An artificial intelligence system using maximum intensity projection MR images facilitates classification of non-mass enhancement breast lesions

Wang, Lijun; Chang, Lufan; Luo, Ran; Cui, Xuee; Liu, Huanhuan; Wu, Haoting; Chen, Yanhong; Zhang, Yuzhen; Wu, Chao; Li, Fangzhen; Líu, Hao; Guan, Wenbin; Wang, Dengbin

doi:10.1007/s00330-022-08553-5

Cited by 17 publications

(13 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our study, a shorter VDT was observed in breast cancer manifesting as a focus or mass lesion when compared to an NME lesion. Because the majority of malignant focus or mass lesions represented invasive breast cancer, only half of malignant NME lesions were invasive breast cancer, which was in line with previous reports 27,28 . Given the shortest VDT of 68 days, breast cancer presented as a spherical focus or mass with a diameter ≤5 mm would still be minimal breast cancer after 6 months but could evolve to the T1c stage after 1 year.…”

Section: Discussionsupporting

confidence: 89%

“…Because the majority of malignant focus or mass lesions represented invasive breast cancer, only half of malignant NME lesions were invasive breast cancer, which was in line with previous reports. 27,28 Given the shortest VDT of 68 days, breast cancer presented as a spherical focus or mass with a diameter ≤5 mm would still be minimal breast cancer after 6 months but could evolve to the T1c stage after 1 year. Edmonds et al reported that most breast cancer lesions that had been initially categorized as BI-RADS 3 progressed within 6 months; therefore, a 6-month follow-up was recommended for these lesions to avoid unnecessary biopsies.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Breast Cancer Growth on Serial MRI: Volume Doubling Time Based on 3‐Dimensional Tumor Volume Assessment

Wang

Luo

Chen

et al. 2023

Magnetic Resonance Imaging

Self Cite

View full text Add to dashboard Cite

BackgroundThe volume doubling time (VDT) of breast cancer was most frequently calculated using the two‐dimensional (2D) diameter, which is not reliable for irregular tumors. It was rarely investigated using three‐dimensional (3D) imaging with tumor volume on serial magnetic resonance imaging (MRI).PurposeTo investigate the VDT of breast cancer using 3D tumor volume assessment on serial breast MRIs.Study TypeRetrospective.SubjectsSixty women (age at diagnosis: 57 ± 10 years) with breast cancer, assessed by two or more breast MRI examinations. The median interval time was 791 days (range: 70–3654 days).Field Strength/Sequence3‐T, fast spin‐echo T2‐weighted imaging (T2WI), single‐shot echo‐planar diffusion‐weighted imaging (DWI), and gradient echo dynamic contrast‐enhanced imaging.AssessmentThree radiologists independently reviewed the morphological, DWI, and T2WI features of lesions. The whole tumor was segmented to measure the volume on contrast‐enhanced images. The exponential growth model was fitted in the 11 patients with at least three MRI examinations. The VDT of breast cancer was calculated using the modified Schwartz equation.Statistical TestsMann–Whitney U test, Kruskal–Wallis test, Chi‐squared test, intraclass correlation coefficients, and Fleiss kappa coefficients. A P‐value <0.05 was considered statistically significant. The exponential growth model was evaluated using the adjusted R2 and root mean square error (RMSE).ResultsThe median tumor diameter was 9.7 mm and 15.2 mm on the initial and final MRI, respectively. The median adjusted R2 and RMSE of the 11 exponential models were 0.97 and 15.8, respectively. The median VDT was 540 days (range: 68–2424 days). For invasive ductal carcinoma (N = 33), the median VDT of the non‐luminal type was shorter than that of the luminal type (178 days vs. 478 days). On initial MRI, breast cancer manifesting as a focus or mass lesion showed a shorter VDT than that of a non‐mass enhancement (NME) lesion (median VDT: 426 days vs. 665 days).Data ConclusionA shorter VDT was observed in breast cancer manifesting as focus or mass as compared to an NME lesion.Level of Evidence3Technical EfficacyStage 2

show abstract

Section: Discussionsupporting

confidence: 89%

Section: Discussionmentioning

confidence: 99%

Breast Cancer Growth on Serial MRI: Volume Doubling Time Based on 3‐Dimensional Tumor Volume Assessment

Wang

Luo

Chen

et al. 2023

Magnetic Resonance Imaging

Self Cite

View full text Add to dashboard Cite

show abstract

“…Thirteen articles described 38 datasets that used the DCE sequence. Ten studies reported 10 datasets evaluating segmental features as the index test ( 5 , 11 , 13 , 28 – 32 , 34 , 35 ). Four datasets assessed heterogeneous characteristics ( 5 , 13 , 32 , 33 ), whereas six datasets used clumped characteristics ( 5 , 11 , 30 – 33 ).…”

Section: Resultsmentioning

confidence: 99%

“…Furthermore, six datasets used the washout/plateau TIC ( 11 , 13 , 19 , 30 – 32 ), while four datasets used DCE combined with DWI as the index test ( 11 , 12 , 14 , 19 ). Twenty-nine studies ( 5 , 12 , 15 – 17 , 28 – 32 , 34 , 36 ) utilized 1.5-T MRI, five studies ( 13 , 14 , 19 , 33 , 35 ) used 3.0-T MRI, and one study used 1.5-T and/or 3.0-T MRI ( 26 ). Half of the included studies were from Asian countries ( 5 , 11 , 12 , 16 , 19 , 30 , 33 – 35 ).…”

Section: Resultsmentioning

confidence: 99%

Meta-analysis of dynamic contrast enhancement and diffusion-weighted MRI for differentiation of benign from malignant non-mass enhancement breast lesions

Zhang,

Li,

Zhang

et al. 2024

Front. Oncol.

View full text Add to dashboard Cite

PurposeThe objective of this study was to conduct a meta-analysis comparing the diagnostic efficacy of models based on diffusion-weighted imaging (DWI)-MRI, dynamic contrast enhancement (DCE)-MRI, and combination models (DCE and DWI) in distinguishing benign from malignant non-mass enhancement (NME) breast lesions.Materials and methodsPubMed, Embase, and Cochrane Library were searched, from inception to January 30, 2023, for studies that used DCE or DWI-MRI for the prediction of NME breast cancer patients. A bivariate random-effects model was used to calculate the meta-analytic sensitivity, specificity, and area under the curve (AUC) of the DCE, DWI, and combination models. Subgroup analysis and meta-regression analysis were performed to find the source of heterogeneity.ResultsOf the 838 articles screened, 18 were eligible for analysis (13 on DCE, five on DWI, and four studies reporting the diagnostic accuracy of both DCE and DWI). The funnel plot showed no publication bias (p > 0.5). The pooled sensitivity and specificity and the AUC of the DCE, DWI, and combination models were 0.58, 0.72, and 0.70, respectively; 0.84, 0.69, and 0.84, respectively; and 0.88, 0.79, 0.90, respectively. The meta-analysis found no evidence of a threshold effect and significant heterogeneity among trials in terms of DCE sensitivity and specificity, as well as DWI specificity alone (I2 > 75%). The meta-regression revealed that different diagnostic criteria contributed to the DCE study’s heterogeneity (p < 0.05). Different reference criteria significantly influenced the heterogeneity of the DWI model (p < 0.05). Subgroup analysis revealed that clustered ring enhancement (CRE) had the highest pooled specificity (0.92) among other DCE features. The apparent diffusion coefficient (ADC) with a mean threshold <1.3 × 10−3 mm2/s had a slightly higher sensitivity of 0.86 compared to 0.82 with an ADC of ≥1.3 × 10−3 mm2/s.ConclusionThe combination model (DCE and DWI) outperformed DCE or DWI alone in identifying benign and malignant NME lesions. The DCE-CRE feature was the most specific test for ruling in NME cancers.

show abstract

“…They find out that operating CNN on MIP can achieve up to 98% AUC. Wang et al [18] applied ResNet-50 with the axial and sagittal MIP of early post-contrast subtracted breast MRI images. Their system achieved comparable diagnostic performance with the senior radiologist and outperformed the junior radiologist.…”

Section: Related Workmentioning

confidence: 99%

DEA-UNet: a dense-edge-attention UNet architecture for medical image segmentation

et al. 2022

View full text Add to dashboard Cite

CNNs have been widely applied for medical image analysis. However, limited memory capacity is one of the most common drawbacks of processing high-resolution 3D volumetric data. 3D volumes are usually cropped or downsized first before processing, which can result in a loss of resolution, increase class imbalance, and affect the performance of the segmentation algorithms. In this paper, we propose an end-to-end deep learning approach called IP-UNet. IP-UNet is a UNet-based model that performs multi-class segmentation on Intensity Projection (IP) of 3D volumetric data instead of the memory-consuming 3D volumes. IP-UNet uses limited memory capability for training without losing the original 3D image resolution. We compare the performance of three models in terms of segmentation accuracy and computational cost: 1) Slice-by-slice 2D segmentation of the CT scan images using a conventional 2D UNet model. 2) IP-UNet that operates on data obtained by merging the extracted Maximum Intensity Projection (MIP), Closest Vessel Projection (CVP), and Average Intensity Projection (AvgIP) representations of the source 3D volumes, then applying the UNet model on the output IP images. 3) 3D-UNet model directly reads the 3D volumes constructed from a series of CT scan images and outputs the 3D volume of the predicted segmentation. We test the performance of these methods on 3D volumetric images for automatic breast calcification detection. Experimental results show that IP-Unet can achieve similar segmentation accuracy with 3D-Unet but with much better performance. It reduces the training time by 70% and memory consumption by 92%.

show abstract

An artificial intelligence system using maximum intensity projection MR images facilitates classification of non-mass enhancement breast lesions

Cited by 17 publications

References 24 publications

Breast Cancer Growth on Serial MRI: Volume Doubling Time Based on 3‐Dimensional Tumor Volume Assessment

Breast Cancer Growth on Serial MRI: Volume Doubling Time Based on 3‐Dimensional Tumor Volume Assessment

Meta-analysis of dynamic contrast enhancement and diffusion-weighted MRI for differentiation of benign from malignant non-mass enhancement breast lesions

DEA-UNet: a dense-edge-attention UNet architecture for medical image segmentation

Contact Info

Product

Resources

About