Deep learning-based synthetic CT generation for MR-only radiotherapy of prostate cancer patients with 0.35T MRI linear accelerator.

PURPOSE: To develop a deep learning model to generate synthetic CT for MR-only radiotherapy of prostate cancer patients treated with 0.35 T MRI linear accelerator. MATERIALS AND METHODS: A U-NET convolutional neural network was developed to translate 0.35 T TRUFI MRI into electron density map using a novel cost function equalizing the contribution of various tissue types including fat, muscle, bone, and background air in training. The impact of training time, dataset size, image standardization, and data augmentation approaches was also quantified. Mean absolute error (MAE) between synthetic and planning CTs was calculated to measure the goodness of the model. RESULTS: With 20 patients in training, our U-NET model has the potential to generate synthetic CT with a MAE of about 29.68 ± 4.41, 16.34 ± 2.67, 23.36 ± 2.85, and 105.90 ± 22.80 HU over the entire body, fat, muscle, and bone tissues, respectively. As expected, we found that the number of patients used for training and MAE are nonlinearly correlated. Data augmentation and our proposed loss function were effective to improve MAE by ~9% and ~18% in bony voxels, respectively. Increasing the training time and image standardization did not improve the accuracy of the model. CONCLUSION: A U-NET model has been developed and tested numerically to generate synthetic CT from 0.35T TRUFI MRI for MR-only radiotherapy of prostate cancer patients. Dosimetric evaluation using a large and independent dataset warrants the validity of the proposed model and the actual number of patients needed for the safe usage of the model in routine clinical workflow.

Immunohistochemical stains for CD3 and CD8 do not improve detection of gluten-sensitive enteropathy in duodenal biopsies.

Patients with gluten-sensitive enteropathy usually have increased numbers of duodenal intraepithelial lymphocytes even if the villous architecture is normal. Some authors advocate the use of CD8 and CD3 immunohistochemical stains to improve detection of intraepithelial lymphocytosis, yet the added value of immunohistochemistry when biopsies appear normal remains unproven. The purpose of this study was to evaluate the utility of CD3 and CD8 immunostains in detecting intraepithelial lymphocytosis among duodenal biopsies originally interpreted to be normal based on routine evaluation. We identified 200 duodenal biopsies from 172 patients, all of which were accompanied by a clinical question of gluten-sensitive enteropathy. Five well-oriented villi from each biopsy were assessed. Intraepithelial lymphocytes present in hematoxylin and eosin (H&E)-stained slides were counted and compared with the number of CD3 and CD8 immunopositive cells present in the villous epithelium. Results were expressed as the mean number of intraepithelial lymphocytes or immunopositive cells present per 20 villous tip enterocytes. Review of H&E-stained slides revealed a mean of 2.1 ± 0.1 intraepithelial lymphocytes, compared with 3.2 ± 0.1 CD3-positive and 2.1 ± 0.1 CD8-positive intraepithelial cells (P=<0.001 and 1, respectively), although none of the cases displayed sufficient numbers of intraepithelial lymphocytes to be considered abnormal (ie, ≥ 12/20 enterocytes) by any method. The number of intraepithelial lymphocytes detected by H&E evaluation or immunohistochemistry did not correlate with results of serologic studies for markers of gluten sensitivity. We conclude that immunostains for T cell markers do not improve detection of gluten-sensitive enteropathy when H&E-stained sections are normal.