• <xmp id="1ykh9"><source id="1ykh9"><mark id="1ykh9"></mark></source></xmp>
      <b id="1ykh9"><small id="1ykh9"></small></b>
    1. <b id="1ykh9"></b>

      1. <button id="1ykh9"></button>
        <video id="1ykh9"></video>
      2. west china medical publishers
        Author
        • Title
        • Author
        • Keyword
        • Abstract
        Advance search
        Advance search

        Search

        find Author "ZHU Weipeng" 2 results
        • Application of hybrid attention guided dual-path residual learning in magnetic resonance imaging diagnosis

          Multimodal magnetic resonance imaging (MRI) often faces challenges such as insufficient model generalization ability and high misdiagnosis rates in intelligent neuroimaging diagnosis. Existing methods, including single-path classification networks such as residual networks and EfficientNet, and segmentation-oriented workflow models such as U-Net, have the following limitations when processing multi-sequence MRI data: ① single-path structures have difficulty in effectively decoupling modality-specific features; ② conventional convolutions lack adaptive enhancement in the channel and spatial dimensions. Therefore, this paper proposes a dual-path residual channel-spatial attention network (DRCSA-Net). The model learns complementary representations through a parallel dual-branch structure (channel enhancement and spatial enhancement), introduces a squeeze-and-excitation module and a channel-spatial attention module to achieve channel recalibration and spatial attention focusing, and finally integrates information and completes classification through lightweight fusion convolution and fully connected layers. To comprehensively evaluate the proposed model, five-fold cross-validation, single-/dual-path comparison, ablation experiments, and robustness analysis were conducted on four public datasets. The experimental results show that DRCSA-Net achieves high accuracy on all four public datasets and demonstrates good effectiveness and stability in the single-/dual-path comparison, ablation experiments, and robustness analysis, providing a structurally clear and stable implementation scheme for brain tumor MRI classification.

          Release date: Export PDF Favorites Scan
        • Non-mechanical suture completely anatomic thoracoscopic lobectomy: A case control study

          ObjectiveTo investigate the safety and feasibility of thoracoscopic lobectomy without mechanical suture.MethodsThe data of 28 consecutive patients (a non-mechanical suture group, 16 males and 12 females at age of 61.23±11.10 years) who underwent non-mechanical suture anatomic thoracoscopic lobectomy performed by the same surgeon from March 2015 to March 2018 were analyzed retrospectively, and 28 patients (18 males and 10 females at age of 59.45±13.39 years) who underwent completely anatomic thoracoscopic lobectomy with endoscopic stapler (a mechanical suture group) in the same period were matched. The clinical effectiveness of the two groups was compared.ResultsThe operation time between the non-mechanical suture group (136.30±53.46 min) and the mechanical suture group (109.63±44.61 min) showed a statistical difference (P<0.05). While in term of intraoperative bleeding volume (65.00 ml vs. 50.00 ml), postoperative thoracic drainage time (3.73 days vs. 3.56 days), thoracic drainage volume (538.60 ml vs. 563.70 ml), postoperative hospital stay (5.58 days vs. 5.35 days) and postoperative complication rate (5/28 vs. 6/28), there was no statistical difference between the two groups. Hospitalization expense was significantly different between the two groups (35 438.30 yuan vs. 51 693.60 yuan).ConclusionNon-mechanical suture thoracoscopic anatomic lobectomy is safe and feasible, and can significantly reduce the medical cost but prolong the operation time.

          Release date:2019-01-23 02:58 Export PDF Favorites Scan
        1 pages Previous 1 Next

        Format

        Content

      3. <xmp id="1ykh9"><source id="1ykh9"><mark id="1ykh9"></mark></source></xmp>
          <b id="1ykh9"><small id="1ykh9"></small></b>
        1. <b id="1ykh9"></b>

          1. <button id="1ykh9"></button>
            <video id="1ykh9"></video>
          2. 射丝袜