RÉSUMÉ
OBJECTIVE: We aimed to develop a deep learning artificial intelligence (AI) algorithm to detect impacted animal bones on lateral neck radiographs and to assess its effectiveness for improving the interpretation of lateral neck radiographs. METHODS: Lateral neck radiographs were retrospectively collected for patients with animal bone impaction between January 2010 and March 2020. Radiographs were then separated into training, validation, and testing sets. A total of 1733 lateral neck radiographs were used to develop the deep learning algorithm. The testing set was assessed for the stand-alone deep learning AI algorithm and for human readers (radiologists, radiology residents, emergency physicians, ENT physicians) with and without the aid of the AI algorithm. Another radiograph cohort, collected from April 1, 2020, to June 30, 2020, was analyzed to simulate clinical application by comparing the deep learning AI algorithm with radiologists' reports. RESULTS: In the testing set, the sensitivity, specificity, and accuracy of the AI model were 96%, 90%, and 93% respectively. Among the human readers, all physicians of different subspecialties achieved a higher accuracy with AI-assisted reading than without. In the simulation set, among the 20 cases positive for animal bones, the AI model accurately identified 3 more cases than the radiologists' reports. CONCLUSION: Our deep learning AI model demonstrated a higher sensitivity for detection of animal bone impaction on lateral neck radiographs without an increased false positive rate. The application of this model in a clinical setting may effectively reduce time to diagnosis, accelerate workflow, and decrease the use of CT.
RÉSUMÉ
Foreign body sensation of the pharynx is a common complaint in otolaryngologic practice. The definite cause of this symptom is seldom precisely diagnosed, leading to chronic pharyngitis. The purpose of the present study was to evaluate the relationship between lump in throat and cervical degenerative disease by their clinical symptoms and the associated image characteristics. All patients with or without the complaint of lump in throat who attended the otolaryngologic clinics of Chang Gung Memorial Hospital-Kaohsiung Medical Center from January 2009 to May 2010 were prospectively eligible for this study. A total of 225 patients who met the inclusion criteria were assigned to either the study group (150 patients with symptoms of lump in the throat) or the control group (75 individuals without symptoms). All patients received plain views of the cervical lateral neck. The study group underwent further diagnostic examinations, including esophagography to exclude other diseases. Symptom scores (range, 0-3) and imaging characteristics such as the number (total spurs) and location (C level) of cervical osteophytes, interspace narrowing, spondylolisthesis, and retrolisthesis were compared between the two groups. It was found that the study group had a significantly greater number of total spurs (P < 0.001), and C45, C56, and C67 were the predominant sites, with significant odds ratios of more than 2 at all these levels. Only C67 revealed a difference in interspace narrowing, and only C45 showed a difference in spondylolisthesis between groups. Hence, a high correlation was found between the lump-in-throat sensation and total number of spurs, especially at the levels of C45, C56, and C67. This result implied that cervical osteophytes might be associated with foreign body sensations of the pharynx.