Evaluation of an artificial intelligence system for detecting vertical root fracture on panoramic radiography

Objectives

The aim of this study was to evaluate the use of a convolutional neural network (CNN) system for detecting vertical root fracture (VRF) on panoramic radiography.

Methods

Three hundred panoramic images containing a total of 330 VRF teeth with clearly visible fracture lines were selected from our hospital imaging database. Confirmation of VRF lines was performed by two radiologists and one endodontist. Eighty percent (240 images) of the 300 images were assigned to a training set and 20% (60 images) to a test set. A CNN-based deep learning model for the detection of VRFs was built using DetectNet with DIGITS version 5.0. To defend test data selection bias and increase reliability, fivefold cross-validation was performed. Diagnostic performance was evaluated using recall, precision, and F measure.

Results

Of the 330 VRFs, 267 were detected. Twenty teeth without fractures were falsely detected. Recall was 0.75, precision 0.93, and F measure 0.83.

Conclusions

The CNN learning model has shown promise as a tool to detect VRFs on panoramic images and to function as a CAD tool.

     

The significance of thymidine phosphorylase activity in hepatocellular carcinoma and chronic diseased livers: a special reference to liver fibrosis and multicentric tumor occurrence

The role of thymidine phosphorylase (TP), an angiogenic factor, in hepatocellular carcinoma (HCC) remains unclear. The aim of this study was to clarify the significance of TP in HCC. Thirty-seven patients with HCC, who underwent hepatectomy, were included. The TP activity in both cancerous and non-cancerous parts of livers were measured by an enzyme-linked immunosorbent assay. Another 11 patients without HCC were used to evaluate the TP activity in the non-cancerous parts of livers. Both the cancerous and non-cancerous TP activities were clinico-pathologically investigated with special reference to the multicentric occurrence of HCCs and the degree of liver fibrosis; consisting of normal, fibrosis and cirrhosis. The TP activity in the cancerous part was 94.6 +/- 70.2 U/mg protein, while that in non-cancerous parts of the liver was 80.9 +/- 48.8 U/mg protein. No significant difference was observed. The TP activity in the cancerous part did not correlate with any clinicopathological variables, such as tumor differentiation, portal vein invasion, intrahepatic metastases and prognosis. However, the TP activity in the non-cancerous parts of the liver correlated with the degree of fibrosis (normal/fibrosis/cirrhosis = 34:74:90 U/ mg protein, respectively). Furthermore, regarding the correlation between TP activity in the non-cancerous parts and the simultaneously multicentric occurrence of HCC, the TP activity in the multicentric group (n = 8; 121 U/mg protein) was significantly higher than that in the non-multicentric group (n = 29; 70 U/mg protein). The TP activity in the non-cancerous parts increased in proportion to the degree of liver fibrosis. Furthermore, it is suggested that the higher TP activity in the non-cancerous part is related to the multicentric occurrence of HCCs.