ABSTRACT
OBJECTIVES@#To investigate the reliability and accuracy of deep learning technology in automatic sex estimation using the 3D reconstructed images of the computed tomography (CT) from the Chinese Han population.@*METHODS@#The pelvic CT images of 700 individuals (350 males and 350 females) of the Chinese Han population aged 20 to 85 years were collected and reconstructed into 3D virtual skeletal models. The feature region images of the medial aspect of the ischiopubic ramus (MIPR) were intercepted. The Inception v4 was adopted as the image recognition model, and two methods of initial learning and transfer learning were used for training. Eighty percent of the individuals' images were randomly selected as the training and validation dataset, and the remaining were used as the test dataset. The left and right sides of the MIPR images were trained separately and combinedly. Subsequently, the models' performance was evaluated by overall accuracy, female accuracy, male accuracy, etc.@*RESULTS@#When both sides of the MIPR images were trained separately with initial learning, the overall accuracy of the right model was 95.7%, the female accuracy and male accuracy were both 95.7%; the overall accuracy of the left model was 92.1%, the female accuracy was 88.6% and the male accuracy was 95.7%. When the left and right MIPR images were combined to train with initial learning, the overall accuracy of the model was 94.6%, the female accuracy was 92.1% and the male accuracy was 97.1%. When the left and right MIPR images were combined to train with transfer learning, the model achieved an overall accuracy of 95.7%, and the female and male accuracies were both 95.7%.@*CONCLUSIONS@#The use of deep learning model of Inception v4 and transfer learning algorithm to construct a sex estimation model for pelvic MIPR images of Chinese Han population has high accuracy and well generalizability in human remains, which can effectively estimate the sex in adults.
Subject(s)
Adult , Female , Humans , Male , Young Adult , Middle Aged , Aged , Aged, 80 and over , Deep Learning , Imaging, Three-Dimensional , Pelvis , Reproducibility of Results , Tomography, X-Ray ComputedABSTRACT
Background@#Extranodal natural killer/T-cell lymphoma (ENKTL), nasal type, is an aggressive entity within the World Health Organization classification of lymphoid tumors. The International Prognostic Index is reported to be prognostically meaningful for ENKTL, but lacks discriminatory power for stage I/II ENKTL with extensive local invasion. This study aimed to evaluate the prognostic effects of local invasion by site and tissue type in patients with ENKTL.@*Methods@#We retrospectively analyzed data of 86 patients who were diagnosed with ENKTL by the Department of Pathology of Beijing Tongren Hospital from June 2002 to April 2016, and ascertained tumor infiltration of adjacent structures (AS), bone, and soft tissue for each patient, using physical findings and imaging scans. We used univariate and multivariate analysis to assess the association of each involved tissue or site with patients’ overall survival (OS).@*Results@#Of the 86 patients, 71(82.6%) experienced invasion of AS, 22(25.6%) of soft tissue, and 26(30.2%) had bone involvement. Overall, patients with AS involvement did not show significantly shorter survival than those without AS involvement (Log rank χ2 = 1.177, P = 0.278); however, patients who had involved eyeballs or brains showed significantly lower 2-year OS rates than those without eyeball involvement (Log rank χ2 = 4.105, P = 0.043) or brain involvement (Log rank χ2 = 7.126, P = 0.008). Patients with involved local soft tissue or bones, respectively, showed lower 2-year OS rates than those without involved local soft tissue (Log rank χ2 = 10.390, P = 0.001) or bones (Log rank χ2 = 8.993, P = 0.003). Multivariate analysis showed that involvement of the cheek or facial muscles (hazard ratio, HR = 5.471, 95% confidence interval [CI]: 1.466–20.416, P = 0.011) and the maxilla bone (HR = 6.120, 95% CI: 1.517–24.694, P = 0.011) were significantly independent predictors of lower 2-year OS rates.@*Conclusions@#Imaging can accurately detect ENKTL invasion of AS, soft tissue, and bone. Involvement of local soft tissue or bone was significantly associated with lower 2-year OS rates. Involvements of the cheek or facial muscle, as well as maxilla bone, are independent predictors of lower 2-year OS rates in ENKTL patients.
ABSTRACT
BACKGROUND@#Extranodal natural killer/T-cell lymphoma (ENKTL), nasal type, is an aggressive entity within the World Health Organization classification of lymphoid tumors. The International Prognostic Index is reported to be prognostically meaningful for ENKTL, but lacks discriminatory power for stage I/II ENKTL with extensive local invasion. This study aimed to evaluate the prognostic effects of local invasion by site and tissue type in patients with ENKTL.@*METHODS@#We retrospectively analyzed data of 86 patients who were diagnosed with ENKTL by the Department of Pathology of Beijing Tongren Hospital from June 2002 to April 2016, and ascertained tumor infiltration of adjacent structures (AS), bone, and soft tissue for each patient, using physical findings and imaging scans. We used univariate and multivariate analysis to assess the association of each involved tissue or site with patients' overall survival (OS).@*RESULTS@#Of the 86 patients, 71 (82.6%) experienced invasion of AS, 22 (25.6%) of soft tissue, and 26 (30.2%) had bone involvement. Overall, patients with AS involvement did not show significantly shorter survival than those without AS involvement (Log rank χ = 1.177, P = 0.278); however, patients who had involved eyeballs or brains showed significantly lower 2-year OS rates than those without eyeball involvement (Log rank χ = 4.105, P = 0.043) or brain involvement (Log rank χ = 7.126, P = 0.008). Patients with involved local soft tissue or bones, respectively, showed lower 2-year OS rates than those without involved local soft tissue (Log rank χ = 10.390, P = 0.001) or bones (Log rank χ = 8.993, P = 0.003). Multivariate analysis showed that involvement of the cheek or facial muscles (hazard ratio, HR = 5.471, 95% confidence interval [CI]: 1.466-20.416, P = 0.011) and the maxilla bone (HR = 6.120, 95% CI: 1.517-24.694, P = 0.011) were significantly independent predictors of lower 2-year OS rates.@*CONCLUSIONS@#Imaging can accurately detect ENKTL invasion of AS, soft tissue, and bone. Involvement of local soft tissue or bone was significantly associated with lower 2-year OS rates. Involvements of the cheek or facial muscle, as well as maxilla bone, are independent predictors of lower 2-year OS rates in ENKTL patients.
ABSTRACT
Purpose To investigate the clinicopathological features of sinonasal primary secretory carcinoma (SC) and its diagnosis, differential diagnosis. Methods Two cases of sin-nasal SC were analyzed by light microscopy with immunohisto-chemical staining (EnVision) for CK, vimentin, S-100 protein, SOX10, PAS, DPAS, Mamaglobin, Calponin, DOG1, p63 and molecular detection of ETV6 gene break. Results Morphologically, SC revealed varying proportions of solid, tubular, acinar, microcystic, tubular growth patterns. All SC cases were cytological low grade with uniform cells, small-to medium-sizes nuclei, occasional small nucleoli, and abundant pink, bubbly cytoplasm. Mitotic figures were rarely encountered. Tumor cells secreted eosinophilic, colloid-like secretions that were PAS positive. There were no DPAS positive zymogen granules in cyto-plasm. This tumor cells were CK, vimentin, S-100, SOX10, PAS positive and Mamaglobin, Calponin and p63 negative. The ETV6 gene rearrangement was confirmed in all cases by fluorescence in situ hybridization (FISH). After excision, all two patients were survival without tumor recurrence for 41 months and53 months. Conclusion Sinonasal primary SC is a low grade malignant tumor. The histological features of SC are overlap with other salivary gland tumors. Immunohistochemical analysis and FISH are useful for the diagnosis and differential diagnosis.