Objective To probe the clinical character,the histopathological classification and misdiagnoses of intraocular tumors. Methods The clinical and pathological data of 359 patients with intraocular tumor diagnosed clinically between 1980～2000 were retrospectively analyzed. Results There were 300 cases of malignant tumor and 23 cases of benign tumor respectively. Non-oncologic malady and benign tumor misdiagnosed as tumor or malignant tumor were 40 cases. The two leading malignant tumors were retinoblastoma and melanoma. Conclusion The clinical and pathological analysis of intraocular tumor is beneficial to the correct clinical diagnosis and treatment. (Chin J Ocul Fundus Dis,2002,18:28-30)
Objective To observe the configuration and viability of full thickness human fetal retina after short-, mid- and long-term preservation. Methods Twenty-two full thickness human fetal retinae of gestational age of 12-24 weeks were coated by glutin and cut into 88 pieces, and then preserved in Ames' solution, DX solution, -80℃ refrigerator or under cryopreservation condition. The cell viability of retinal neuroepithelial layer was determined by trypan blue staining, retinal configuration was determined by light microscope and electromicroscope. Results The viability of neuroepithelial layer was (94.79plusmn;2.85) % in fresh fetal retina, gt;80% in Ames' solution within 4 hours, and gt;77% in DX solution within 2 days. There was no significant difference between those solution-preservations and the fresh fetal. In -80℃ refrigerator, the viability was (65.83plusmn;5.06)% after 7 days, and then dropped to (57.54plusmn;16.18)% at the end of the first month. Under the cryopreservation condition, the viability was (69.46plusmn;9.31)% at the end of first month. Light and transmission electron microscopy had not deteced any abnormals in the full thickness human fetal retina preserved in Ames' solution within 2 hours, but showed clear retinal layers with bigger intercellular space after preserved in DX solution for 2 days, in -80℃ refrigerator for 7 days and under cryopreservation condition for 1 month. Conclusion Ames' solution and DX solution can preserve good viability and configuration of full thickness human fetal retina in a certain time period.
Pathological images of gastric cancer serve as the gold standard for diagnosing this malignancy. However, the recurrence prediction task often encounters challenges such as insignificant morphological features of the lesions, insufficient fusion of multi-resolution features, and inability to leverage contextual information effectively. To address these issues, a three-stage recurrence prediction method based on pathological images of gastric cancer is proposed. In the first stage, the self-supervised learning framework SimCLR was adopted to train low-resolution patch images, aiming to diminish the interdependence among diverse tissue images and yield decoupled enhanced features. In the second stage, the obtained low-resolution enhanced features were fused with the corresponding high-resolution unenhanced features to achieve feature complementation across multiple resolutions. In the third stage, to address the position encoding difficulty caused by the large difference in the number of patch images, we performed position encoding based on multi-scale local neighborhoods and employed self-attention mechanism to obtain features with contextual information. The resulting contextual features were further combined with the local features extracted by the convolutional neural network. The evaluation results on clinically collected data showed that, compared with the best performance of traditional methods, the proposed network provided the best accuracy and area under curve (AUC), which were improved by 7.63% and 4.51%, respectively. These results have effectively validated the usefulness of this method in predicting gastric cancer recurrence.