Artificial intelligence to detect tympanic membrane perforations.

OBJECTIVE: To explore the feasibility of constructing a proof-of-concept artificial intelligence algorithm to detect tympanic membrane perforations, for future application in under-resourced rural settings. METHODS: A retrospective review was conducted of otoscopic images analysed using transfer learning with Google's Inception-V3 convolutional neural network architecture. The 'gold standard' 'ground truth' was defined by otolaryngologists. Perforation size was categorised as less than one-third (small), one-third to two-thirds (medium), or more than two-thirds (large) of the total tympanic membrane diameter. RESULTS: A total of 233 tympanic membrane images were used (183 for training, 50 for testing). The algorithm correctly identified intact and perforated tympanic membranes (overall accuracy = 76.0 per cent, 95 per cent confidence interval = 62.1-86.0 per cent); the area under the curve was 0.867 (95 per cent confidence interval = 0.771-0.963). CONCLUSION: A proof-of-concept image-classification artificial intelligence algorithm can be used to detect tympanic membrane perforations and, with further development, may prove to be a valuable tool for ear disease screening. Future endeavours are warranted to develop a point-of-care tool for healthcare workers in areas distant from otolaryngology.

An artificial intelligence algorithm that identifies middle turbinate pneumatisation (concha bullosa) on sinus computed tomography scans.

OBJECTIVE: Convolutional neural networks are a subclass of deep learning or artificial intelligence that are predominantly used for image analysis and classification. This proof-of-concept study attempts to train a convolutional neural network algorithm that can reliably determine if the middle turbinate is pneumatised (concha bullosa) on coronal sinus computed tomography images. METHOD: Consecutive high-resolution computed tomography scans of the paranasal sinuses were retrospectively collected between January 2016 and December 2018 at a tertiary rhinology hospital in Australia. The classification layer of Inception-V3 was retrained in Python using a transfer learning method to interpret the computed tomography images. Segmentation analysis was also performed in an attempt to increase diagnostic accuracy. RESULTS: The trained convolutional neural network was found to have diagnostic accuracy of 81 per cent (95 per cent confidence interval: 73.0-89.0 per cent) with an area under the curve of 0.93. CONCLUSION: A trained convolutional neural network algorithm appears to successfully identify pneumatisation of the middle turbinate with high accuracy. Further studies can be pursued to test its ability in other clinically important anatomical variants in otolaryngology and rhinology.

An artificial intelligence algorithm that differentiates anterior ethmoidal artery location on sinus computed tomography scans.

OBJECTIVE: Deep learning using convolutional neural networks represents a form of artificial intelligence where computers recognise patterns and make predictions based upon provided datasets. This study aimed to determine if a convolutional neural network could be trained to differentiate the location of the anterior ethmoidal artery as either adhered to the skull base or within a bone 'mesentery' on sinus computed tomography scans. METHODS: Coronal sinus computed tomography scans were reviewed by two otolaryngology residents for anterior ethmoidal artery location and used as data for the Google Inception-V3 convolutional neural network base. The classification layer of Inception-V3 was retrained in Python (programming language software) using a transfer learning method to interpret the computed tomography images. RESULTS: A total of 675 images from 388 patients were used to train the convolutional neural network. A further 197 unique images were used to test the algorithm; this yielded a total accuracy of 82.7 per cent (95 per cent confidence interval = 77.7-87.8), kappa statistic of 0.62 and area under the curve of 0.86. CONCLUSION: Convolutional neural networks demonstrate promise in identifying clinically important structures in functional endoscopic sinus surgery, such as anterior ethmoidal artery location on pre-operative sinus computed tomography.

Periostin as a marker of mucosal remodelling in chronic rhinosinusitis.

BACKGROUND: Although extracellular matrix (ECM) proteins are associated with irreversible lower airway changes, the relationship with upper airway remodelling which occurs during chronic rhinosinusitis (CRS) is poorly understood. This study assessed the expression of ECM proteins periostin, fibulin-1, fibronectin and collagenIV in nasal mucosa of patients with and without histologic features of remodelling. METHODS: A cross-sectional study of sinonasal mucosal biopsies taken from patients, undergoing surgery for CRS was performed, where patients were grouped according to remodelling, defined by basement membrane thickening (BMT over 7.5 micrometer) and subepithelial fibrosis. An overall view and three random fields of immunostained tissue sections that included epithelium, basement membrane and submucosa, were imaged using Zeiss Zen software. The area and intensity of positive staining were scored by two blinded observers, using a 12-point ordinal scale of weak to strong. RESULTS: 65 patients (47.6 +/- 13.4years, 44.6% female) were assessed. Patients were grouped as controls 26.2%, BMT/no fibrosis 38.5% or BMT and fibrosis 33.8%. Stronger grade of periostin expression was associated with remodelling changes and tissue eosinophilia over 10/HPF. Fibulin-1, fibronectin and collagenIV did not differ. CONCLUSION: Periostin expression was associated with the presence of BMT, fibrosis and tissue eosinophilia and may identify patients undergoing remodelling changes.

The effect of two sphenoidotomy approaches on the sense of smell: trans-ethmoidal versus trans-sphenoethmoidal.

BACKGROUND: There are generally two methods to access the sphenoid sinus: either through the natural ostium {trans-sphenoethmoidalor via sphenoethmoidal recess), or by creating a second opening through the posterior ethmoids (trans-ethmoidal).This study psychophysically and subjectively evaluates the effect of the trans-sphenoethmoidal technique to the trans-ethmoidal technique for sphenoid sinusotomy on olfactory function. METHODS: Prospective cohort analysis of 48 patients with comparable sinus disease underwent primary sphenoidotomy via transsphenoethmoidal(n = 24) versus trans-ethmiodal (n = 24) technique between September 2011 and February 2012. The patients had their olfaction measured psychophysically with "Sniffin' Sticks" and subjectively with a visual analogue scale (VAS) pre-operatively and at 5 weeks post-operatively. RESULTS: Psychophysical scores from the Sniffin' sticks provide a Threshold, Discrimination and Identification (TDI) score out of 48.The TDI change (post-operative TDI score minus pre-operative score) as well as VAS change (post-operative VAS minus pre-operativeVAS) were analyzed using t-test analysis, which showed no significant difference between the two measurements. CONCLUSION: If the trans-sphenoethmoidal technique is done meticulously, patients have the same olfactory relief, psychophysically and subjectively, as those undergoing the trans-ethmoidal technique.

Combined topical flucytosine and amphotericin B for refractory vaginal Candida glabrata infections.

Patients with vaginitis due to highly azole resistant Candida glabrata can be particularly difficult to treat. We describe three cases of longstanding vaginal candidiasis due to C glabrata. These had failed to respond to local and systemic antifungals. Flucytosine (1 g) and amphotericin B (100 mg) formulated in lubricating jelly base in a total 8 g delivered dose, was used per vagina once daily for 14 days with significant improvement, both clinically and microbiologically.