SW-UNet: a U-Net fusing sliding window transformer block with CNN for segmentation of lung nodules.

Medical images are information carriers that visually reflect and record the anatomical structure of the human body, and play an important role in clinical diagnosis, teaching and research, etc. Modern medicine has become increasingly inseparable from the intelligent processing of medical images. In recent years, there have been more and more attempts to apply deep learning theory to medical image segmentation tasks, and it is imperative to explore a simple and efficient deep learning algorithm for medical image segmentation. In this paper, we investigate the segmentation of lung nodule images. We address the above-mentioned problems of medical image segmentation algorithms and conduct research on medical image fusion algorithms based on a hybrid channel-space attention mechanism and medical image segmentation algorithms with a hybrid architecture of Convolutional Neural Networks (CNN) and Visual Transformer. To the problem that medical image segmentation algorithms are difficult to capture long-range feature dependencies, this paper proposes a medical image segmentation model SW-UNet based on a hybrid CNN and Vision Transformer (ViT) framework. Self-attention mechanism and sliding window design of Visual Transformer are used to capture global feature associations and break the perceptual field limitation of convolutional operations due to inductive bias. At the same time, a widened self-attentive vector is used to streamline the number of modules and compress the model size so as to fit the characteristics of a small amount of medical data, which makes the model easy to be overfitted. Experiments on the LUNA16 lung nodule image dataset validate the algorithm and show that the proposed network can achieve efficient medical image segmentation on a lightweight scale. In addition, to validate the migratability of the model, we performed additional validation on other tumor datasets with desirable results. Our research addresses the crucial need for improved medical image segmentation algorithms. By introducing the SW-UNet model, which combines CNN and ViT, we successfully capture long-range feature dependencies and break the perceptual field limitations of traditional convolutional operations. This approach not only enhances the efficiency of medical image segmentation but also maintains model scalability and adaptability to small medical datasets. The positive outcomes on various tumor datasets emphasize the potential migratability and broad applicability of our proposed model in the field of medical image analysis.

Monitoring the Early Antiproliferative Effect of the Analgesic-Antitumor Peptide, BmK AGAP on Breast Cancer Using Intravoxel Incoherent Motion With a Reduced Distribution of Four b-Values.

Background: The present study aimed to investigate the possibility of using intravoxel incoherent motion (IVIM) diffusion magnetic resonance imaging (MRI) to quantitatively assess the early therapeutic effect of the analgesic-antitumor peptide BmK AGAP on breast cancer and also evaluate the medical value of a reduced distribution of four b-values. Methods: IVIM diffusion MRI using 10 b-values and 4 b-values (0-1,000 s/mm2) was performed at five different time points on BALB/c mice bearing xenograft breast tumors treated with BmK AGAP. Variability in Dslow, Dfast, PF, and ADC derived from the set of 10 b-values and 4 b-values was assessed to evaluate the antitumor effect of BmK AGAP on breast tumor. Results: The data showed that PF values significantly decreased in rBmK AGAP-treated mice on day 12 (P = 0.044). PF displayed the greatest AUC but with a poor medical value (AUC = 0.65). The data showed no significant difference between IVIM measurements acquired from the two sets of b-values at different time points except in the PF on the day 3. The within-subject coefficients of variation were relatively higher in Dfast and PF. However, except for a case noticed on day 0 in PF measurements, the results indicated no statistically significant difference at various time points in the rBmK AGAP-treated or the untreated group (P < 0.05). Conclusion: IVIM showed poor medical value in the early evaluation of the antiproliferative effect of rBmK AGAP in breast cancer, suggesting sensitivity in PF. A reduced distribution of four b-values may provide remarkable measurements but with a potential loss of accuracy in the perfusion-related parameter PF.

Soft Tissue Sarcomas: Preoperative Predictive Histopathological Grading Based on Radiomics of MRI.

RATIONALE AND OBJECTIVES: The purpose of this study is to develop a radiomics model for predicting the histopathological grades of soft tissue sarcomas preoperatively through magnetic resonance imaging (MRI). MATERIALS AND METHODS: Thirty-five patients who were pathologically diagnosed with soft tissue sarcomas and their histological grades were recruited. All patients had undergone MRI before surgery on a 3.0T MRI scanner. Radiomics features were extracted from fat-suppressed T2-weighted imaging. We used the least absolute shrinkage and selection operator (LASSO) regression method to select features. Then three machine learning classification methods, including random forests, k-nearest neighbor, and support vector machine algorithm were trained using the 5-fold cross validation strategy to separate the soft tissue sarcomas with low- and high-histopathological grades. RESULTS: The radiomics features were significantly associated with the histopathological grades. Quantitative imaging features (n = 1049) were extracted from fat-suppressed T2-weighted imaging, and five features were selected to construct the radiomics model. The model that used support vector machine classification method achieved the best performance among the three methods, with areas under the receiver operating characteristic curves Area Under Curve (AUC) values of 0.92 ± 0.07, accuracy of 0.88. CONCLUSION: Good accuracy and AUC could be obtained using only five radiomic features. Therefore, we proposed that three-dimensional imaging features from fat-suppressed T2-weighted imaging could be used as candidate biomarkers for preoperative prediction of histopathological grades of soft tissue sarcomas noninvasively.

Intravoxel Incoherent Motion MR Imaging: Comparison of Diffusion and Perfusion Characteristics for Differential Diagnosis of Soft Tissue Tumors.

We used intravoxel incoherent motion (IVIM) magnetic resonance imaging (MRI) to explore the possibility of preoperative diagnosis of soft tissue tumors (STTs). This prospective study enrolled 23 patients. Conventional MRI and IVIM examinations were performed on a 3.0T MR imager. Eight (35%) hemangiomas, 11 (47%) benign soft tissue tumors excluding hemangiomas (BSTTEHs) and 4 soft tissue sarcomas (STSs) were assessed. The mean tumor size was about 1652.36 ±â€Š233.66  mm(2). Ten b values (0-800  s/mm(2)) were used to evaluate diffusion and perfusion characteristics of IVIM. IVIM parameters (ADC(standard), ADC(slow), ADC(fast), and f) of STTs were measured and evaluated for differentiating hemangiomas, BSTTEHs, and STSs. ADC(slow) and ADC(fast) value were different for hemangiomas, BSTTEHs, and STSs separately (P < 0.001, P < 0.001, and P = 0.001). ADC(slow), cut-off value smaller than 0.93 × 10(-3)  mm(2)/s, was the best parameter to differ STSs (0.689 ±â€Š0.173 × 10 (-3)mm(2)/s) from hemangiomas (0.933 ±â€Š0.237 × 10 (-3)mm(2)/s) and BSTTEHs (1.156 ±â€Š0.120 × 10 (-3)mm(2)/s) (P = 0.001). ADC(slow) (0.93 × 10(-3)  mm(2)/s