Author Interviews, Brigham & Women's - Harvard, Cancer Research, JAMA, Lancet, Lung Cancer, Medical Imaging, Technology / 07.09.2022
Lung Cancer: Human-AI Collaboration Can Accelerate Time to Treatment
MedicalResearch.com Interview with:
Raymond H. Mak, MD
Radiation Oncology Disease Center Leader for Thoracic Oncology
Director of Patient Safety and QualityDirector of Clinical Innovation
Associate Professor, Harvard Medical School
Cancer - Radiation Oncology, Radiation Oncology
Department of Radiation Oncology
Brigham and Women's Hospital
MedicalResearch.com: What is the background for this study? What is the algorithm detecting?
Response: Lung cancer, the most common cancer worldwide is highly lethal, but can be treated and cured in some cases with radiation therapy. Nearly half of lung cancer patients will eventually require some form of radiation therapy, but the planning for a course of radiation therapy currently entails manual, time-consuming, and resource-intensive work by highly trained physicians to segment (target) the cancerous tumors in the lungs and adjacent lymph nodes on three-dimensional images (CT scans). Prior studies have shown substantial variation in how expert clinicians delineate these targets, which can negatively impact outcomes and there is a projected shortage of skilled medical staff to perform these tasks worldwide as cancer rates increase.
To address this critical gap, our team developed deep learning algorithms that can automatically target lung cancer in the lungs and adjacent lymph nodes from CT scans that are used for radiation therapy planning, and can be deployed in seconds.
We trained these artificial intelligence (AI) algorithms using expert-segmented targets from over 700 cases and validated the performance in over 1300 patients in external datasets (including publicly available data from a national trial), benchmarked its performance against expert clinicians, and then further validated the clinical usefulness of the algorithm in human-AI collaboration experiments that measured accuracy, task speed, and end-user satisfaction.
(more…)