Validation of a fully automated liver segmentation algorithm using multi-scale deep reinforcement learning and comparison versus manual segmentation

Winkel, David; Weikert, Thomas; Breit, Hanns‐Christian; Chabin, Guillaume; Gibson, Eli; Heye, Tobias; Comanicìu, Dorin; Boll, Daniel T.

doi:10.1016/j.ejrad.2020.108918

Cited by 38 publications

(21 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One reason for the lack of such studies is certainly that organ volumetric analyses have been performed in the past using manual contour segmentation, partially employing techniques to speed up the process, such as semiautomated contour interpolation [8,9]. Nevertheless, in order to create a meaningful reference database for organ volumes, the number of cases that needs to be processed would exceed manual segmentation capabilities.…”

Section: Discussionmentioning

confidence: 99%

“…Using DRL to fill this gap in clinical practice has one prerequisite, namely, that the outputted values are correct and that outliers on both sides-including missing organ-are accurately captured and make sense. As we did not perform a dedicated comparison between the automatically and manually derived organ volumes, serving as a reference standard, we developed two strategies in order to review the usefulness and correctness of our values: (i) referencing our organ volumes to values published in the literature, except for liver volumes, since previous work [9], using the same framework, showed an excellent agreement between the averaged liver volumes of three human readers and the AI approach and (ii) internally validating outlier cases from the AI solution by comparison with manual contour segmentation.…”

Section: Discussionmentioning

confidence: 99%

“…With this technology, whole body organ volumetric analyses can be derived in a short amount of time. Furthermore, applying this to liver volumetric analyses has recently been shown to yield an excellent agreement with human readers [9]. We hypothesize that the approach is capable of building databases that can assist reporting.…”

Section: Introductionmentioning

confidence: 87%

“…The creation of such organ-specific databases for whole-body organs is a currently an unsolved task; one of the main reasons is that manual organ segmentation is a time-consuming task that is difficult to incorporate in high-throughput clinical routine [8]. Furthermore, it is prone to a relevant amount of inter-and intra-reader variability [9].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Building Large-Scale Quantitative Imaging Databases with Multi-Scale Deep Reinforcement Learning: Initial Experience with Whole-Body Organ Volumetric Analyses

et al. 2021

Self Cite

View full text Add to dashboard Cite

To explore the feasibility of a fully automated workflow for whole-body volumetric analyses based on deep reinforcement learning (DRL) and to investigate the influence of contrast-phase (CP) and slice thickness (ST) on the calculated organ volume. This retrospective study included 431 multiphasic CT datasets—including three CP and two ST reconstructions for abdominal organs—totaling 10,508 organ volumes (10,344 abdominal organ volumes: liver, spleen, and kidneys, 164 lung volumes). Whole-body organ volumes were determined using multi-scale DRL for 3D anatomical landmark detection and 3D organ segmentation. Total processing time for all volumes and mean calculation time per case were recorded. Repeated measures analyses of variance (ANOVA) were conducted to test for robustness considering CP and ST. The algorithm calculated organ volumes for the liver, spleen, and right and left kidney (mean volumes in milliliter (interquartile range), portal venous CP, 5 mm ST: 1868.6 (1426.9, 2157.8), 350.19 (45.46, 395.26), 186.30 (147.05, 214.99) and 181.91 (143.22, 210.35), respectively), and for the right and left lung (2363.1 (1746.3, 2851.3) and 1950.9 (1335.2, 2414.2)). We found no statistically significant effects of the variable contrast phase or the variable slice thickness on the organ volumes. Mean computational time per case was 10 seconds. The evaluated approach, using state-of-the art DRL, enables a fast processing of substantial amounts irrespective of CP and ST, allowing building up organ-specific volumetric databases. The thus derived volumes may serve as reference for quantitative imaging follow-up.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 87%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Building Large-Scale Quantitative Imaging Databases with Multi-Scale Deep Reinforcement Learning: Initial Experience with Whole-Body Organ Volumetric Analyses

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Surrogate endpoints for physicians’ fatigue, like detection of pathology and diagnosis accuracy [ 15 ], could benefit from the help of artificial intelligence (AI) [ 16 , 17 ]. Over the past few decades, several AI-algorithms have proven their performance in radiology [ 18 , 19 , 20 , 21 , 22 ], reducing the number of missed findings and false-positive findings (FPs) [ 23 ]. Furthermore, automated pathology detection allows radiologists to put their capacities into more complex tasks, such as making the final diagnosis [ 24 , 25 ].…”

Section: Introductionmentioning

confidence: 99%

Automated Detection of Pancreatic Cystic Lesions on CT Using Deep Learning

et al. 2021

Self Cite

View full text Add to dashboard Cite

Pancreatic cystic lesions (PCL) are a frequent and underreported incidental finding on CT scans and can transform into neoplasms with devastating consequences. We developed and evaluated an algorithm based on a two-step nnU-Net architecture for automated detection of PCL on CTs. A total of 543 cysts on 221 abdominal CTs were manually segmented in 3D by a radiology resident in consensus with a board-certified radiologist specialized in abdominal radiology. This information was used to train a two-step nnU-Net for detection with the performance assessed depending on lesions’ volume and location in comparison to three human readers of varying experience. Mean sensitivity was 78.8 ± 0.1%. The sensitivity was highest for large lesions with 87.8% for cysts ≥220 mm3 and for lesions in the distal pancreas with up to 96.2%. The number of false-positive detections for cysts ≥220 mm3 was 0.1 per case. The algorithm’s performance was comparable to human readers. To conclude, automated detection of PCL on CTs is feasible. The proposed model could serve radiologists as a second reading tool. All imaging data and code used in this study are freely available online.

show abstract

Reinforcement learning in medical image analysis: Concepts, applications, challenges, and future directions

Zhang

Luke

et al. 2023

J Applied Clin Med Phys

View full text Add to dashboard Cite

Motivation Medical image analysis involves a series of tasks used to assist physicians in qualitative and quantitative analyses of lesions or anatomical structures which can significantly improve the accuracy and reliability of medical diagnoses and prognoses. Traditionally, these tedious tasks were finished by experienced physicians or medical physicists and were marred with two major problems, low efficiency and bias. In the past decade, many machine learning methods have been applied to accelerate and automate the image analysis process. Compared to the enormous deployments of supervised and unsupervised learning models, attempts to use reinforcement learning in medical image analysis are still scarce. We hope that this review article could serve as the stepping stone for related research in the future. Significance We found that although reinforcement learning has gradually gained momentum in recent years, many researchers in the medical analysis field still find it hard to understand and deploy in clinical settings. One possible cause is a lack of well‐organized review articles intended for readers without professional computer science backgrounds. Rather than to provide a comprehensive list of all reinforcement learning models applied in medical image analysis, the aim of this review is to help the readers formulate and solve their medical image analysis research through the lens of reinforcement learning. Approach & Results We selected published articles from Google Scholar and PubMed. Considering the scarcity of related articles, we also included some outstanding newest preprints. The papers were carefully reviewed and categorized according to the type of image analysis task. In this article, we first reviewed the basic concepts and popular models of reinforcement learning. Then, we explored the applications of reinforcement learning models in medical image analysis. Finally, we concluded the article by discussing the reviewed reinforcement learning approaches’ limitations and possible future improvements.

show abstract

Validation of a fully automated liver segmentation algorithm using multi-scale deep reinforcement learning and comparison versus manual segmentation

Cited by 38 publications

References 22 publications

Building Large-Scale Quantitative Imaging Databases with Multi-Scale Deep Reinforcement Learning: Initial Experience with Whole-Body Organ Volumetric Analyses

Building Large-Scale Quantitative Imaging Databases with Multi-Scale Deep Reinforcement Learning: Initial Experience with Whole-Body Organ Volumetric Analyses

Automated Detection of Pancreatic Cystic Lesions on CT Using Deep Learning

Reinforcement learning in medical image analysis: Concepts, applications, challenges, and future directions

Contact Info

Product

Resources

About