Technical and Clinical Factors Affecting Success Rate of a Deep Learning Method for Pancreas Segmentation on CT

Bagheri, Mohammad Hadi; Roth, Holger R.; Kovacs, William; Yao, Jianhua; Farhadi, Faraz; Li, Xiaobai; Summers, Ronald M.

doi:10.1016/j.acra.2019.08.014

Cited by 18 publications

(21 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multiphasic CT scans may further increase the accuracy of pancreas segmentation. 15,49 Our proposed segmentation method is for morphologically normal pancreas. Thus, the performance of our model for segmentation of diseased pancreas remains to be seen.…”

Section: Discussionmentioning

confidence: 99%

“…Manual segmentation of pancreas is a time-consuming, cumbersome process, and is, therefore, not scalable. 15,16 The pancreas is a small and deformable organ with considerable variability and complexity in anatomy, location, shape, and attenuation. 17 Compared to segmentation of other solid abdominal organs, pancreas segmentation has been consistently less accurate even with deep learning (DL) methods based on convolutional neural networks (CNNs).…”

Section: Introductionmentioning

confidence: 99%

“…Manual segmentation of pancreas is a time‐consuming, cumbersome process, and is, therefore, not scalable 15,16 . The pancreas is a small and deformable organ with considerable variability and complexity in anatomy, location, shape, and attenuation 17 .…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Two‐stage deep learning model for fully automated pancreas segmentation on computed tomography: Comparison with intra‐reader and inter‐reader reliability at full and reduced radiation dose on an external dataset

et al. 2021

View full text Add to dashboard Cite

To develop a two-stage three-dimensional (3D) convolutional neural networks (CNNs) for fully automated volumetric segmentation of pancreas on computed tomography (CT) and to further evaluate its performance in the context of intra-reader and inter-reader reliability at full dose and reduced radiation dose CTs on a public dataset. Methods: A dataset of 1994 abdomen CT scans (portal venous phase, slice thickness ≤ 3.75-mm, multiple CT vendors) was curated by two radiologists (R1 and R2) to exclude cases with pancreatic pathology, suboptimal image quality, and image artifacts (n = 77). Remaining 1917 CTs were equally allocated between R1 and R2 for volumetric pancreas segmentation [ground truth (GT)]. This internal dataset was randomly divided into training (n = 1380), validation (n = 248), and test (n = 289) sets for the development of a two-stage 3D CNN model based on a modified U-net architecture for automated volumetric pancreas segmentation. Model's performance for pancreas segmentation and the differences in model-predicted pancreatic volumes vs GT volumes were compared on the test set. Subsequently, an external dataset from The Cancer Imaging Archive (TCIA) that had CT scans acquired at standard radiation dose and same scans reconstructed at a simulated 25% radiation dose was curated (n = 41). Volumetric pancreas segmentation was done on this TCIA dataset by R1 and R2 independently on the full dose and then at the reduced radiation dose CT images. Intra-reader and inter-reader reliability, model's segmentation performance, and reliability between model-predicted pancreatic volumes at full vs reduced dose were measured. Finally, model's performance was tested on the benchmarking National Institute of Health (NIH)-Pancreas CT (PCT) dataset. Results: Three-dimensional CNN had mean (SD) Dice similarity coefficient (DSC): 0.91 (0.03) and average Hausdorff distance of 0.15 (0.09) mm on the test set. Model's performance was equivalent between males and females (P = 0.08) and across different CT slice thicknesses (P > 0.05) based on noninferiority statistical testing. There was no difference in model-predicted and GT pancreatic volumes [mean predicted volume 99 cc (31cc); GT volume 101 cc (33 cc), P = 0.33]. Mean pancreatic volume difference was À2.7 cc (percent difference: À2.4% of GT volume) with excellent correlation between model-predicted and GT volumes [concordance correlation coefficient (CCC)=0.97]. In the external TCIA dataset, the model had higher reliability than R1 and R2 on full vs reduced dose CT scans [model mean (SD) DSC: 0.96 (0.02), CCC = 0.995 vs R1 DSC: 0.83 (0.07), CCC = 0.89, and R2 DSC:0.87 (0.04), CCC = 0.97]. The DSC and volume concordance correlations for R1 vs R2 (inter-reader reliability) were 0.85 (0.07), CCC = 0.90 at full dose and 0.83 (0.07), CCC = 0.96 at reduced dose datasets. There was good reliability between model and R1 at both full and reduced dose CT [full dose: DSC: 0.81 (0.07), CCC = 0.83 and reduced dose DSC:0.81 (0.08), CCC = 0.87]. Likewise, there was good reliability betw...

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Two‐stage deep learning model for fully automated pancreas segmentation on computed tomography: Comparison with intra‐reader and inter‐reader reliability at full and reduced radiation dose on an external dataset

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Therefore, validated methods for automated segmentation of pancreas in clinical practice are necessary. Automated pancreas segmentation will also have potential applications in surgical and radiation therapy planning, and for early detection of pancreatic cancer [5]. Although technologists gain a working knowledge of key anatomical landmarks during their routine clinical assignments, the skills needed for fine segmentations of organs such as pancreas on cross-sectional imaging are not part of their portfolio.…”

Section: Discussionmentioning

confidence: 99%

“…The pancreas is a solid retroperitoneal organ that can be hard to segment because of its small size, complex anatomy, and variability in location, morphology, and attenuation [4]. Furthermore, the variable degrees of peripancreatic fat, contrast enhancement, and subadjacent iso-attenuating structures such as collapsed bowel can further confound delineation of its exact boundaries [5][6][7]. These factors make manual segmentation of the pancreas a challenge and at least partly contribute to the underutilization of pancreas morphometrics and radiomics in both endocrine and exocrine diseases despite promising results [8][9][10][11].…”

Section: Introductionmentioning

confidence: 99%

Development of a volumetric pancreas segmentation CT dataset for AI applications through trained technologists: a study during the COVID 19 containment phase

et al. 2020

View full text Add to dashboard Cite

Purpose To evaluate the performance of trained technologists vis-à-vis radiologists for volumetric pancreas segmentation and to assess the impact of supplementary training on their performance. Methods In this IRB-approved study, 22 technologists were trained in pancreas segmentation on portal venous phase CT through radiologist-led interactive videoconferencing sessions based on an image-rich curriculum. Technologists segmented pancreas in 188 CTs using freehand tools on custom image-viewing software. Subsequent supplementary training included multimedia videos focused on common errors, which were followed by second batch of 159 segmentations. Two radiologists reviewed all cases and corrected inaccurate segmentations. Technologists' segmentations were compared against radiologists' segmentations using Dice-Sorenson coefficient (DSC), Jaccard coefficient (JC), and Bland-Altman analysis. Results Corrections were made in 71 (38%) cases from first batch [26 (37%) oversegmentations and 45 (63%) undersegmentations] and in 77 (48%) cases from second batch [12 (16%) oversegmentations and 65 (84%) undersegmentations]. DSC, JC, false positive (FP), and false negative (FN) [mean (SD)] in first versus second batches were 0.63 (0.15) versus 0.63 (0.16), 0.48 (0.15) versus 0.48 (0.15), 0.29 (0.21) versus 0.21 (0.10), and 0.36 (0.20) versus 0.43 (0.19), respectively. Differences were not significant (p > 0.05). However, range of mean pancreatic volume difference reduced in the second batch [− 2.74 cc (min − 92.96 cc, max 87.47 cc) versus − 23.57 cc (min − 77.32, max 30.19)]. Conclusion Trained technologists could perform volumetric pancreas segmentation with reasonable accuracy despite its complexity. Supplementary training further reduced range of volume difference in segmentations. Investment into training technologists could augment and accelerate development of body imaging datasets for AI applications.

show abstract

Multiscale unsupervised domain adaptation for automatic pancreas segmentation in CT volumes using adversarial learning

Zhu

et al. 2022

Medical Physics

View full text Add to dashboard Cite

Computer-aided automatic pancreas segmentation is essential for early diagnosis and treatment of pancreatic diseases. However, the annotation of pancreas images requires professional doctors and considerable expenditure. Due to imaging differences among various institution population, scanning devices, imaging protocols, and so on, significant degradation in the performance of model inference results is prone to occur when models trained with domain-specific (usually institution-specific) datasets are directly applied to new (other centers/institutions) domain data. In this paper, we propose a novel unsupervised domain adaptation method based on adversarial learning to address pancreas segmentation challenges with the lack of annotations and domain shift interference. Methods: A 3D semantic segmentation model with attention module and residual module is designed as the backbone pancreas segmentation model. In both segmentation model and domain adaptation discriminator network, a multiscale progressively weighted structure is introduced to acquire different field of views. Features of labeled data and unlabeled data are fed in pairs into the proposed multiscale discriminator to learn domain-specific characteristics. Then the unlabeled data features with pseudodomain label are fed to the discriminator to acquire domain-ambiguous information. With this adversarial learning strategy, the performance of the segmentation network is enhanced to segment unseen unlabeled data. Results: Experiments were conducted on two public annotated datasets as source datasets, respectively, and one private dataset as target dataset, where annotations were not used for the training process but only for evaluation. The 3D segmentation model achieves comparative performance with state-of -theart pancreas segmentation methods on source domain. After implementing our domain adaptation architecture, the average dice similarity coefficient (DSC) of the segmentation model trained on the NIH-TCIA source dataset increases from 58.79% to 72.73% on the local hospital dataset, while the performance of the target domain segmentation model transferred from the medical segmentation decathlon (MSD) source dataset rises from 62.34% to 71.17%. Conclusions: Correlations of features across data domains are utilized to train the pancreas segmentation model on unlabeled data domain, improving the generalization of the model. Our results demonstrate that the proposed method enables the segmentation model to make meaningful segmentation for unseen data of the training set. In the future, the proposed method has the potential to

show abstract

Technical and Clinical Factors Affecting Success Rate of a Deep Learning Method for Pancreas Segmentation on CT

Cited by 18 publications

References 27 publications

Two‐stage deep learning model for fully automated pancreas segmentation on computed tomography: Comparison with intra‐reader and inter‐reader reliability at full and reduced radiation dose on an external dataset

Two‐stage deep learning model for fully automated pancreas segmentation on computed tomography: Comparison with intra‐reader and inter‐reader reliability at full and reduced radiation dose on an external dataset

Development of a volumetric pancreas segmentation CT dataset for AI applications through trained technologists: a study during the COVID 19 containment phase

Multiscale unsupervised domain adaptation for automatic pancreas segmentation in CT volumes using adversarial learning

Contact Info

Product

Resources

About