Research on Target Image Classification in Low-Light Night Vision.

In extremely dark conditions, low-light imaging may offer spectators a rich visual experience, which is important for both military and civic applications. However, the images taken in ultra-micro light environments usually have inherent defects such as extremely low brightness and contrast, a high noise level, and serious loss of scene details and colors, which leads to great challenges in the research of low-light image and object detection and classification. The low-light night vision image used as the study object in this work has an excessively dim overall picture and very little information about the screen's features. Three algorithms, HE, AHE, and CLAHE, were used to enhance and highlight the image. The effectiveness of these image enhancement methods is evaluated using metrics such as the peak signal-to-noise ratio and mean square error, and CLAHE was selected after comparison. The target image includes vehicles, people, license plates, and objects. The gray-level co-occurrence matrix (GLCM) was used to extract the texture features of the enhanced images, and the extracted image texture features were used as input to construct a backpropagation (BP) neural network classification model. Then, low-light image classification models were developed based on VGG16 and ResNet50 convolutional neural networks combined with low-light image enhancement algorithms. The experimental results show that the overall classification accuracy of the VGG16 convolutional neural network model is 92.1%. Compared with the BP and ResNet50 neural network models, the classification accuracy was increased by 4.5% and 2.3%, respectively, demonstrating its effectiveness in classifying low-light night vision targets.

Prone position versus supine position in postoperative radiotherapy for breast cancer: A meta-analysis.

BACKGROUND: This meta-analysis evaluates the difference of sparing organs at risk (OAR) in different position (Prone position and Supine position) with different breathing patterns (Free breathing, FB/Deep inspiration breath hold, DIBH) for breast cancer patients receiving postoperative radiotherapy and provides a useful reference for clinical practice. METHOD: The relevant controlled trials of prone position versus supine position in postoperative radiotherapy for breast cancer were retrieved from the sources of PubMed, Cochrane Library, Embase, Web of Science and ClinicalTrails.gov. The principal outcome of interest was OAR doses (heart dose, left anterior descending coronary artery dose and ipsilateral lung dose) and target coverage. We mainly compared the effects of P-FB (Prone position FB) and S-FB (Supine position FB) and discussed the effects of DIBH combined with different positions on OAR dose in postoperative radiotherapy. We calculated summary standardized mean difference (SMD) and 95% confidence intervals (CI). The meta-analysis was performed using RevMan 5.4 software. RESULTS: The analysis included 751 patients from 19 observational studies. Compared with the S-FB, the P-FB can have lower heart dose, left anterior descending coronary artery (LADCA) dose, and ipsilateral lung dose (ILL) more effectively, and the difference was statistically significant (heart dose, SMD = - 0.51, 95% CI - 0.66 ∼ - 0.36, P < .00001. LADCA dose, SMD = - 0.58, 95% CI - 0.85 ∼ - 0.31, P < .0001. ILL dose, SMD = - 2.84, 95% CI - 3.2 ∼ - 2.48, P < .00001). And there was no significant difference in target coverage between the S-FB and P-FB groups (SMD = - 0.1, 95% CI - 0.57 ∼ 0.36, P = .66). Moreover, through descriptive analysis, we found that P-DIBH (Prone position DIBH) has better sparing OAR than P-FB and S-DIBH (Supine position DIBH). CONCLUSION: By this meta-analysis, compared with the S-FB we found that implementation of P-FB in postoperative radiotherapy for breast cancer can reduce irradiation of heart dose, LADCA dose and ILL dose, without compromising mean dose of target coverage. Moreover, P-DIBH might become the most promising way for breast cancer patients to undergo radiotherapy.

Meta-analysis of deep inspiration breath hold (DIBH) versus free breathing (FB) in postoperative radiotherapy for left-side breast cancer.

OBJECTIVES: This meta-analysis evaluates the difference in deep inspiration breath hold (DIBH) versus free breathing (FB) for patients receiving postoperative radiotherapy for left breast cancer and provides a useful reference for clinical practice. METHODS: The relevant controlled trials of DIBH versus FB in postoperative radiotherapy for left-side breast cancer were retrieved from the databases of PubMed, Science Direct, Cochrane Library, and Web of Science databases. The principal outcome of interest was heart dose, left anterior descending coronary artery (LADCA) dose, and left lung dose and target coverage. We calculated summary standardized mean difference (SMD) and 95% confidence intervals (CI). The meta-analysis was performed using RevMan 5.3 software. RESULTS: The analysis included 1019 patients from 12 observational studies, of which 576 cases were in the DIBH group and 443 cases in the FB group. Compared with the FB group, the DIBH group can have lower heart dose, left anterior descending coronary artery (LADCA) dose, and left lung dose more effectively, and the difference was statistically significant (heart dose, SMD = - 1.36, 95% CI - 1.64 ~ - 1.09, P < 0.01. LADCA dose, SMD = - 1.45, 95% CI - 1.62 ~ - 1.27, P < 0.01. Left lung dose, SMD = - 0.52, 95% CI - 0.81 ~ - 0.23, P < 0.01). There was no significant difference in target coverage between the two groups (SMD = 0.03, 95% CI - 0.11 ~ 0.18, P = 0.64). CONCLUSION: By this meta-analysis, we found that implementation of DIBH in postoperative radiotherapy for left-side breast cancer can reduce irradiation of heart dose, LADCA dose and left lung dose, without compromising target coverage.

Abdominal DIBH reduces the cardiac dose even further: a prospective analysis.

BACKGROUND: Deep inspiration breath hold (DIBH) can be performed using different breathing maneuvers, such as DIBH with a thoracic breathing maneuver (T-DIBH) and DIBH with an abdominal breathing maneuver (A-DIBH). Dosimetric benefits of A-DIBH were investigated in the treatment of left-sided breast cancer radiotherapy (RT) with both 3-Dimensional conformal radiation therapy (3D-CRT) and intensity-modulated radiotherapy (IMRT) techniques. METHODS: Twenty-two patients with left-sided breast cancer were enrolled in this study. 3D-CRT and IMRT plans were generated for each patient with three different CT scans of free breathing (FB), T-DIBH and A-DIBH. There were total of six treatment plans generated for each patient: FB_3D-CRT; TDIBH_3D-CRT; ADIBH_3D-CRT; FB-IMRT; TDIBH-IMRT; ADIBH-IMRT. Doses to the heart, left anterior descending coronary artery (LADCA), and ipsilateral lung were evaluated and compared using the Wilcoxon signed-rank test. RESULTS: The mean doses to the heart, LADCA and ipsilateral lung in 3D-CRT plans generated from 3D-CRT with FB, T-DIBH and A-DIBH were (2.89 ± 1.30), (1.67 ± 0.90) and (1.34 ± 0.43) Gy (all P < 0.05), respectively, with FB; (29.08 ± 16.72), (13.94 ± 14.74) and (10.22 ± 10.30) Gy (all P < 0.05), respectively, with T-DIBH; and (7.77 ± 2.71), (7.32 ± 1.42) and (6.90 ± 1.60) Gy (all P < 0.05), respectively, with A-DIBH. The mean doses to the heart, LADCA and ipsilateral lung in IMRT plans were generated from IMRT with FB, T-DIBH and A-DIBH were (1.96 ± 2.25), (1.37 ± 0.44) and (1.18 ± 0.26) Gy (all P < 0.05), respectively, with FB; (16.10 ± 7.45), (8.6 ± 6.60) and (7.35 ± 5.42) Gy (all P < 0.05), respectively, with T-DIBH; and (5.90 ± 2.24), (5.65 ± 1.58) and (5.62 ± 1.05) Gy (all P > 0.05), respectively, with A-DIBH. CONCLUSIONS: This study indicates that both 3D-CRT and IMRT plans with A-DIBH achieved lower cardiac and LADCA doses than plans with FB and T-DIBH; 3D-CRT plans with A-DIBH achieved lower ipsilateral lung doses than plans with FB and T-DIBH; and IMRT plans with A-DIBH had better outcomes than 3D-CRT plans with A-DIBH with respect to the mean dose to the heart, LADCA and ipsilateral lung. IMRT plans with A-DIBH should be incorporated into the daily routine for left-sided breast RT.