中文摘要：

肺部电阻抗成像（EIT）通过配置于人体体表一组阵列电极，利用边界测量信息重建胸腔内部二维断面电导率分布，具有非侵入、无辐射等特点，可用于临床监护。针对EIT逆问题求解的欠定性和病态性，提出一种基于总变差正则化（TV）的两步迭代（TWIST）算法。该算法利用迭代引入TV去噪算子，达到解的双重正则化效果。通过仿真构建不同程度肺痿陷EIT模型，利用该算法进行呼吸状态差分图像再构；同时基于图像提出肺通气总量指数作为客观评价指标。结果表明，与传统Tikhonov正则化算法相比，该算法可达到较好的图像重建质量和鲁棒性，其相应肺通气量指数也更接近于仿真肺痿陷模型变化。从而验证了该算法用于临床肺痿陷EIT通气量监测的可行性。

英文摘要：

Thoracic electrical impedance tomography （EIT） is a non-invasive, radiation-free monitoring technique for reconstruction of the cross-seetional image of the internal spatial distribution of eonductivity from electrical measurements made by injecting small alternating currents via an electrode array placed on the surface of the thorax. To solve the ill-posed problem of EIT, a two-step iterative shrinkage/threshholding algorithm （TWIST） based on total variation （TV） regularization was proposed in this paper. This algorithm used an iterative method with TV denoising operator, so it can achieve double regularization effect. Three dimensional （3D） EIT models of collapsed lung under different conditions were built up, and this algorithm was adopted to reconstruct differential images of respiratory status. And EIT-derived numeric index was also employed for evaluation of lung ventilation. The results showed that the TV-TwlST method can further improve image quality and stability compared with traditional Tikhonov regularization method. And the corresponding evaluation index based on TV-TwlST algorithm also behaved better than that based on Tikhonov regularization. Therefore, the results verified the feasibility of this algorithm applied to clinical use for ventilation monitoring of patients with collapsed-lung.