Abstract
Although artificial intelligence algorithms are often developed and applied for narrow tasks, their implementation in other medical settings could help to improve patient care. Here we assess 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. 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 5677 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. With deep-learning assistance, segmentation time significantly decreased (4.0 min [IQR 3.1–5.0] vs. 2.0 min [IQR 1.3–3.5]; p < 0.001), and agreement increased (Dice 0.95 [IQR = 0.02]; vs. 0.97 [IQR = 0.02], p < 0.001). Expert accuracy was similar with and without deep-learning assistance (Dice 0.92 [IQR = 0.02] vs. 0.92 [IQR = 0.02]; p = 0.48), and not significantly different from deep-learning-only segmentations (Dice 0.92 [IQR = 0.02]; p ≥ 0.1). In comparison to real-world data, the system showed high concordance (Dice 0.89 [IQR = 0.06]) across 5677 patients and a significantly lower failure rate (p < 0.001). These results suggest that deep-learning algorithms can successfully be applied across medical specialties and improve clinical care beyond the original field of interest.
Highlights
Medical knowledge is increasing exponentially with an estimated doubling every few months as of 20201
To evaluate the deep-learning system for a clinical radiation oncology implementation, we compared the time needed to generate a clinically acceptable segmentation without and with the assistance of the deep learning system, and found a significant reduction by 50% for the deep-learning-assisted approach compared to the current manual clinical workflow (Fig. 2a)
We demonstrate that expert knowledge encapsulated in a deep-learning system can be disseminated across medical domains to help optimize the treatment of patients with breast cancer in radiation oncology
Summary
Medical knowledge is increasing exponentially with an estimated doubling every few months as of 20201. Deep-learning applications are developed using labeled data generated by medical experts for domain-specific problems. As a result, this expert knowledge is encapsulated in the deep-learning system, providing the opportunity to disseminate this highly skilled expertise across medical domains, institutions and countries, with the potential to optimize patient care and reducing knowledge and economic disparities in undersupplied settings. One area that could benefit from this concept are imagingrelated specialties, such as radiology and radiation oncology While the former uses imaging studies primarily for diagnosis, the latter relies on the same information for organ and tumor targeting, treatment planning and delivery, and monitoring. 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[6]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
