ABSTRACT
Introduction: Organ-at-risk segmentation for head and neck cancer radiation therapy is a complex and time-consuming process (requiring up to 42 individual structure, and may delay start of treatment or even limit access to function-preserving care. Feasibility of using a deep learning (DL) based autosegmentation model to reduce contouring time without compromising contour accuracy is assessed through a blinded randomized trial of radiation oncologists (ROs) using retrospective, de-identified patient data. Methods: Two head and neck expert ROs used dedicated time to create gold standard (GS) contours on computed tomography (CT) images. 445 CTs were used to train a custom 3D U-Net DL model covering 42 organs-at-risk, with an additional 20 CTs were held out for the randomized trial. For each held-out patient dataset, one of the eight participant ROs was randomly allocated to review and revise the contours produced by the DL model, while another reviewed contours produced by a medical dosimetry assistant (MDA), both blinded to their origin. Time required for MDAs and ROs to contour was recorded, and the unrevised DL contours, as well as the RO-revised contours by the MDAs and DL model were compared to the GS for that patient. Results: Mean time for initial MDA contouring was 2.3 hours (range 1.6-3.8 hours) and RO-revision took 1.1 hours (range, 0.4-4.4 hours), compared to 0.7 hours (range 0.1-2.0 hours) for the RO-revisions to DL contours. Total time reduced by 76% (95%-Confidence Interval: 65%-88%) and RO-revision time reduced by 35% (95%-CI,-39%-91%). All geometric and dosimetric metrics computed, agreement with GS was equivalent or significantly greater (p<0.05) for RO-revised DL contours compared to the RO-revised MDA contours, including volumetric Dice similarity coefficient (VDSC), surface DSC, added path length, and the 95%-Hausdorff distance. 32 OARs (76%) had mean VDSC greater than 0.8 for the RO-revised DL contours, compared to 20 (48%) for RO-revised MDA contours, and 34 (81%) for the unrevised DL OARs. Conclusion: DL autosegmentation demonstrated significant time-savings for organ-at-risk contouring while improving agreement with the institutional GS, indicating comparable accuracy of DL model. Integration into the clinical practice with a prospective evaluation is currently underway.
ABSTRACT
As medical science advances and the population ages, the prevalence of chronic conditions has also grown. The traditional model of care, with its focus on acute and episodic issues within the office visit, is not designed to meaningfully address long-term patient needs. With COVID-19 has come unprecedented digital adoption, bringing health care delivery to a critical juncture. While digital tools and technologies present vast opportunities for democratizing and decentralizing care experiences, their piecemeal application to the existing "sick care" model and its information technology infrastructure will not only limit their value, but will inevitably add cost, inefficiency, and burden to care teams. In order to build upon this momentum and reap the full benefits of practice digitization, care model transformation must occur. This entails holistically reexamining how every component of the health care experience, from the digital tools to visit interactions, synchronizes to address the full continuum of patient needs throughout the journey. By doing this, care shifts away from one-size-fits-all, fragmented strings of visits, toward seamless experiences that adapt to patients' needs in real-time while integrating within their daily lives. Rather than acting as a substitute for care, technology instead is vital to promoting and amplifying the impact of all those involved. To achieve this, this paper outlines 10 principles for restructuring care to incorporate digital health capabilities. Each describes how all care model components work as a system that aligns with patient needs. By doing this, technology is now an integral in supporting relationships across the full continuum of care.