Joint estimation of relaxation and diffusion tissue parameters for prostate cancer grading with relaxation-VERDICT MRI
Joint estimation of relaxation and diffusion tissue parameters for prostate cancer grading with relaxation-VERDICT MRI
复制标题
使用松弛-VERDICT MRI 联合估计松弛和扩散组织参数以进行前列腺癌分级
DOI：
10.1101/2021.06.24.21259440
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
E. Panagiotaki
中科院分区：
文献类型：
--
作者：
M. Palombo;V. Valindria;S. Singh;E. Chiou;F. Giganti;H. Pye;H. Whitaker;D. Atkinson;S. Punwani;D. Alexander;E. Panagiotaki
目标。用于肿瘤细胞术的血管、细胞外和限制扩散（VERDICT）是一种非侵入性MRI技术，在临床环境中区分正常组织和恶性组织以及Gleason分级3+3和3+4显示出令人鼓舞的结果。然而，VERDICT目前并没有考虑到组织的固有松弛特性，其量化可以提供额外的信息并提高其诊断能力。这项工作的目的是为前列腺引入松弛-判决（rVERDICT），这是一个模型，用于联合估计基于扩散的参数（如细胞内体积分数，fic）和松弛时间（如T2）从VERDICT MRI采集；并评估其可重复性和区分Gleason分级3和4的诊断效用。材料与方法：招募了72名疑似前列腺癌（PCa）或正在接受主动监测的男性。所有患者均行多参数MRI （mp-MRI）和VERDICT MRI检查。采用深度神经网络对rVERDICT参数进行超快速拟合。44名男性接受了靶向活检，从而能够评估rVERDICT参数区分Gleason分级的准确性、f1评分和卷积神经网络分类器的Cohen kappa。为了评估新模型的可重复性，对五名男性进行了两次成像。结果。rVERDICT细胞内体积分数区分Gleason分级（5级分类）的准确性，f1评分和kappa评分比经典VERDICT高8,7和3个百分点，比mp-MRI的表观扩散系数（ADC）高12,13和24个百分点。rVERDICT参数重复性高（R2=0.74 ~ 0.99，变异系数=1% ~ 10%，类内相关系数=78% ~ 98%）。rVERDICT估计的T2值与独立多te采集估计的T2值无显著差异（p < 0.05）。深度神经网络拟合方法提供了所有rVERDICT参数的稳定拟合，并大大减少了处理时间（~35秒，而使用经典的VERDICT则需要~15分钟），从而能够实时生成rVERDICT地图。结论。新的rVERDICT允许对PCa的扩散和弛豫特性进行稳健和超快速的微观结构估计，并实现Gleason评分。
Objectives. The Vascular, Extracellular, and Restricted Diffusion for Cytometry in Tumours (VERDICT) is a non-invasive MRI technique that has shown promising results in clinical settings discriminating normal from malignant tissue and Gleason grade 3+3 from 3+4. However, VERDICT currently does not account for the inherent relaxation properties of the tissue, whose quantification could provide additional information and enhance its diagnostic power. The aim of this work is to introduce relaxation-VERDICT (rVERDICT) for prostate, a model for the joint estimation of diffusion-based parameters (e.g. the intracellular volume fraction, fic) and relaxation times (e.g. T2) from a VERDICT MRI acquisition; and to evaluate its repeatability and diagnostic utility for differentiating Gleason grades 3 and 4. Materials and Methods. 72 men with suspected prostate cancer (PCa) or undergoing active surveillance were recruited. All men underwent multiparametric MRI (mp-MRI) and VERDICT MRI. Deep neural network was used for ultra-fast fitting of the rVERDICT parameters. 44 men underwent targeted biopsy, which enabled assessment of rVERDICT parameters in differentiating Gleason grades measured with accuracy, F1-score and Cohen's kappa of a convolutional neural network classifier. To assess the repeatability of the new model, five men were imaged twice. Results. The rVERDICT intracellular volume fraction fic discriminated between Gleason grades (5-class classification) with accuracy, F1-score and kappa 8, 7 and 3 percentage points higher than classic VERDICT, and 12, 13 and 24 percentage points higher than the Apparent Diffusion Coefficient (ADC) from mp-MRI. Repeatability of rVERDICT parameters was high (R2=0.74-0.99, coefficient of variation=1%-10% and intraclass correlation coefficient=78%-98%). T2 values estimated with rVERDICT were not significantly different from those estimated with an independent multi-TE acquisition (p>0.05). The deep neural network fitting approach provided stable fitting of all the rVERDICT parameters with dramatic reduction of the processing time (~35 seconds vs ~15 minutes using classic VERDICT), enabling on-the-fly rVERDICT map generation. Conclusions. The new rVERDICT allows for robust and ultra-fast microstructural estimation of diffusion and relaxation properties of PCa and enables Gleason scoring.
登录
查看更多内容
影响因子：
11.2
作者：
White NS;McDonald C;Farid N;Kuperman J;Karow D;Schenker-Ahmed NM;Bartsch H;Rakow-Penner R;Holland D;Shabaik A;Bjørnerud A;Hope T;Hattangadi-Gluth J;Liss M;Parsons JK;Chen CC;Raman S;Margolis D;Reiter RE;Marks L;Kesari S;Mundt AJ;Kane CJ;Carter BS;Bradley WG;Dale AM
通讯作者：
Dale AM
影响因子：
19.7
作者：
Yu, Alice C.;Badve, Chaitra;Gulani, Vikas
通讯作者：
Gulani, Vikas
影响因子：
2.5
作者：
Grgac K;Li W;Huang A;Qin Q;van Zijl PC
通讯作者：
van Zijl PC
影响因子：
23.4
作者：
Weinreb JC;Barentsz JO;Choyke PL;Cornud F;Haider MA;Macura KJ;Margolis D;Schnall MD;Shtern F;Tempany CM;Thoeny HC;Verma S
通讯作者：
Verma S
影响因子：
45.3
作者：
Stark, Jennifer R.;Perner, Sven;Mucci, Lorelei A.
通讯作者：
Mucci, Lorelei A.