The performance of the CNNs were evaluated and compared. Of the 844, 717 (85%) were used for training and 127 (15%) were used for testing the three convolutional neural networks (CNNs) (VGG19, Inception V3, and ResNet18).
occlusal or proximal) did not affect the clinical and radiographic outcome of full pulpotomy performed in mature permanent mandibular teeth with partial irreversible pulpitis.Īn accurate diagnosis of deep caries and pulpitis on periapical radiographs is a clinical challenge.Ī total of 844 radiographs were included in this study. For both groups, the combined success rate of full pulpotomy was 93.75%. At the end of follow up period (12 months), success of full pulpotomy in occlusal and proximal caries group was 95% and 92.5% respectively (p=0.644). The significance level was predetermined at p0.05). The data were analysed using the Shapiro-Wilk W test, Two-sample Wilcoxon rank-sum test, Pearson Chi-Square test, Fisher's exact test and Kappa coefficient. Asymptomatic teeth without any radiographic evidence of periapical rarefaction were considered successful. Patients were followed at 6 and 12 months for clinical and radiographic evaluation. Pre and post- operative pain was assessed at 24, 48 and 72 hours using a 11 point Numerical Rating Scale (NRS). Teeth were restored with resin modified glass ionomer cement (RMGIC) and bulk-fill composite resin. Mineral trioxide aggregate (MTA) was compacted over the radicular pulp. The pulp tissue was removed until the root canal orifices and 2.5% sodium hypochlorite (NaOCl) was applied to arrest pulpal bleeding. Full pulpotomy was performed under local anaesthesia and aseptic conditions. Depending upon the location of deep carious lesion, the teeth were distributed equally into occlusal (n=40) and proximal caries groups (n=40). To assess the influence of occlusal and proximal deep carious lesion on the outcome of full pulpotomy performed in mandibular teeth with pulpal diagnosis of symptomatic partial irreversible pulpitis.Įighty deep carious mandibular molar teeth with clinical and radiographic diagnosis of symptomatic partial irreversible pulpitis without periapical rarefaction from patients of either gender between the age of 16–35 years were included. A radiographic threshold between deep and extremely deep lesions is suggested as indicator of the bacterial penetration level and the severity of the pulpal response prior to intervention. Unlike deep lesions, extremely deep carious lesions were often associated with severe pulp inflammation and infection. Hyperplastic pulps were seen only in extremely deep lesions. Whereas radiographically defined deep lesions tended to have bacteria only in the primary dentine (P = 0.000, OR = 20.55, 95% CI ), extremely deep carious lesions tended to have bacteria in contact with the pulpal tissue (P = 0.007, OR =6.84, 95% CI ), presence of an inflammatory infiltrate (Fisher's Exact P = 0.000) and partial pulp necrosis.
There were significant associations between a closed environment (1-2 surfaces involved) and the presence of biofilm, retrograde demineralization and light-coloured demineralized dentine. The effect size was reported by odds ratio (OR) and associated 95% confidence interval (CI). Associations between variables were assessed using Pearson's χ2 or Fisher's Exact test. Interrater agreement was assessed using Cohen´s kappa. The material was scored with regard to coronal break-down, macroscopic variables describing caries activity, and histological variables describing the dentine-pulp complex (bacteria, inflammatory infiltrate, partial pulp necrosis, hyperplastic changes, hard tissue/ectopic presence of connective tissue). After extraction, the teeth were processed for histology. The material comprised 68 untreated cavitated permanent teeth divided into well-defined radiographic categories based on the lesion-penetration depth: a) Deep lesions ( ≥¾ of the dentine thickness with a radio-dense zone separating the lesion from the pulp) and b) extremely deep lesions (the carious lesion penetrated the entire thickness of the dentine, without a radio-dense zone). To investigate the relationship between radiographically and macroscopically well-defined carious lesions and the dentine-pulp complex with regard to: a) level of bacterial penetration, b) inflammatory status including the presence of hyperplastic pulp stroma and c) formation of hard and/or ectopic connective tissue.