Scholarlink

Objective <p>To develop a novel apparent diffusion coefficient (ADC) and signal intensity (SI) based model to quantify the subtle internal difference between tumor-adjacent (TAL) and -distant liver tissue (TDL) in T<sub>3</sub>-staged resectable gallbladder carcinoma (GBC).</p> Methods <p>65 consecutive patients with T<sub>3</sub>-staged GBC invading liver undergoing preoperative MRI were retrospectively included, among which 54 from hospital 1 were randomly assigned to training (TC, <i>n</i> = 43) and internal validation cohorts (IVC, <i>n</i> = 11), while the remaining 11 from hospital 2 constituted external validation cohort (EVC, <i>n</i> = 11). Mean ADC and its standard deviation (SD) of TAL and TDL were measured on DWI at b-values of 0 and 600&#xa0;s/mm<sup>2</sup>, 0 and 800&#xa0;s/mm<sup>2</sup>, and 0 and 1000&#xa0;s/mm<sup>2</sup>. SIs of TAL, TDL and erector spinae (ES) on T<sub>1</sub>WI, T<sub>2</sub>WI, and arterial, portal-venous and delayed phases enhanced images were measured, and signal intensity ratios (SIRs) of TAL and TDL to ES were calculated. The <i>t</i>-test, Mann-Whitney <i>U</i> test and binary logistic regression analyses were conducted sequentially to determine independent index for differentiating TAL from TDL, and a model was constructed for the differentiation. Predictive value of model was assessed using the receiver operating characteristic (ROC) curve.</p> Results <p>In TC, SIRs on arterial phase (SIR<sub>AP</sub>) and portal-venous phase (SIR<sub>PP</sub>), SIs on portal-venous phase (SI<sub>PP</sub>) and delayed phase (SI<sub>DP</sub>), and SD at b-values of 0 and 1000&#xa0;s/mm<sup>2</sup> (SD<sub>1000</sub>) were independent differentiating indexes with odds ratios of 0.008 (95% confidence interval [CI], 0.001–0.131), 0.132 (95%CI, 0.033–0.533), 1.002 (95%CI, 1.000-1.003), 0.998 (95%CI, 0.997–0.999), and 1.472 (95%CI, 0.006-355.856), respectively. ROC analysis showed that the model by integrating the previous indexes obtained excellent performance with areas under the ROC curve of 0.879, 0.934 and 0.909 in TC, IVC and EVC, respectively.</p> Conclusion <p>The novel model could be helpful for quantifying the subtle difference between TAL and TDL in T<sub>3</sub>-staged GBC.</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

A novel model based on apparent diffusion coefficient and signal intensity on MRI to quantify subtle internal difference between tumor-adjacent and -distant liver tissues in T3-staged resectable gallbladder carcinoma

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月27日

查看全文
  • Zhao Tang,
  • Xiao-fang Zhu,
  • Jing Ou,
  • Yu-ping Wu,
  • Xiao-ming Zhang,
  • Bang-guo Tan,
  • Tian-wu Chen

摘要

Objective

To develop a novel apparent diffusion coefficient (ADC) and signal intensity (SI) based model to quantify the subtle internal difference between tumor-adjacent (TAL) and -distant liver tissue (TDL) in T3-staged resectable gallbladder carcinoma (GBC).

Methods

65 consecutive patients with T3-staged GBC invading liver undergoing preoperative MRI were retrospectively included, among which 54 from hospital 1 were randomly assigned to training (TC, n = 43) and internal validation cohorts (IVC, n = 11), while the remaining 11 from hospital 2 constituted external validation cohort (EVC, n = 11). Mean ADC and its standard deviation (SD) of TAL and TDL were measured on DWI at b-values of 0 and 600 s/mm2, 0 and 800 s/mm2, and 0 and 1000 s/mm2. SIs of TAL, TDL and erector spinae (ES) on T1WI, T2WI, and arterial, portal-venous and delayed phases enhanced images were measured, and signal intensity ratios (SIRs) of TAL and TDL to ES were calculated. The t-test, Mann-Whitney U test and binary logistic regression analyses were conducted sequentially to determine independent index for differentiating TAL from TDL, and a model was constructed for the differentiation. Predictive value of model was assessed using the receiver operating characteristic (ROC) curve.

Results

In TC, SIRs on arterial phase (SIRAP) and portal-venous phase (SIRPP), SIs on portal-venous phase (SIPP) and delayed phase (SIDP), and SD at b-values of 0 and 1000 s/mm2 (SD1000) were independent differentiating indexes with odds ratios of 0.008 (95% confidence interval [CI], 0.001–0.131), 0.132 (95%CI, 0.033–0.533), 1.002 (95%CI, 1.000-1.003), 0.998 (95%CI, 0.997–0.999), and 1.472 (95%CI, 0.006-355.856), respectively. ROC analysis showed that the model by integrating the previous indexes obtained excellent performance with areas under the ROC curve of 0.879, 0.934 and 0.909 in TC, IVC and EVC, respectively.

Conclusion

The novel model could be helpful for quantifying the subtle difference between TAL and TDL in T3-staged GBC.