Background The overall prognosis of oral cancer remains poor because over half of patients are diagnosed at advanced-stages. Previously reported screening and earlier detection methods for oral cancer still largely rely on health workers’ clinical experience and as yet there is no established method. We aimed to develop a rapid, non-invasive, cost-effective, and easy-to-use deep learning approach for identifying oral cavity squamous cell carcinoma (OCSCC) patients using photographic images. Methods We developed an automated deep learning algorithm using cascaded convolutional neural networks to detect OCSCC from photographic images. We included all biopsy-proven OCSCC photographs and normal controls of 44,409 clinical images collected from 11 hospitals around China between April 12, 2006, and Nov 25, 2019. We trained the algorithm on a randomly selected part of this dataset (development dataset) and used the rest for testing (internal validation dataset). Additionally, we curated an external validation dataset comprising clinical photographs from six representative journals in the field of dentistry and oral surgery. We also compared the performance of the algorithm with that of seven oral cancer specialists on a clinical validation dataset. We used the pathological reports as gold standard for OCSCC identification. We evaluated the algorithm performance on the internal, external, and clinical validation datasets by calculating the area under the receiver operating characteristic curves (AUCs), accuracy, sensitivity, and specificity with two-sided 95% CIs. Findings 1469 intraoral photographic images were used to validate our approach. The deep learning algorithm achieved an AUC of 0·983 (95% CI 0·973–0·991), sensitivity of 94·9% (0·915–0·978), and specificity of 88·7% (0·845–0·926) on the internal validation dataset ( n = 401), and an AUC of 0·935 (0·910–0·957), sensitivity of 89·6% (0·847–0·942) and specificity of 80·6% (0·757–0·853) on the external validation dataset ( n = 402). For a secondary analysis on the internal validation dataset, the algorithm presented an AUC of 0·995 (0·988–0·999), sensitivity of 97·4% (0·932–1·000) and specificity of 93·5% (0·882–0·979) in detecting early-stage OCSCC. On the clinical validation dataset ( n = 666), our algorithm achieved comparable performance to that of the average oral cancer expert in terms of accuracy (92·3% [0·902–0·943] vs 92.4% [0·912–0·936]), sensitivity (91·0% [0·879–0·941] vs 91·7% [0·898–0·934]), and specificity (93·5% [0·909–0·960] vs 93·1% [0·914–0·948]). The algorithm also achieved significantly better performance than that of the average medical student (accuracy of 87·0% [0·855–0·885], sensitivity of 83·1% [0·807–0·854], and specificity of 90·7% [0·889–0·924]) and the average non-medical student (accuracy of 77·2% [0...

LLNs are rare in patients with SCC of the tongue and the floor of the mouth, and they would be ready to be omitted. The dissection of these LLNs would be of benefit to those patients with advanced pathological grade.

No abstract

Objective: One of the major impediments in tissue-engineered oral mucosa (TEOM) is the lack of rete ridge (RR) structures that can weaken the connection between the epidermis and dermis. This study aimed to investigate the native morphology of RRs as well as the expression of extracellular signal-regulated kinase 1/2 (ERK1/2), Ki67, and keratin-19, which are related to cell mechanotransduction, proliferation, and stemness in the oral epidermis, respectively. Methods: RR characteristics, including type, density, length, and width, were analyzed in the masticatory mucosa (Mm) and lining mucosa (Lm) sites of 52 specimens. The expression of ERK1/2, Ki67, and keratin-19 was assessed by immunohistochemistry. ERK1/2 activation by masticatory stimuli was confirmed in vitro by loading pressure onto cultured keratinocytes isolated from the specimens. Results: Three types of RR were found. The RRs in the Mm and Lm differed. The length and percentage of ERK1/2-positive (%ERK1/2+) basal layer cells had a negative correlation (p = 0.004), whereas the length and %Ki67+ basal layer cells had a positive correlation (p = 0.013). The %ERK1/2+ basal layer cells and %keratin-19+ basal layer cells had a negative relationship (p = 0.011). ERK1/2 activation in the oral epithelium was induced by pressure and propagated in cultured keratinocytes. Conclusion: RRs are longer in the Mm, which may result from the topical basal cell proliferation and migration induced by masticatory pressure via ERK1/2 activation. Our findings preliminarily interpret RR histomorphology as influenced by oral masticatory pressure. Results may benefit future studies on RR development and reconstruction in TEOM models to enhance the epidermis-dermis connection.

Rationale:Only 4.5% of brown tumors involve facial bones; of these, solitary bone involvement is usual. Brown tumors of multiple facial bones are extremely rare. Here, we report the case of a brown tumor of multiple facial bones initially misdiagnosed as an odontogenic cyst.Patient concerns:A pregnant 26-year-old woman was referred to our hospital with painful swelling of multiple facial bones, anemia, urinary calculi, marasmus, and a history of multiple bone fractures. Laboratory examination revealed an elevated serum calcium level of 3.09 mmol/L (normal range: 2.0–2.8 mmol/L) and a low phosphorus level of 0.62 mmol/L (normal range: 0.81–1.65 mmol/L). The serum alkaline phosphatase concentration was 397 IU/L (normal range: 24–82 IU/L) and parathyroid hormone level was 267 pg/mL (normal range: 14–72 pg/mL). Cone beam computed tomography revealed multiple ossifying fibromas of the maxilla and mandible. Incisional biopsy revealed abundant spindle cells with areas of hemorrhage and haphazardly arranged diffuse multinucleated giant cells.Diagnoses:The patient was diagnosed with primary hyperparathyroidism (HPT).Interventions:She was treated by parathyroidectomy.Outcomes:The multiple osteitis fibrosa cystica gradually resolved as bone re-mineralized. The patient has been followed up for 2 years without evidence of tumor recurrence.Lessons:As multiple osteolytic lesions of facial bones can be caused by primary HPT, serum calcium and parathyroid hormone assays should be performed routinely when investigating these lesions.