Scholarlink

<p>Left ventricular ejection fraction (LVEF) is cardiovascular function's most important clinical parameter. The accuracy in estimating this parameter depends on precisely segmenting the left ventricle (LV) structure at the end-diastole and -systole phases. Therefore, developing robust algorithms for precisely segmenting the heart structure during different phases is crucial. In this work, an improved 3D UNet model is proposed to segment the LV and myocardium while excluding papillary muscles, as per the recommendation of the Society for Cardiovascular Magnetic Resonance. For the practical testing of the proposed framework, 8,400 cardiac MRI images were collected and analyzed from the military hospital in Tunis (HMPIT) and the popular ACDC public dataset. The Dice coefficient and the F1 score were used as performance metrics to validate the LV and myocardium segmentations. The data was split into 70%, 10%, and 20% for training, validation, and testing, respectively. It is worth noting that the proposed segmentation model was tested across three axis views: basal, medio basal and apical, at two different cardiac phases: end-diastole and -systole instances. The experimental results showed a Dice coefficient of 0.965 and 0.945 and an F1 score of 0.801 and 0.799 at the end-diastolic and -systolic phases, respectively. The clinical evaluation outcomes revealed a significant difference in the LVEF and other clinical parameters when the papillary muscles were included or excluded. The proposed framework outperforms state-of-the-art methods by around 0.1 in terms of Dice coefficient, demonstrating its accuracy in assessing the left ventricular function precisely.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An Improved Approach for Cardiac MRI Segmentation based on a 3D UNet Combined with Papillary Muscle Exclusion

  • 期刊论文

  • 发表日期：

    2025年5月24日

查看全文
  • Narjes Benameur,
  • Ramzi Mahmoudi,
  • Mohamed Deriche,
  • Amira Fayouka,
  • Imene Masmoudi,
  • Nessrine Zoghlami

摘要

Left ventricular ejection fraction (LVEF) is cardiovascular function's most important clinical parameter. The accuracy in estimating this parameter depends on precisely segmenting the left ventricle (LV) structure at the end-diastole and -systole phases. Therefore, developing robust algorithms for precisely segmenting the heart structure during different phases is crucial. In this work, an improved 3D UNet model is proposed to segment the LV and myocardium while excluding papillary muscles, as per the recommendation of the Society for Cardiovascular Magnetic Resonance. For the practical testing of the proposed framework, 8,400 cardiac MRI images were collected and analyzed from the military hospital in Tunis (HMPIT) and the popular ACDC public dataset. The Dice coefficient and the F1 score were used as performance metrics to validate the LV and myocardium segmentations. The data was split into 70%, 10%, and 20% for training, validation, and testing, respectively. It is worth noting that the proposed segmentation model was tested across three axis views: basal, medio basal and apical, at two different cardiac phases: end-diastole and -systole instances. The experimental results showed a Dice coefficient of 0.965 and 0.945 and an F1 score of 0.801 and 0.799 at the end-diastolic and -systolic phases, respectively. The clinical evaluation outcomes revealed a significant difference in the LVEF and other clinical parameters when the papillary muscles were included or excluded. The proposed framework outperforms state-of-the-art methods by around 0.1 in terms of Dice coefficient, demonstrating its accuracy in assessing the left ventricular function precisely.