|
This paper discusses the crucial role of endometrial glands in early human pregnancy and the potential applications of medical information technology in their identification and analysis. The glands serve as a source of nutrients, growth factors, and immunoregulatory proteins, which maintain fertilized embryos and facilitate maternal-fetal transport. The document highlights the production of glycodelin A and growth factors such as EGF and LIF by the glands, which regulate placental morphogenesis and modulate the interactions between uNK cells and extravillous trophoblast cells. The use of medical imaging techniques and machine learning algorithms can enhance diagnostic accuracy and efficiency in identifying and analyzing these glands. Challenges remain in acquiring high-quality data and integrating medical information technology with traditional clinical expertise. The paper concludes with the proposal of a novel glandular ostium labeling algorithm based on medical information technology for identifying and analyzing endometrial-decidual glands. The algorithm is effective in removing uneven brightness backgrounds, extracting potential glandular ostium regions, enhancing glandular openings, and labeling them using the regional maximum value method in mathematical morphology. The labeling algorithm showed a sensitivity of 84.72%±8.00% and a Dice similarity coefficient of 83.80%±4.56% in evaluating 32 endometrial images.
|
