RESUMO
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.
RESUMO
BACKGROUND: Invasive fungal rhinosinusitis (IFS) with orbital complications has remained a challenging disease over the past few decades. Only a few studies have been conducted to investigate the factors associated with orbital complications in fungal rhinosinusitis (FRS). We aimed to review the characteristics between IFS and non-invasive fungal rhinosinusitis (NIFS) and determine clinical factors associated with orbital complications and overall survival. METHODS: A multi-institutional database review study was conducted using the Chang Gung Research Database (CGRD) from January 2001 to January 2019. We identified FRS patients using International Classification of Diseases diagnosis codes and SNOMED CT. We categorized patients into IFS and NIFS groups and analyzed the demographic data, underlying diseases, clinical symptoms, laboratory data, image findings, fungal infection status, and survival outcomes. RESULTS: We included 1624 patients in our study, with 59 IFS patients and 1565 NIFS patients. The history of an organ or hematopoietic cell transplantation had a significant prognostic effect on the survival outcomes, with surgical intervention and high hemoglobin (Hb) and albumin levels recognized as positive predictors. Posterior ethmoid sinus involvement, sphenoid sinus involvement, facial pain, blurred vision, and periorbital swelling were risk factors of orbital complications. CONCLUSIONS: In NIFS patients, orbital complications were found to be associated with old age, a high WBC count, high blood glucose, and a high CRP level. For the risk factors of orbital complications in IFS patients, posterior ethmoid sinus involvement, sphenoid sinus involvement, facial pain, blurred vision, and periorbital swelling were recognized as predictors. Among IFS patients, a history of organ or hematopoietic cell transplantation was a risk factor for poor survival, while, conversely, surgical intervention and high Hb and albumin levels were related to improved survival. As predictors of orbital complications in IFS patients, posterior ethmoid sinus involvement, sphenoid sinus involvement, facial pain, blurred vision, and periorbital swelling upon the first visit should raise attention, with close monitoring.
