Cone-beam CT image quality improvement using Cycle-Deblur consistent adversarial networks (Cycle-Deblur GAN) for chest CT imaging in breast cancer patients.

Cone-beam computed tomography (CBCT) integrated with a linear accelerator is widely used to increase the accuracy of radiotherapy and plays an important role in image-guided radiotherapy (IGRT). For comparison with fan-beam computed tomography (FBCT), the image quality of CBCT is indistinct due to X-ray scattering, noise, and artefacts. We proposed a deep learning model, "Cycle-Deblur GAN", combined with CycleGAN and Deblur-GAN models to improve the image quality of chest CBCT images. The 8706 CBCT and FBCT image pairs were used for training, and 1150 image pairs were used for testing in deep learning. The generated CBCT images from the Cycle-Deblur GAN model demonstrated closer CT values to FBCT in the lung, breast, mediastinum, and sternum compared to the CycleGAN and RED-CNN models. The quantitative evaluations of MAE, PSNR, and SSIM for CBCT generated from the Cycle-Deblur GAN model demonstrated better results than the CycleGAN and RED-CNN models. The Cycle-Deblur GAN model improved image quality and CT-value accuracy and preserved structural details for chest CBCT images.

Clinical value of combined detection of serum matrix metalloproteinase-9, heparanase, and cathepsin for determining ovarian cancer invasion and metastasis.

AIM: This study evaluated the clinical value of the combined detection of serum cathepsin L (CL), heparanase (Hpa), and matrix metalloproteinase-9 (MMP-9) for determining the degree of ovarian cancer invasion and metastasis before surgery. PATIENTS AND METHODS: Enzyme-linked immunosorbent assays were used to measure the serum content of CL, Hpa, and MMP-9 in 217 patients with untreated ovarian cancer before surgery, 100 patients with benign ovarian tumors, and 101 healthy women as controls. In addition, the degrees of invasion and metastasis were assessed by the 'gold standard' of clinicopathological diagnosis. The associations of the preoperative serum CL, Hpa, and MMP-9 levels with the clinicopathological factors and metastatic status were analyzed. Receiver operating characteristic (ROC) curve analysis was used to evaluate the usefulness of these markers for determining the degree of ovarian cancer invasion before surgery. RESULTS: The serum CL, Hpa, and MMP-9 levels were significantly higher (p=0.001) in patients with malignant ovarian cancer compared with patients with benign ovarian tumors and healthy controls. The serum CL level was significantly higher in patients with epithelial ovarian carcinoma compared with non-epithelial ovarian carcinoma (p=0.048), whereas the serum levels of Hpa (p=0.109) and MMP-9 (p=0.544) did not differ significantly between these two groups. The serum CL, Hpa, and MMP-9 levels correlated with the degree of differentiation and the FIGO staging (p>0.05). The serum CL (p=0.030) and MMP-9 (p=0.010) levels were significantly associated with peritoneal metastasis, and the serum Hpa level (p=0.042) was associated with distant metastasis. A ROC curve analysis revealed sensitivity of 60.9%, 69.6%, and 72.2%, and specificity of 57.4%, 67.2%, and 68.9% for the preoperative serum levels of CL, Hpa, and MMP-9, respectively, as tumor markers for the degree of extra-pelvic metastasis. CONCLUSION: Elevated serum CL, Hpa, and MMP-9 levels are correlated with malignant invasion and progression in ovarian cancer. The combined detection of serum CL, Hpa, and MMP-9 may be useful for determining the extent of ovarian cancer metastasis before surgery.