Patient-Specific Finetuning of Deep Learning Models for Adaptive Radiotherapy in Prostate CT

Elmahdy, Mohamed S.; Ahuja, Tanuj; Heide, Uulke A. van der; Staring, Marius

doi:10.1109/isbi45749.2020.9098702

Cited by 15 publications

(17 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The numbers on the test set are slightly higher than the validation set, but this is due to the variance between the deformations between both sets and the fact that the network has not seen the test set before. This can be addressed using transfer learning as suggested by Elmahdy et al [25] or by using synthetic deformations that mimic the one presented in the EMC dataset.…”

Section: Discussionmentioning

confidence: 99%

“…Recently, deep learning-based networks have shown unprecedented success in many fields especially in the medical image analysis domain [19], [20], [21], [22], [23], [24], [25], [26], where deep learning models perform on par with medical experts or even surpassing them in some tasks [27], [28], [29], [30], [31]. Several deep learning-based approaches have been proposed for joint registration and segmentation.…”

Section: A Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Joint Registration and Segmentation via Multi-Task Learning for Adaptive Radiotherapy of Prostate Cancer

et al. 2021

Self Cite

View full text Add to dashboard Cite

Medical image registration and segmentation are two of the most frequent tasks in medical image analysis. As these tasks are complementary and correlated, it would be beneficial to apply them simultaneously in a joint manner. In this paper, we formulate registration and segmentation as a joint problem via a Multi-Task Learning (MTL) setting, allowing these tasks to leverage their strengths and mitigate their weaknesses through the sharing of beneficial information. We propose to merge these tasks not only on the loss level, but on the architectural level as well. We studied this approach in the context of adaptive image-guided radiotherapy for prostate cancer, where planning and follow-up CT images as well as their corresponding contours are available for training. At testing time the contours of the follow-up scans are not available, which is a common scenario in adaptive radiotherapy. The study involves two datasets from different manufacturers and institutes. The first dataset was divided into training (12 patients) and validation (6 patients), and was used to optimize and validate the methodology, while the second dataset (14 patients) was used as an independent test set. We carried out an extensive quantitative comparison between the quality of the automatically generated contours from different network architectures as well as loss weighting methods. Moreover, we evaluated the quality of the generated deformation vector field (DVF). We show that MTL algorithms outperform their Single-Task Learning (STL) counterparts and achieve better generalization on the independent test set. The best algorithm achieved a mean surface distance of 1.06 ± 0.3 mm, 1.27 ± 0.4 mm, 0.91 ± 0.4 mm, and 1.76 ± 0.8 mm on the validation set for the prostate, seminal vesicles, bladder, and rectum, respectively. The high accuracy of the proposed method combined with the fast inference speed, makes it a promising method for automatic re-contouring of follow-up scans for adaptive radiotherapy, potentially reducing treatment related complications and therefore improving patients quality-of-life after treatment.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: A Related Workmentioning

confidence: 99%

Joint Registration and Segmentation via Multi-Task Learning for Adaptive Radiotherapy of Prostate Cancer

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…To our knowledge, there have been several related studies utilizing the patient-specific fine-tuning strategy. 12,19,20 Kim et al proposed continual DLbased segmentation for personalized adaptive radiation therapy in the HN area. However, they could not see a significant improvement in performance because there was a consistency problem between the ground truths of pCT and rpCT.…”

Section: Discussionmentioning

confidence: 99%

Intentional deep overfit learning (IDOL): A novel deep learning strategy for adaptive radiation therapy

et al. 2021

View full text Add to dashboard Cite

Purpose: Applications of deep learning (DL) are essential to realizing an effective adaptive radiotherapy (ART) workflow. Despite the promise demonstrated by DL approaches in several critical ART tasks, there remain unsolved challenges to achieve satisfactory generalizability of a trained model in a clinical setting. Foremost among these is the difficulty of collecting a task-specific training dataset with high-quality, consistent annotations for supervised learning applications. In this study, we propose a tailored DL framework for patient-specific performance that leverages the behavior of a model intentionally overfitted to a patient-specific training dataset augmented from the prior information available in an ART workflow-an approach we term Intentional Deep Overfit Learning (IDOL). Methods: Implementing the IDOL framework in any task in radiotherapy consists of two training stages: (1) training a generalized model with a diverse training dataset of N patients, just as in the conventional DL approach, and (2) intentionally overfitting this general model to a small training dataset-specific the patient of interest (N + 1) generated through perturbations and augmentations of the available task-and patient-specific prior information to establish a personalized IDOL model. The IDOL framework itself is task-agnostic and is, thus, widely applicable to many components of the ART workflow, three of which we use as a proof of concept here: the autocontouring task on replanning CTs for traditional ART, the MRI super-resolution (SR) task for MRI-guided ART, and the synthetic CT (sCT) reconstruction task for MRI-only ART. Results: In the replanning CT autocontouring task, the accuracy measured by the Dice similarity coefficient improves from 0.847 with the general model to 0.935 by adopting the IDOL model. In the case of MRI SR, the mean absolute error (MAE) is improved by 40% using the IDOL framework over the conventional model. Finally, in the sCT reconstruction task, the MAE is reduced from 68 to 22 HU by utilizing the IDOL framework. Conclusions: In this study, we propose a novel IDOL framework for ART and demonstrate its feasibility using three ART tasks.We expect the IDOL framework to be especially useful in creating personally tailored models in situations with limited availability of training data but existing prior information, which is usually true in the medical setting in general and is especially true in ART. K E Y W O R D S adaptive radiotherapy, deep learning, overfitting, personalized model 488

show abstract

“…In the medical imaging domain, transfer learning from natural image datasets, particularly ImageNet [39], using standard large models and corresponding pretrained weights has become a de-facto method to speed up training convergence and to improve accuracy [40]. Transfer learning can also be used to leverage personalized anatomical knowledge accumulated over time to improve the accuracy of pre-trained CNNs for specific patients [41], i.e., to perform patient-specific model tuning. This is an important topic in biomedical application domains, which will be further discussed in IV-F.…”

Section: B Deep Artificial Neural Networkmentioning

confidence: 99%

Hardware Implementation of Deep Network Accelerators Towards Healthcare and Biomedical Applications

Azghadi

Lammie

Eshraghian

et al. 2020

IEEE Trans. Biomed. Circuits Syst.

149

View full text Add to dashboard Cite

The advent of dedicated Deep Learning (DL) accelerators and neuromorphic processors has brought on new opportunities for applying both Deep and Spiking Neural Network (SNN) algorithms to healthcare and biomedical applications at the edge. This can facilitate the advancement of medical Internet of Things (IoT) systems and Point of Care (PoC) devices. In this paper, we provide a tutorial describing how various technologies including emerging memristive devices, Field Programmable Gate Arrays (FPGAs), and Complementary Metal Oxide Semiconductor (CMOS) can be used to develop efficient DL accelerators to solve a wide variety of diagnostic, pattern recognition, and signal processing problems in healthcare. Furthermore, we explore how spiking neuromorphic processors can complement their DL counterparts for processing biomedical signals. The tutorial is augmented with case studies of the vast literature on neural network and neuromorphic hardware as applied to the healthcare domain. We benchmark various hardware platforms by performing a sensor fusion signal processing task combining electromyography (EMG) signals with computer vision. Comparisons are made between dedicated neuromorphic processors and embedded AI accelerators in terms of inference latency and energy. Finally, we provide our analysis of the field and share a perspective on the advantages, disadvantages, challenges, and opportunities that various accelerators and neuromorphic processors introduce to healthcare and biomedical domains.

show abstract

Patient-Specific Finetuning of Deep Learning Models for Adaptive Radiotherapy in Prostate CT

Cited by 15 publications

References 7 publications

Joint Registration and Segmentation via Multi-Task Learning for Adaptive Radiotherapy of Prostate Cancer

Joint Registration and Segmentation via Multi-Task Learning for Adaptive Radiotherapy of Prostate Cancer

Intentional deep overfit learning (IDOL): A novel deep learning strategy for adaptive radiation therapy

Hardware Implementation of Deep Network Accelerators Towards Healthcare and Biomedical Applications

Contact Info

Product

Resources

About