To address the challenges faced by current brain midline segmentation techniques, such as insufficient accuracy and poor segmentation continuity, this paper proposes a deep learning network model based on a two-stage framework. On the first stage of the model, prior knowledge of the feature consistency of adjacent brain midline slices under normal and pathological conditions is utilized. Associated midline slices are selected through slice similarity analysis, and a novel feature weighting strategy is adopted to collaboratively fuse the overall change characteristics and spatial information of these associated slices, thereby enhancing the feature representation of the brain midline in the intracranial region. On the second stage, the optimal path search strategy for the brain midline is employed based on the network output probability map, which effectively addresses the problem of discontinuous midline segmentation. The method proposed in this paper achieved satisfactory results on the CQ500 dataset provided by the Center for Advanced Research in Imaging, Neurosciences and Genomics, New Delhi, India. The Dice similarity coefficient (DSC), Hausdorff distance (HD), average symmetric surface distance (ASSD), and normalized surface Dice (NSD) were 67.38 ± 10.49, 24.22 ± 24.84, 1.33 ± 1.83, and 0.82 ± 0.09, respectively. The experimental results demonstrate that the proposed method can fully utilize the prior knowledge of medical images to effectively achieve accurate segmentation of the brain midline, providing valuable assistance for subsequent identification of the brain midline by clinicians.
Objective To analyze the disease burden, risk factor attribution, and trends through 2030 of female breast cancer in China from 1990 to 2023. MethodsBased on data from the Global Burden of Disease Study 2023, the average annual percent change (AAPC) was used to evaluate long-term trends. An age-period-cohort (APC) model was applied to assess age, period, and cohort effects. Population attributable fractions, attributable disability-adjusted life years (DALYs), and decomposition analysis were used to evaluate the contributions of risk factors. A Bayesian APC model was used to project trends from 2024 to 2030. ResultsFrom 1990 to 2023, the age-standardized incidence rate of female breast cancer in China increased from 24.54 per 100 000 to 31.62 per 100 000, whereas the age-standardized mortality rate decreased from 10.65 per 100 000 to 6.74 per 100 000 and the age-standardized DALY rate decreased from 372.92 per 100 000 to 228.43 per 100 000. The AAPCs in incidence rates were negative among women aged 15–39 years (P<0.05) and became positive after 45 years of age (P<0.05), with the largest increase observed in the 85–89 year age group (AAPC=1.72%). In the ≥95 year age group, mortality, DALY, and years of life lost rates all increased, with AAPCs of 0.58%, 0.52%, and 0.52%, respectively (P<0.05). APC analysis showed that the net drift in age-specific incidence rates was 0.315% per year (P<0.001), whereas the net drift in age-specific mortality rates was ?2.316% per year (P<0.001). In 2023, the leading attributable risk factors were high red meat intake, secondhand smoke, high fasting plasma glucose, high body mass index, and low physical activity. Bayesian APC projections indicated that the age-standardized incidence, mortality, and DALY rates would reach 38.91 per 100 000, 7.84 per 100 000, and 244.24 per 100 000, respectively, by 2030. ConclusionsThe incidence burden of female breast cancer in China continued to increase, whereas the age-standardized mortality and premature mortality burdens declined over the long term but showed a recent rebound. The disease burden increasingly shifted toward older women, and the overall burden is expected to remain substantial through 2030.