Deep convolutional neural network algorithm for the automatic segmentation of oral potentially malignant disorders and oral cancers.

This study aimed to develop an algorithm to automatically segment the oral potentially malignant diseases (OPMDs) and oral cancers (OCs) of all oral subsites with various deep convolutional neural network applications. A total of 510 intraoral images of OPMDs and OCs were collected over 3 years (2006-2009). All images were confirmed both with patient records and histopathological reports. Following the labeling of the lesions the dataset was arbitrarily split, using random sampling in Python as the study dataset, validation dataset, and test dataset. Pixels were classified as the OPMDs and OCs with the OPMD/OC label and the rest as the background. U-Net architecture was used and the model with the best validation loss was chosen for the testing among the trained 500 epochs. Dice similarity coefficient (DSC) score was noted. The intra-observer ICC was found to be 0.994 while the inter-observer reliability was 0.989. The calculated DSC and validation accuracy across all clinical images were 0.697 and 0.805, respectively. Our algorithm did not maintain an excellent DSC due to multiple reasons for the detection of both OC and OPMDs in oral cavity sites. A better standardization for both 2D and 3D imaging (such as patient positioning) and a bigger dataset are required to improve the quality of such studies. This is the first study which aimed to segment OPMDs and OCs in all subsites of oral cavity which is crucial not only for the early diagnosis but also for higher survival rates.

Craniofacial Fibrous Dysplasia Involvements of Mccune-Albright Syndrome: A Review with an Additional Case.

BACKGROUND: McCune-Albright Syndrome (MAS) is a genetic disorder with a triad of endocrine diseases, café-au-lait macules and fibrous dysplasias. Craniofacial fibrous dysplasia is a term that is used to describe the fibrous dysplasia, which was localized at the craniofacial skeleton and is common in MAS patients. OBJECTIVE: The objective of this review is to determine the involvement frequency of cranial and facial bones in patients with MAS and CFD. METHODS: Articles in PubMed was searched with the following details "(mccune[Title/Abstract] OR albright[Title/Abstract]) OR ("craniofacial fibrous dysplasia"[MeSH Terms] OR ("craniofacial"[ All Fields] AND "fibrous"[All Fields] AND "dysplasia"[All Fields]) OR "craniofacial fibrous dysplasia"[All Fields])". The articles in which the authors did not state the involved bones or did not add any radiographic images were excluded from the study. RESULTS: 26 cases in 25 articles met the inclusion criteria. Among the 26 cases and our case, sphenoid and frontal bones were involved in 17 cases, parietal and occipital bones were involved in 15 cases, mandible and ethmoid bone were involved in 14 cases, maxilla-zygoma-temporal and palate was involved in 13, 11, 6 and 3 cases, respectively. Palate was involved in cases where maxilla was also involved. Our case was the only case that was evaluated with CBCT. CONCLUSION: Routine follow-ups are important since new CFDs can occur in different cranial or facial bones. 2D imaging techniques may not be able to demonstrate early CFDs; thus, an advanced imaging technique should be used after MAS diagnosis.

A review with an additional case: amelanotic malignant melanoma at mandibular gingiva.

The purpose of this review with an additional case is to evaluate the clinical, ultrasonographic and histopathological features of a rare case of Amelanotic Malignant Melanoma (AMM) at mandibular gingiva and to compare our case with other published AMMs at mandibular gingiva. A 52-year-old male patient with no systemic diseases was referred to our clinic with a soft tissue lesion at mandibular gingiva. Ultrasonographic examination was performed and a lesion with malignant features was observed. A periapical radiograph was taken to investigate bone destruction and biopsy was planned. Histopathological examination revealed AMM and a literature search was performed to congregate reports which were indexed in PubMed, ScienceDirect, and ResearchGate. Three AMM cases at mandibular gingiva were found. Doppler Ultrasound examination suggested bone destruction and a 1.8 cm × 0.6 cm soft tissue mass with well-defined borders and increased vascularity. Due to its hypervascularity, depth of invasion and destruction at the bone, the lesion was prediagnosed as a malignancy. Lack of melanin pigmentation caused the large immunohistochemical panel study. The tumour cells showed HMB45 and S100 positivity and they were negative with SMA, Desmin, CK1.3, and CK20. Routine ultrasound examination of all soft tissue lesions is very important for assessing features such as vascularity, bone destruction and depth of invasion to detect malignancy. Melanocytic-associated immunohistochemical markers are crucial for AMM diagnosis.

Wetting properties of blood lipid fractions on different titanium surfaces.

BACKGROUND: Blood is the first tissue contacting the implant surface and starting the biological interactions to enhance osseointegration and stimulate bone formation with the progenitor cytokines, chemokines, and growth factors. The coagulation cascade initiates the first step of osseointegration between implant and neighboring tissues. The wound healing may be inadequate unless the blood wets the implant surface properly. Wettability is one of the most important features of the implant surface while lipid level constitutes a milestone that may change the energy of blood, which determines its distribution on implant material. Thus, the aim of this study was to evaluate the effect of lipid component of blood as cholesterol and its treatment on their wetting behavior of titanium surfaces. METHODS: Five surface groups were formed including grade 4 titanium-machined, grade 4 titanium-SLA, grade 4 titanium-SLActive, Roxolid-SLA, and Roxolid-SLActive. In healthy, hyperlipidemic, and treatment situations, blood was taken from eight rabbits and dropped to the disc surfaces. Contact angles were measured between the blood samples and disc surfaces. RESULTS: A significant difference was found between both machined and SLActive surfaces, SLA and SLActive surfaces in the hyperlipidemic period, and only Roxolid-SLA and SLActive surfaces during the treatment period. When evaluated according to time, only grade 4-machined and Grade 4-SLA surfaces showed a significant difference. CONCLUSIONS: Our findings indicated that each period has its own characteristics and showed the importance of cholesterol in blood structure on applicability of implant surfaces.