Abstract
ObjectiveIn this paper, we studied the feasibility of automatic segmentation of bone structures for rehabilitation and disease prevention from diffusion-weighted and apparent diffusion coefficient images obtained from prostate multi-parameter magnetic resonance imaging (mpMRI) in the healthy promotion of Chinese Taijiquan martial arts. MethodsThe mpMRI images of 15 patients practicing Taijiquan martial arts were analyzed retrospectively. We manually annotate bone structures on DWI (b = 800 s/mm2), DWI (b = 0S/mm2), and ADC images. Then, we use different sequence combinations as input data to test the segmentation model, and evaluate the impact of six different sequence combinations on the region-based segmentation performance, such as the watershed model. The model evaluation indicators include quantitative indicators (DICE coefficient, label capacity) and qualitative indicators (subjective score). The model evaluation standard calculates the coincidence rate of all sequences in the test set, and more than 80% are considered to meet the clinical application requirements. ResultsThe DICE value of the watershed segmentation model was 0.75 (0.70–0.81) −0.81 (0.73–0.85) on DWI images, and the ADC value was 0.79 (0.78–0.81) −0.83 (0.80–0.85). However, there was no significant difference in DICE value between different models (HDWI = 2.978, PDWI>0.05; HADC = 1.140, PADC>0.05). There was no significant difference in the volume difference between model prediction and manual labeling among different models (HDWI = 2.900, PDWI>0.05; HADC = 2.236, PADC>0.05). Qualitative scoring models 1 and 3 achieved the highest standard rate in DWI and ADC image segmentation, both above 80%. ConclusionAfter Taijiquan martial arts exercise, it is found that the DWI segmentation model based on watershed region combined with ADC sequence can achieve high-performance segmentation of pelvic bone structure in prostate mpMRI, meet the needs of clinical application, and is conducive to healthy sports for all.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: Journal of Radiation Research and Applied Sciences
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.