OBJECTIVE: This review aimed to identify self-care approaches, domains, and their effectiveness for a proper self-care educational guide package for women with GDM.
DESIGN: A systematic review using electronic literature databases published between January 2016 and December 2022 was conducted.
DATA SOURCES: Web of Science, Scopus, and Ovid databases were used.
REVIEW METHODS: This review utilized the PICO (Population, Intervention, Comparison, and Outcomes) framework to screen the retrieved articles for eligibility in which mothers with GDM, educational materials, standard practice or intervention, and effectiveness were considered the PICO, respectively. The CIPP (Context, Input, Process, Product) model served as a framework for adopting the education development model. Mixed methods appraisal tool was used for quality assessment. Data extraction and synthesis without meta-analysis were presented as evidence tables.
RESULTS: A total of 19 articles on GDM were included in the final analysis (16 Intervention studies, two qualitative studies, and one mixed-methods study). Four broad domains emerged from the analysis: 1) information or knowledge of GDM, 2) monitoring of blood glucose levels, 3) practice of healthy lifestyles, and 4) other non-specific activities. The majority of the articles employed a face-to-face approach in executing the educational group sessions, and most studies disclosed their positive effects on GDM management. Other methods of evaluating intervention effectiveness were described as improved self-care behavior, increased satisfaction score, enhanced self-efficacy, good glucose control, and better pregnancy outcome.
CONCLUSION: Knowledge or information about GDM, healthy diet, and exercise or physical activity was found to be the most applied domains of intervention. Framework domains based on the present review can be used in the future development of any interventional program for GDM women in enhancing health information reaching the targeted group in promoting self-efficacy.
PROSPERO REGISTRATION NUMBER: CRD42021229610.
METHODS: This study included 100 patients diagnosed with prostate adenocarcinoma who had undergone multi-parametric MRI and prostatectomy. From the T2-weighted MR images, ground truth segmentation masks were established by consensus from two expert radiologists. The prostate was then automatically contoured with six different methods: (1) a multi-atlas algorithm, (2) a proprietary algorithm in the Syngo.Via medical imaging software, and four deep learning models: (3) a V-net trained from scratch, (4) a pre-trained 2D U-net, (5) a GAN extension of the 2D U-net, and (6) a segmentation-adapted EfficientDet architecture. The resulting segmentations were compared and scored against the ground truth masks with one 70/30 and one 50/50 train/test data split. We also analyzed the association between segmentation performance and clinical variables.
RESULTS: The best performing method was the adapted EfficientDet (model 6), achieving a mean Dice coefficient of 0.914, a mean absolute volume difference of 5.9%, a mean surface distance (MSD) of 1.93 pixels, and a mean 95th percentile Hausdorff distance of 3.77 pixels. The deep learning models were less prone to serious errors (0.854 minimum Dice and 4.02 maximum MSD), and no significant relationship was found between segmentation performance and clinical variables.
CONCLUSIONS: Deep learning-based segmentation techniques can consistently achieve Dice coefficients of 0.9 or above with as few as 50 training patients, regardless of architectural archetype. The atlas-based and Syngo.via methods found in commercial clinical software performed significantly worse (0.855[Formula: see text]0.887 Dice).