Artificial intelligence, as the breakthrough of current information technology, is gaining importance and being applied in
more and more industries. Research on the application of artificial intelligence to the medical field has gradually matured
and some products have come out. Relevant products in the United States have been approved by the FDA, mainly for
image-based software products. In this paper, we studied the cases of image-based artificial intelligence aided diagnosis
software approved by the US FDA, and analyzed the listing routes, clinical evaluation methods and clinical data
processing of representative artificial intelligence products, and summarized the clinical evaluation characteristics of FDA
for image-based artificial intelligence aided diagnosis software. Finally, problems that may be encountered in the clinical
evaluation of similar products in China were considered, and relevant suggestions were put forward.
HU Kai, ZHEN Hui, YANG Hui, XIA Jiansong, HE Rui.Study on the Clinical Evaluation of Image-based Artificial Intelligence Aided Diagnosis Software Approved in the United States[J] Chinese Journal of Medical Instrumentation, 2019,V43(5): 379-383
White House Fact Sheet: President Donald J. Trump is accelerating America’s leadership in Artificial Intelligence[R/OL]. (2019-02-11)[2019-03-01]. https://www.energy.gov/articles/white-housefact- sheet-president-donald-j-trump-accelerating-america-sleadership.
MARKUS R, ELLEDGE J, SIMADER C, et al. Evaluation of optical coherence tomography findings in age-related macular degeneration: a reproducibility study of two independent reading centres[J].Br J Ophthalmol.2011, 95 (3):381-385
GRÉGOIRE G. Logisticregression[C]. Statistics for Astrophysics Methods and Applications of the Regression, 2013: 89-120.
Use of real-world evidence to support regulatory decision-making for medical devices[Z]. (2017-08-31)[2019-03-01].
ClinicalTrials.gov is a database of privately and publicly funded clinical studies conducted around the world[Z]. [2019-03-01].https://www.clinicaltrials.gov/.
KE Y, WANG X S, LU L, et al. DeepLesion: automated mining of large-scale lesion annotations and universal lesion detection with deep learning [J]. J Med Imaging, 2018(5): 036501
RAFAEL W, MARTIN B, EKTA D, et al. Agreement of CAD features with expert observer ratings for characterization of pulmonary nodules in CT using the LIDC-IDRI database[C]. Medical Imaging 2009: Computer-Aided Diagnosis, 2009, 20092512133657.