Stacked dilated convolutions and asymmetric architecture for U-Net-based medical image segmentation

Wang, Shuhang; Singh, Vivek Kumar; Cheah, Eugene; Wang, Xiaohong; Li, Qian; Chou, Shinn-Huey S.; Lehman, Constance D.; Kumar, Viksit; Samir, Anthony E.

doi:10.1016/j.compbiomed.2022.105891

Cited by 16 publications

(7 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, deep learning has provided promising solutions for many data-driven clinical challenges (Protonotarios et al 2022, Wang et al 2022a, Luan et al 2021. As a result, an increasing number of deep learning-based commercial artificial intelligence software (DL-CAIS) are being introduced into clinical practice.…”

Section: Introductionmentioning

confidence: 99%

A multi-modal vision-language pipeline strategy for contour quality assurance and adaptive optimization

Luan,

Ou-yang,

Yang

et al. 2024

Phys. Med. Biol.

View full text Add to dashboard Cite

Objective: Accurate delineation of organs-at-risk (OARs) is a critical step in radiotherapy. The deep learning generated segmentations usually need to be reviewed and corrected by oncologists manually, which is time-consuming and operator-dependent. Therefore, an automated quality assurance (QA) and adaptive optimization correction strategy was proposed to identify and optimize “incorrect” auto-segmentations. Approach: A total of 586 CT images and labels from nine institutions were used. The OARs included the brainstem, parotid, and mandible. The deep learning generated contours were compared with the manual ground truth delineations. In this study, we proposed a novel Contour Quality Assurance and Adaptive Optimization (CQA-AO) strategy, which consists of the following three main components: 1) The contour QA module classified the deep learning generated contours as either accepted or unaccepted; 2) The unacceptable contour categories analysis module provided the potential error reasons (five unacceptable category) and locations (attention heatmaps); 3) The adaptive correction of unacceptable contours module integrate vision-language representations and utilize convex optimization algorithms to achieve adaptive correction of “incorrect” contours. Main results: In the contour quality assurance tasks, the sensitivity (accuracy, precision) of CQA-AO strategy reached 0.940 (0.945, 0.948), 0.962 (0.937, 0.913), and 0.967 (0.962, 0.957) for brainstem, parotid and mandible, respectively. The unacceptable contour category analysis, the (F_I,〖Acc〗_I,F_micro,F_macro) of CQA-AO strategy reached (0.901,0.763,0.862,0.822), (0.855,0.737, 0.837, 0.784), and (0.907, 0.762, 0.858, 0.821) for brainstem, parotid and mandible, respectively. After adaptive optimization correction, the DSC values of brainstem, parotid and mandible have been improved by 9.4%, 25.9%, and 13.5%, and HD values decreased by 62%, 70.6%, and 81.6%, respectively. Significance: The proposed CQA-AO strategy, which combines quality assurance of contour and adaptive optimization correction for OARs contouring, demonstrated superior performance compare to conventional methods. This method can be implemented in the clinical contouring procedures and improve the efficiency of delineating and reviewing workflow.

show abstract

Section: Introductionmentioning

confidence: 99%

A multi-modal vision-language pipeline strategy for contour quality assurance and adaptive optimization

Luan,

Ou-yang,

Yang

et al. 2024

Phys. Med. Biol.

View full text Add to dashboard Cite

show abstract

“…According to the statistics of the Grand Challenges competition in biomedical image analysis, there were 10, 14 and 13 tasks related to image segmentation in 2018, 2019 and 2020, respectively [ 1 ]. Deep learning has achieved huge success in image processing due to its accuracy and generality [ 2 , 3 , 4 , 5 , 6 ]. That is why most biomedical annotation tasks employ deep learning segmentation models such as U-Net [ 7 ], SegNet [ 8 ], DeepLab v3+ [ 9 ], etc.…”

Section: Introductionmentioning

confidence: 99%

BézierSeg: Parametric Shape Representation for Fast Object Segmentation in Medical Images

Haichou

Deng

et al. 2023

Life

View full text Add to dashboard Cite

Background: Delineating the lesion area is an important task in image-based diagnosis. Pixel-wise classification is a popular approach to segmenting the region of interest. However, at fuzzy boundaries, such methods usually result in glitches, discontinuity or disconnection, inconsistent with the fact that lesions are solid and smooth. Methods: To overcome these problems and to provide an efficient, accurate, robust and concise solution that simplifies the whole segmentation pipeline in AI-assisted applications, we propose the BézierSeg model which outputs Bézier curves encompassing the region of interest. Results: Directly modeling the contour with analytic equations ensures that the segmentation is connected and continuous, and that the boundary is smooth. In addition, it offers sub-pixel accuracy. Without loss of precision, the Bézier contour can be resampled and overlaid with images of any resolution. Moreover, clinicians can conveniently adjust the curve’s control points to refine the result. Conclusions: Our experiments show that the proposed method runs in real time and achieves accuracy competitive with pixel-wise segmentation models.

show abstract

“…[58][59][60] Most of these deep learning (DL) approaches take advantage of U-Net variants. [61][62][63][64][65] Recent application of DL-based kidney segmentation has focused on automation of renal cyst and kidney volume measurements in healthy subjects and patients with autosomal-dominant PKD and CKD. 54,56,60,[66][67][68][69][70] The feasibility and reliability of dynamic or longitudinal MRIbased KS monitoring using DL in acute pathophysiological scenarios, where changes may be more subtle than in autosomal-dominant PKD or CKD, has not yet been investigated.…”

Section: Introductionmentioning

confidence: 99%

“…MRI studies using neural networks for renal segmentation reported processing times as good as 1–10 s per subject 58–60 . Most of these deep learning (DL) approaches take advantage of U‐Net variants 61–65 . Recent application of DL‐based kidney segmentation has focused on automation of renal cyst and kidney volume measurements in healthy subjects and patients with autosomal‐dominant PKD and CKD 54,56,60,66–70 .…”

Section: Introductionmentioning

confidence: 99%

“…Although the manual and the deep learning (DL)-based segmentations accurately include the pelvis and exclude the extrarenal vessels, the analytical model (D) draws a straight border from both poles of the kidney.four convolutional layers with growing dilation and residual connections. This results in an increased receptive field and an asymmetric network architecture with a larger encoding part 62,65. A schematic representation of the network architecture is shown in Figure3A.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Dynamic parametric MRI and deep learning: Unveiling renal pathophysiology through accurate kidney size quantification

Klein,

Gladytz,

Millward

et al. 2023

NMR in Biomedicine

View full text Add to dashboard Cite

Renal pathologies often manifest as alterations in kidney size, providing a valuable avenue for employing dynamic parametric MRI as a means to derive kidney size measurements for the diagnosis, treatment, and monitoring of renal disease. Furthermore, this approach holds significant potential in supporting MRI data‐driven preclinical investigations into the intricate mechanisms underlying renal pathophysiology. The integration of deep learning algorithms is crucial in achieving rapid and precise segmentation of the kidney from temporally resolved parametric MRI, facilitating the use of kidney size as a meaningful (pre)clinical biomarker for renal disease. To explore this potential, we employed dynamic parametric T2 mapping of the kidney in rats in conjunction with a custom‐tailored deep dilated U‐Net (DDU‐Net) architecture. The architecture was trained, validated, and tested on manually segmented ground truth kidney data, with benchmarking against an analytical segmentation model and a self‐configuring no new U‐Net. Subsequently, we applied our approach to in vivo longitudinal MRI data, incorporating interventions that emulate clinically relevant scenarios in rats. Our approach achieved high performance metrics, including a Dice coefficient of 0.98, coefficient of determination of 0.92, and a mean absolute percentage error of 1.1% compared with ground truth. The DDU‐Net enabled automated and accurate quantification of acute changes in kidney size, such as aortic occlusion (−8% ± 1%), venous occlusion (5% ± 1%), furosemide administration (2% ± 1%), hypoxemia (−2% ± 1%), and contrast agent‐induced acute kidney injury (11% ± 1%). This approach can potentially be instrumental for the development of dynamic parametric MRI‐based tools for kidney disorders, offering unparalleled insights into renal pathophysiology.

show abstract

Stacked dilated convolutions and asymmetric architecture for U-Net-based medical image segmentation

Cited by 16 publications

References 28 publications

A multi-modal vision-language pipeline strategy for contour quality assurance and adaptive optimization

A multi-modal vision-language pipeline strategy for contour quality assurance and adaptive optimization

BézierSeg: Parametric Shape Representation for Fast Object Segmentation in Medical Images

Dynamic parametric MRI and deep learning: Unveiling renal pathophysiology through accurate kidney size quantification

Contact Info

Product

Resources

About