Objective We compared the image quality and diagnostic performance of deep learning reconstruction (DLR) accelerated reduced field-of-view (rFOVDL) diffusion-weighted imaging (DWI) with standard-reconstructed full field-of-view (fFOVSTA) DWI in rectal cancer. Materials and methods This prospective study enrolled 173 participants with biopsy-confirmed rectal adenocarcinoma from November 2022 to August 2023 undergoing rFOVDL and fFOVSTA DWI scans. Two radiologists evaluated qualitative image quality, objective image quality, and apparent diffusion coefficient (ADC) independently. T and N staging were evaluated in 94 participants undergoing radical surgery. Diagnostic sensitivity, specificity, and accuracy were calculated using histopathologic results as the gold standard. ADC values were analyzed for correlations with histopathologic staging. Results We observed that rFOVDL DWI reduced acquisition time by 30% compared to fFOVSTA DWI. rFOVDL DWI outperformed fFOVSTA DWI in all qualitative image quality metrics (p ≤ 0.013), especially in mucosa-submucosa-muscularis visualization, spatial resolution, overall image quality, and diagnostic confidence, accompanied by comparable objective image quality (p ≥ 0.054). When applied with T2-weighted imaging, rFOVDL DWI significantly enhanced primary T-staging accuracy than fFOVSTA DWI (p < 0.001), especially for early-stage tumors (T1 or T2). Tumor ADC values of rFOVDL DWI were lower than those of fFOVSTA DWI, yet remained solid inverse correlations with histopathologic T-staging (p < 0.001). Higher inter-reader agreements of locoregional staging and ADC measurements were obtained by rFOVDL DWI. Conclusion rFOVDL DWI significantly improved image quality than fFOVSTA DWI, with a 30% reduced acquisition time. rFOVDL DWI facilitated higher primary T-staging accuracy, especially for early-stage rectal cancer (T1–T2). Relevance statement Reduced acquisition time and improved imaging quality highlighted the clinical feasibility of applying DLR to rFOV DWI. rFOVDL DWI could significantly enhance primary T-staging accuracy, especially for early-stage rectal cancer (T1–T2), facilitating more precise treatment management. Key Points <UnorderedList Mark="Bullet"> <ItemContent> Applying deep learning reconstruction (DLR) to reduced field-of-view (rFOV) diffusion-weighted imaging (DWI) improved mucosa-submucosa-muscularis visualization and reduced acquisition time. </ItemContent> <ItemContent> DLR-based rFOV DWI significantly enhanced primary T-staging accuracy for rectal cancer, especially for early-stage tumors (T1 or T2). </ItemContent> <ItemContent> DLR-based rFOV DWI facilitated higher inter-reader agreements for locoregional staging and apparent diffusion coefficient measurement in rectal cancer. </ItemContent> </UnorderedList> Graphical Abstract

Deep learning reconstruction accelerated reduced field-of-view DWI in rectal cancer: mucosa-submucosa-muscularis visualization and T staging

期刊论文
出版模式：

开放获取
发表日期：

2026年1月26日

查看全文

Wenjing Peng,
Fan Yang,
Diliang Li,
Rui Zhao,
Lijuan Wan,
Shuang Chen,
Xiangchun Liu,
Sicong Wang,
Yuanlong Li,
Min Li,
Yuan Liu,
Hongmei Zhang

摘要

Objective

We compared the image quality and diagnostic performance of deep learning reconstruction (DLR) accelerated reduced field-of-view (rFOV_DL) diffusion-weighted imaging (DWI) with standard-reconstructed full field-of-view (fFOV_STA) DWI in rectal cancer.

Materials and methods

This prospective study enrolled 173 participants with biopsy-confirmed rectal adenocarcinoma from November 2022 to August 2023 undergoing rFOV_DL and fFOV_STA DWI scans. Two radiologists evaluated qualitative image quality, objective image quality, and apparent diffusion coefficient (ADC) independently. T and N staging were evaluated in 94 participants undergoing radical surgery. Diagnostic sensitivity, specificity, and accuracy were calculated using histopathologic results as the gold standard. ADC values were analyzed for correlations with histopathologic staging.

Results

We observed that rFOV_DL DWI reduced acquisition time by 30% compared to fFOV_STA DWI. rFOV_DL DWI outperformed fFOV_STA DWI in all qualitative image quality metrics (p ≤ 0.013), especially in mucosa-submucosa-muscularis visualization, spatial resolution, overall image quality, and diagnostic confidence, accompanied by comparable objective image quality (p ≥ 0.054). When applied with T2-weighted imaging, rFOV_DL DWI significantly enhanced primary T-staging accuracy than fFOV_STA DWI (p < 0.001), especially for early-stage tumors (T1 or T2). Tumor ADC values of rFOV_DL DWI were lower than those of fFOV_STA DWI, yet remained solid inverse correlations with histopathologic T-staging (p < 0.001). Higher inter-reader agreements of locoregional staging and ADC measurements were obtained by rFOV_DL DWI.

Conclusion

rFOV_DL DWI significantly improved image quality than fFOV_STA DWI, with a 30% reduced acquisition time. rFOV_DL DWI facilitated higher primary T-staging accuracy, especially for early-stage rectal cancer (T1–T2).

Relevance statement

Reduced acquisition time and improved imaging quality highlighted the clinical feasibility of applying DLR to rFOV DWI. rFOV_DL DWI could significantly enhance primary T-staging accuracy, especially for early-stage rectal cancer (T1–T2), facilitating more precise treatment management.

Key Points

Applying deep learning reconstruction (DLR) to reduced field-of-view (rFOV) diffusion-weighted imaging (DWI) improved mucosa-submucosa-muscularis visualization and reduced acquisition time.

DLR-based rFOV DWI significantly enhanced primary T-staging accuracy for rectal cancer, especially for early-stage tumors (T1 or T2).

DLR-based rFOV DWI facilitated higher inter-reader agreements for locoregional staging and apparent diffusion coefficient measurement in rectal cancer.

Graphical Abstract