ABSTRACT
Magnetic resonance imaging(MRI) can obtain multi-modal images with different contrast, which provides rich information for clinical diagnosis. However, some contrast images are not scanned or the quality of the acquired images cannot meet the diagnostic requirements due to the difficulty of patient's cooperation or the limitation of scanning conditions. Image synthesis techniques have become a method to compensate for such image deficiencies. In recent years, deep learning has been widely used in the field of MRI synthesis. In this paper, a synthesis network based on multi-modal fusion is proposed, which firstly uses a feature encoder to encode the features of multiple unimodal images separately, and then fuses the features of different modal images through a feature fusion module, and finally generates the target modal image. The similarity measure between the target image and the predicted image in the network is improved by introducing a dynamic weighted combined loss function based on the spatial domain and K-space domain. After experimental validation and quantitative comparison, the multi-modal fusion deep learning network proposed in this paper can effectively synthesize high-quality MRI fluid-attenuated inversion recovery (FLAIR) images. In summary, the method proposed in this paper can reduce MRI scanning time of the patient, as well as solve the clinical problem of missing FLAIR images or image quality that is difficult to meet diagnostic requirements.
Subject(s)
Humans , Deep Learning , Magnetic Resonance Imaging/methods , Image Processing, Computer-Assisted/methodsABSTRACT
BACKGROUND:During ordinary plate fixation, the soft tissues around the fracture of the proximal humerus are nearly stripped to impact blood supply, and moreover, an ordinary steel plate cannot meet with the fixed requirements for severe osteoporosis, large bone defects and comminuted fractures. OBJECTIVE: To observe the functional recovery and complications in middle-aged patients with proximal humeral fractures undergoing stainless steel locking plate implantation. METHODS:From March 2011 to March 2014, 48 patients with proximal humeral fractures were treated in the 306th RESULTS AND CONCLUSION: The 48 patients were folowed up for 6-17 years, and the mean healing time was (15.3±1.2) weeks. At the last folow-up, the Neer scores were excelent in 12 cases, good in 22 cases, fair in 11 cases and poor in 3 cases, with an excelent-good rate of 71%. After internal fixation, there was one case of Hospital of PLA, including 20 males and 28 females, with an average age of 58 years. Of the 48 cases, there were 9 cases of Neer 2 fractures, 26 cases of Neer 3 fractures, and 13 cases of Neer 4 cases, al of which belonged to closed injuries. Al the patients were subject to locking plate implantation for repair of fractures of the proximal humerus, and evaluated based on Neer scores. soft tissue infection, two cases of traumatic arthritis, and no bone ununion and osteomyelitis. These findings suggest that the stainless steel locking plate implantation for repair of fractures of the proximal humerus can achieve a good anatomic reduction under minimaly invasive conditions and produce a stable rehabilitation environment for the soft tissue around the shoulder joint by internal fixation. Hence, the factures can recover faster with less complications. The stainless steel locking plate implantation can obtain good achievements in the repair of proximal humeral fractures of Neer 2, 3 as wel as Neer 4 in young patients with good bone quality.