Head and Neck Surgery, Phoenix Children's Hospital, Phoenix, Arizona, U.S.A.
Objectives: To investigate the relationship between social determinants of health and timeliness of management, adherence to follow-up, and outcomes of treatment with interarytenoid injection augmentation (IAIA).
Methods: Retrospective cohort study of all pediatric patients treated with IAIA at a large pediatric institution's multidisciplinary aerodigestive clinic between August 2022 and February 2024. Retrieved demographic factors, dates of referral, consultation, treatment, and follow-up, as well as objective measures of dysphagia and aspiration via videofluoroscopic swallow study (VFSS) reports, using dysphagia outcome and severity scale (DOSS) scores and the greatest unsafe thickness, or 'aspiration score'.
Laryngoscope Investig Otolaryngol
December 2024
Department of Surgery, Otolaryngology/Head and Neck Surgery Section Virginia Tech Carilion School of Medicine Roanoke Virginia USA.
Objectives: Present a clinically challenging case of an immunocompetent 74-year-old male who presented with marked dyspnea and hemoptysis. After the airway was secured, direct laryngoscopy revealed a large, fungating, hemorrhagic mass of the left lateral pharyngeal wall and surrounding structures.
Methods: Chart review of a single patient.
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!