RESUMO
In tomographic reconstruction, the image quality of the reconstructed images can be significantly degraded by defects in the measured two-dimensional (2D) raw image data. Despite the importance of screening defective 2D images for robust tomographic reconstruction, manual inspection and rule-based automation suffer from low-throughput and insufficient accuracy, respectively. Here, we present deep learning-enabled quality control for holographic data to produce robust and high-throughput optical diffraction tomography (ODT). The key idea is to distil the knowledge of an expert into a deep convolutional neural network. We built an extensive database of optical field images with clean/noisy annotations, and then trained a binary-classification network based upon the data. The trained network outperformed visual inspection by non-expert users and a widely used rule-based algorithm, with >90% test accuracy. Subsequently, we confirmed that the superior screening performance significantly improved the tomogram quality. To further confirm the trained model's performance and generalisability, we evaluated it on unseen biological cell data obtained with a setup that was not used to generate the training dataset. Lastly, we interpreted the trained model using various visualisation techniques that provided the saliency map underlying each model inference. We envision the proposed network would a powerful lightweight module in the tomographic reconstruction pipeline.
RESUMO
BACKGROUNDS: Dual immunocytochemistry (DIC) with cytokeratin (CK) 20 and p53 in liquid-based cytology is a tool for improving the accuracy of urine cytology (UC). This study was conducted to compare the diagnostic accuracy of UC alone with that of UC combined with CK20/p53 DIC. METHODS: We retrieved urine samples collected between January 2015 and March 2016 stored in PreservCyt®solution that were from cases categorized as malignant, highly suspicious, suspicious, and atypical and that were matched with a subsequent biopsy. We re-prepared 63 samples of 28 patients for DIC and blindly evaluated 63 pairs of original Papanicolaou smears and DIC. RESULTS: Of the 63 samples, 11 could not be analyzed because of the low number of atypical urothelial cells, and the results of the remaining 52 samples were as follows: 34 positive and 18 negative. The positive predictive value of DIC was 100%, and the negative predictive value was 78%. Fifteen DIC-positive cases, histologically proven as malignant were originally diagnosed as highly suspicious (4), suspicious (8), and atypical (3), which were strongly suggestive of "urothelial carcinoma". Four negative cases, histologically confirmed as non-neoplastic cases, were filtered from false positivity. CONCLUSIONS: Despite the small sample size, this study demonstrated the diagnostic utility, high sensitivity, and positive predictive value of CK20/p53 DIC, especially in cases with a small number of single malignant cells or cellular clusters of reactive atypical urothelial cells. Thus, CK20/p53 DIC can be used for improving diagnostic accuracy of UC, either as an ancillary method to cytology or as a part of a potential future diagnostic panel to improve patient diagnosis and management.