Search

A convolution neural network (CNN) uses kernels to filter applied images. These kernels learn their coefficients' values during the training process, thus they do not possess any centrosymmetry. Hence, the phase responses for these kernels are neither zero-phase nor linear-phase. This technique adds a group delay distortion to the filtered images. In this paper, we constrain the values of the kernels' coefficients to be centrosymmetric. This scheme guarantees the prevention of any distortion in the filtered images. In the proposed method, the CNN trains all the kernel coefficients as normal. Then every two-centrosymmetric coefficients are set to their average. This does not affect much the accuracy of the CNN. The proposed algorithm may be used to improve images generated using generative adversarial networks (GAN), autoencoders, image segmentation, and all other algorithms that generate images or video using CNN. This point still requires further study.

BACKGROUND AND OBJECTIVES Many patients with relapsed metastatic breast cancer are pre-treated with taxanes and anthracyclines, which are usually given in the neoadjuvant/adjuvant setting or as first-line treatment for metastatic disease. The primary objective of this study was to determine the overall response rate for combination treatment with gemcitabine and cisplatin in patients with locally advanced or metastatic breast cancer who had relapsed after receiving one adjuvant/neoadjuvant or first-line metastatic chemotherapy regimen containing an anthracycline with/without a taxane. Secondary endpoints included duration of response, time to progression, one-year survival probability, and toxicity. DESIGN AND SETTING A single-arm, open-label, phase 2 study conducted at 17 investigative sites in Egypt. PATIENTS AND METHODS Treatment consisted of gemcitabine (1250 mg/m 2 ) on Days 1 and 8 and cisplatin (70 mg/m 2 ) on Day 1 of each 21-day cycle. Treatment continued until disease progression or a maximum of 6 cycles. RESULTS Of 144 patients all were evaluable for safety and 132 patients were evaluable for efficacy. The overall response rate was 33.3% and 45.5% of the patients with stable disease as their best response. The median time to progression was 5.1 months and the one-year survival probability was 73%. The most common grade 3/4 adverse events were nausea/vomiting (20.1%), neutropenia (19.4%), anemia (13.9%), asthenia (11.1%), diarrhea (9.7%), stomatitis (7.6%), leucopenia (7.6%), and thrombocytopenia (6.2%). Twelve (8.3%) patients had serious adverse events. CONCLUSIONS The results of this study indicate that gemcitabine and cisplatin were active and generally well tolerated in pretreated patients with locally advanced or metastatic breast cancer.