Evaluation of Deep Learning to Augment Image-Guided Radiotherapy for Head and Neck and Prostate Cancers

Published on Nov 2, 2020
· DOI :10.1001/JAMANETWORKOPEN.2020.27426
Ozan Oktay23
Estimated H-index: 23
Jay Nanavati4
Estimated H-index: 4
+ 12 AuthorsYvonne L. Rimmer7
Estimated H-index: 7
(Cambridge University Hospitals NHS Foundation Trust)
Importance Personalized radiotherapy planning depends on high-quality delineation of target tumors and surrounding organs at risk (OARs). This process puts additional time burdens on oncologists and introduces variability among both experts and institutions. Objective To explore clinically acceptable autocontouring solutions that can be integrated into existing workflows and used in different domains of radiotherapy. Design, Setting, and Participants This quality improvement study used a multicenter imaging data set comprising 519 pelvic and 242 head and neck computed tomography (CT) scans from 8 distinct clinical sites and patients diagnosed either with prostate or head and neck cancer. The scans were acquired as part of treatment dose planning from patients who received intensity-modulated radiation therapy between October 2013 and February 2020. Fifteen different OARs were manually annotated by expert readers and radiation oncologists. The models were trained on a subset of the data set to automatically delineate OARs and evaluated on both internal and external data sets. Data analysis was conducted October 2019 to September 2020. Main Outcomes and Measures The autocontouring solution was evaluated on external data sets, and its accuracy was quantified with volumetric agreement and surface distance measures. Models were benchmarked against expert annotations in an interobserver variability (IOV) study. Clinical utility was evaluated by measuring time spent on manual corrections and annotations from scratch. Results A total of 519 participants’ (519 [100%] men; 390 [75%] aged 62-75 years) pelvic CT images and 242 participants’ (184 [76%] men; 194 [80%] aged 50-73 years) head and neck CT images were included. The models achieved levels of clinical accuracy within the bounds of expert IOV for 13 of 15 structures (eg, left femur, κ = 0.982; brainstem, κ = 0.806) and performed consistently well across both external and internal data sets (eg, mean [SD] Dice score for left femur, internal vs external data sets: 98.52% [0.50] vs 98.04% [1.02];P = .04). The correction time of autogenerated contours on 10 head and neck and 10 prostate scans was measured as a mean of 4.98 (95% CI, 4.44-5.52) min/scan and 3.40 (95% CI, 1.60-5.20) min/scan, respectively, to ensure clinically accepted accuracy. Manual segmentation of the head and neck took a mean 86.75 (95% CI, 75.21-92.29) min/scan for an expert reader and 73.25 (95% CI, 68.68-77.82) min/scan for a radiation oncologist. The autogenerated contours represented a 93% reduction in time. Conclusions and Relevance In this study, the models achieved levels of clinical accuracy within expert IOV while reducing manual contouring time and performing consistently well across previously unseen heterogeneous data sets. With the availability of open-source libraries and reliable performance, this creates significant opportunities for the transformation of radiation treatment planning.
#1Yueyue Wang (Fudan University)H-Index: 1
#2Liang Zhao (Fudan University)H-Index: 1
Last. Zhijian SongH-Index: 8
view all 4 authors...
Accurate segmentation of organs at risk (OARs) plays a critical role in the treatment planning of image-guided radiotherapy of head and neck cancer. This segmentation task is challenging for both humans and automated algorithms because of the relatively large number of OARs to be segmented, the large variability in size and morphology across different OARs, and the low contrast between some OARs and the background. In this study, we propose a two-stage segmentation framework based on 3D U-Net. I...
29 CitationsSource
#1Jan-Jakob Sonke (NKI-AVL: Netherlands Cancer Institute)H-Index: 55
#2Marianne C. Aznar (University of Manchester)H-Index: 31
Last. Coen R. N. Rasch (UvA: University of Amsterdam)H-Index: 13
view all 3 authors...
The anatomy of cancer patients changes between radiation treatment planning and delivery as well as over the course of radiotherapy. Adaptive radiotherapy (ART) aims to deliver radiation accurately and precisely in the presence of such changes. To that end, ART uses an imaging feedback loop to quantify these changes and modify the treatment plan accordingly. This paper provides an overview of anatomical changes occurring over the course of therapy and various adaptive strategies developed to acc...
51 CitationsSource
#1Giacomo Tarroni (Imperial College London)H-Index: 13
#2Ozan Oktay (Imperial College London)H-Index: 23
Last. Daniel Rueckert (Imperial College London)H-Index: 109
view all 12 authors...
The effectiveness of a cardiovascular magnetic resonance (CMR) scan depends on the ability of the operator to correctly tune the acquisition parameters to the subject being scanned and on the potential occurrence of imaging artifacts, such as cardiac and respiratory motion. In the clinical practice, a quality control step is performed by visual assessment of the acquired images; however, this procedure is strongly operator-dependent, cumbersome, and sometimes incompatible with the time constrain...
22 CitationsSource
#1Shuai Wang (UNC: University of North Carolina at Chapel Hill)H-Index: 9
#2Kelei He (NU: Nanjing University)H-Index: 7
Last. Dinggang Shen (UNC: University of North Carolina at Chapel Hill)H-Index: 113
view all 6 authors...
Abstract Accurate segmentation of the prostate and organs at risk (e.g., bladder and rectum) in CT images is a crucial step for radiation therapy in the treatment of prostate cancer. However, it is a very challenging task due to unclear boundaries, large intra- and inter-patient shape variability, and uncertain existence of bowel gases and fiducial markers. In this paper, we propose a novel automatic segmentation framework using fully convolutional networks with boundary sensitive representation...
39 CitationsSource
#1Wentao Zhu (UCI: University of California, Irvine)H-Index: 17
#2Yufang HuangH-Index: 6
Last. Xiaohui Xie (UCI: University of California, Irvine)H-Index: 60
view all 9 authors...
Methods: Our deep learning model, called AnatomyNet, segments OARs from head and neck CT images in an end-to-end fashion, receiving whole-volume HaN CT images as input and generating masks of all OARs of interest in one shot. AnatomyNet is built upon the popular 3D U-net architecture, but extends it in three important ways: 1) a new encoding scheme to allow auto-segmentation on whole-volume CT images instead of local patches or subsets of slices, 2) incorporating 3D squeeze-and-excitation residu...
180 CitationsSource
#1Anjali Balagopal (UTSW: University of Texas Southwestern Medical Center)H-Index: 5
#2Samaneh Kazemifar (UTSW: University of Texas Southwestern Medical Center)H-Index: 5
Last. Steve B Jiang (UTSW: University of Texas Southwestern Medical Center)H-Index: 83
view all 7 authors...
: Accurate segmentation of prostate and surrounding organs at risk is important for prostate cancer radiotherapy treatment planning. We present a fully automated workflow for male pelvic CT image segmentation using deep learning. The architecture consists of a 2D organ volume localization network followed by a 3D segmentation network for volumetric segmentation of prostate, bladder, rectum, and femoral heads. We used a multi-channel 2D U-Net followed by a 3D U-Net with encoding arm modified with...
51 CitationsSource
#1Patrik RaudaschlH-Index: 7
#2Paolo Zaffino (Magna Græcia University)H-Index: 12
Last. Karl D. FritscherH-Index: 16
view all 29 authors...
Purpose Automated delineation of structures and organs is a key step in medical imaging. However, due to the large number and diversity of structures and the large variety of segmentation algorithms, a consensus is lacking as to which automated segmentation method works best for certain applications. Segmentation challenges are a good approach for unbiased evaluation and comparison of segmentation algorithms. Methods In this work, we describe and present the results of the Head and Neck Auto-Seg...
123 CitationsSource
7 CitationsSource
#1Hubert Y. Pan (University of Texas MD Anderson Cancer Center)H-Index: 21
#2Bruce G. Haffty (UMDNJ: University of Medicine and Dentistry of New Jersey)H-Index: 83
Last. Benjamin Smith (University of Texas MD Anderson Cancer Center)H-Index: 91
view all 7 authors...
Purpose Prior studies have forecasted demand for radiation therapy to grow 10 times faster than the supply between 2010 and 2020. We updated these projections for 2015 to 2025 to determine whether this imbalance persists and to assess the accuracy of prior projections. Methods and Materials The demand for radiation therapy between 2015 and 2025 was estimated by combining current radiation utilization rates determined by the Surveillance, Epidemiology, and End Results data with population project...
54 CitationsSource
#1Waldo Valenzuela (University of Bern)H-Index: 6
#2Stephen J. Ferguson (ETH Zurich)H-Index: 55
Last. Mauricio Reyes (University of Bern)H-Index: 35
view all 9 authors...
BACKGROUND AND PURPOSE In clinical diagnosis, medical image segmentation plays a key role in the analysis of pathological regions. Despite advances in automatic and semi-automatic segmentation techniques, time-effective correction tools are commonly needed to improve segmentation results. Therefore, these tools must provide faster corrections with a lower number of interactions, and a user-independent solution to reduce the time frame between image acquisition and diagnosis. METHODS We present a...
7 CitationsSource
Cited By10
#1C. Robert (Institut Gustave Roussy)H-Index: 21
#2A. MunozH-Index: 1
Last. P. MeyerH-Index: 1
view all 11 authors...
Deep-learning (DL)-based auto-contouring solutions have recently been proposed as a convincing alternative to decrease workload of target volumes and organs-at-risk (OAR) delineation in radiotherapy planning and improve inter-observer consistency. However, there is minimal literature of clinical implementations of such algorithms in a clinical routine. In this paper we first present an update of the state-of-the-art of DL-based solutions. We then summarize recent recommendations proposed by the ...
Deep learning research has recently witnessed an impressively fast-paced progress in a wide range of tasks including computer vision, natural language processing, and reinforcement learning. The extraordinary performance of these systems often gives the impression that they can be used to revolutionise our lives for the better. However, as recent works point out, these systems suffer from several issues that make them unreliable for use in the real world, including vulnerability to adversarial a...
Adoption of artificial intelligence medical imaging applications is often impeded by barriers between healthcare systems and algorithm developers given that access to both private patient data and commercial model IP is important to perform pre-deployment evaluation. This work investigates a framework for secure, privacy-preserving and AI-enabled medical imaging inference using CrypTFlow2, a state-of-the-art end-to-end compiler allowing cryptographically secure 2-party Computation (2PC) protocol...
#1Junaid Bajwa (Microsoft)
#2Usman Munir (Microsoft)
Last. Bryan Williams (UCLH: University College Hospital)H-Index: 149
view all 4 authors...
ABSTRACT null Artificial intelligence (AI) is a powerful and disruptive area of computer science, with the potential to fundamentally transform the practice of medicine and the delivery of healthcare. In this review article, we outline recent breakthroughs in the application of AI in healthcare, describe a roadmap to building effective, reliable and safe AI systems, and discuss the possible future direction of AI augmented healthcare systems.
#1Ché L ReddyH-Index: 5
Last. Rifat AtunH-Index: 82
view all 3 authors...
#3Wei Li (SYSU: Sun Yat-sen University)H-Index: 17
#4Shuling Chen (SYSU: Sun Yat-sen University)H-Index: 11
#1Margherita Mottola (UNIBO: University of Bologna)H-Index: 1
#2Stephan Ursprung (University of Cambridge)H-Index: 6
Last. Alessandro Bevilacqua (UNIBO: University of Bologna)H-Index: 22
view all 9 authors...
Computed Tomography (CT) is widely used in oncology for morphological evaluation and diagnosis, commonly through visual assessments, often exploiting semi-automatic tools as well. Well-established automatic methods for quantitative imaging offer the opportunity to enrich the radiologist interpretation with a large number of radiomic features, which need to be highly reproducible to be used reliably in clinical practice. This study investigates feature reproducibility against noise, varying resol...
#1Seung Yeun Chung (Yonsei University)H-Index: 5
#2Jee Suk Chang (Yonsei University)H-Index: 17
Last. Yong Bae Kim (Yonsei University)H-Index: 22
view all 9 authors...
BACKGROUND In breast cancer patients receiving radiotherapy (RT), accurate target delineation and reduction of radiation doses to the nearby normal organs is important. However, manual clinical target volume (CTV) and organs-at-risk (OARs) segmentation for treatment planning increases physicians' workload and inter-physician variability considerably. In this study, we evaluated the potential benefits of deep learning-based auto-segmented contours by comparing them to manually delineated contours...
4 CitationsSource
#1Nalee Kim (Yonsei University)H-Index: 8
#2Jaehee Chun (Yonsei University)H-Index: 4
Last. Jin Sung Kim (Yonsei University)H-Index: 20
view all 6 authors...
This study investigated the feasibility of deep learning-based segmentation (DLS) and continual training for adaptive radiotherapy (RT) of head and neck (HN (ii) using data obtained from 60 identical patients with 20 unmatched patients in the test set (DLSu). Manually contoured OARs in adaptive planning CT for independent 20 patients were provided as test sets. Deformable image registration (DIR) was also performed. All 23 OARs were compared using quantitative measurements, and nine OARs were al...
3 CitationsSource