Deep learning system to improve the quality and efficiency of volumetric heart segmentation for breast cancer
Although artificial intelligence algorithms are often developed and applied for narrow tasks, their implementation in other medical settings could help to improve patient care. This project assessed whether a deep learning system for volumetric heart segmentation on computed tomography (CT) scans developed in cardiovascular radiology can optimize treatment planning in radiation oncology. An integral part of radiotherapy treatment planning is segmenting organs at risk in the radiation field on computed tomography (CT) scans. If appropriate resources are available, this is done manually by trained experts who require considerable time and are prone to inter- and intra-observer variability. If time or knowledge are limited, this crucial step to ensure treatment quality and patient safety may be neglected. Therefore, automating and optimizing this process of organ at risk segmentation by deep learning could improve clinical care at high speed and low additional cost, especially in underprivileged healthcare settings.
The system was trained using multi-center data (n=858) with manual heart segmentations provided by cardiovascular radiologists. Validation of the system was performed in an independent real-world dataset of 5,677 breast cancer patients treated with radiation therapy at the Dana-Farber/Brigham and Women's Cancer Center between 2008-2018. In a subset of 20 patients, the performance of the system was compared to eight radiation oncology experts by assessing segmentation time, agreement between experts, and accuracy with and without deep learning assistance. To compare the performance to segmentations used in the clinic, concordance and failures (defined as Dice<0.85) of the system were evaluated in the entire dataset. The system was successfully applied without retraining. The results of this study suggest that deep learning algorithms can successfully be applied across medical specialties and improve clinical care beyond the original field of interest.
Publication
Deep convolutional neural networks to predict cardiovascular risk from computed tomography, Roman Zeleznik, Jakob Weiss, Jana Taron, Christian Guthier, Danielle S Bitterman, Cindy Hancox, Benjamin H Kann, Daniel W Kim, Rinaa S Punglia, Jeremy Bredfeldt, Borek Foldyna, Parastou Eslami, Michael T Lu, Udo Hoffmann, Raymond Mak, Hugo Aerts, npj Digital Medicine volume 4, 43 (2021)
Code availability
All code of the deep learning system including the trained models are publicly available under the open MIT license and can be found here.
Statistical Code
This link contains the code to reproduce the statistical analysis of our paper Deep learning system to improve the quality and efficiency of volumetric heart segmentation for breast cancer. More information about the statistical analysis can be found in the Methods section.
Data availability
Example data of computed tomography (CT) images for testing the deep learning system can be found here. For this we included four lung cancer screening thoracic CT scans where expert readers manually segmented the heart. The image data and manual segmentations are available in the nrrd format.
AIM Investigators
Acknowledgements
We would like to thank the Framingham Heart Study, NCI, ACRIN, NLST, Prospective Multicenter Imaging Study for Evaluation of Chest Pain, and Rule Out Myocardial Infarction Using Computer Assisted Tomography II trial for access to trial data.The authors acknowledge financial support from NIH (NIH-USA U24CA194354, NIH-USA U01CA190234, NIH-USA U01CA209414, NIH-USA R35CA22052; 5R01-HL109711, NIH/NHLBI 5K24HL113128, NIH/NHLBI 5T32HL076136, NIH/NHLBI 5U01HL123339), the European Union - European Research Council (866504), as well as the German Research Foundation (DFG 1438/1-1 and 6405/2-1), American Heart Association Institute for Precision Cardiovascular Medicine (8UNPG34030172), Fulbright Visiting Researcher Grant (E0583118), Rosztoczy Foundation Grant. The Framingham Heart Study (FHS) acknowledges the support of contracts NO1-HC-25195, HHSN268201500001I and 75N92019D00031 from the National Heart, Lung and Blood Institute.