Design and Analysis Methods for Trials with AI-Based Diagnostic Devices for Breast Cancer

Imaging is important in cancer diagnostics. It takes a long period of medical training and clinical experience for radiologists to be able to accurately interpret diagnostic images. With the advance of big data analysis, machine learning and AI-based devices are currently under development and takin...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Lu Liu, Kevin J. Parker, Sin-Ho Jung
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	artificial intelligence (AI) breast cancer clinical device trial concordance rate generalized estimating equation sample size calculation Medicine R
Acceso en línea:	https://doaj.org/article/c35788de3bda478abde59c4b49110efe
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:c35788de3bda478abde59c4b49110efe
record_format	dspace
spelling	oai:doaj.org-article:c35788de3bda478abde59c4b49110efe2021-11-25T18:07:34ZDesign and Analysis Methods for Trials with AI-Based Diagnostic Devices for Breast Cancer10.3390/jpm111111502075-4426https://doaj.org/article/c35788de3bda478abde59c4b49110efe2021-11-01T00:00:00Zhttps://www.mdpi.com/2075-4426/11/11/1150https://doaj.org/toc/2075-4426Imaging is important in cancer diagnostics. It takes a long period of medical training and clinical experience for radiologists to be able to accurately interpret diagnostic images. With the advance of big data analysis, machine learning and AI-based devices are currently under development and taking a role in imaging diagnostics. If an AI-based imaging device can read the image as accurately as experienced radiologists, it may be able to help radiologists increase the accuracy of their reading and manage their workloads. In this paper, we consider two potential study objectives of a clinical trial to evaluate an AI-based device for breast cancer diagnosis by comparing its concordance with human radiologists. We propose statistical design and analysis methods for each study objective. Extensive numerical studies are conducted to show that the proposed statistical testing methods control the type I error rate accurately and the design methods provide required sample sizes with statistical powers close to pre-specified nominal levels. The proposed methods were successfully used to design and analyze a real device trial.Lu LiuKevin J. ParkerSin-Ho JungMDPI AGarticleartificial intelligence (AI)breast cancerclinical device trialconcordance rategeneralized estimating equationsample size calculationMedicineRENJournal of Personalized Medicine, Vol 11, Iss 1150, p 1150 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	artificial intelligence (AI) breast cancer clinical device trial concordance rate generalized estimating equation sample size calculation Medicine R
spellingShingle	artificial intelligence (AI) breast cancer clinical device trial concordance rate generalized estimating equation sample size calculation Medicine R Lu Liu Kevin J. Parker Sin-Ho Jung Design and Analysis Methods for Trials with AI-Based Diagnostic Devices for Breast Cancer
description	Imaging is important in cancer diagnostics. It takes a long period of medical training and clinical experience for radiologists to be able to accurately interpret diagnostic images. With the advance of big data analysis, machine learning and AI-based devices are currently under development and taking a role in imaging diagnostics. If an AI-based imaging device can read the image as accurately as experienced radiologists, it may be able to help radiologists increase the accuracy of their reading and manage their workloads. In this paper, we consider two potential study objectives of a clinical trial to evaluate an AI-based device for breast cancer diagnosis by comparing its concordance with human radiologists. We propose statistical design and analysis methods for each study objective. Extensive numerical studies are conducted to show that the proposed statistical testing methods control the type I error rate accurately and the design methods provide required sample sizes with statistical powers close to pre-specified nominal levels. The proposed methods were successfully used to design and analyze a real device trial.
format	article
author	Lu Liu Kevin J. Parker Sin-Ho Jung
author_facet	Lu Liu Kevin J. Parker Sin-Ho Jung
author_sort	Lu Liu
title	Design and Analysis Methods for Trials with AI-Based Diagnostic Devices for Breast Cancer
title_short	Design and Analysis Methods for Trials with AI-Based Diagnostic Devices for Breast Cancer
title_full	Design and Analysis Methods for Trials with AI-Based Diagnostic Devices for Breast Cancer
title_fullStr	Design and Analysis Methods for Trials with AI-Based Diagnostic Devices for Breast Cancer
title_full_unstemmed	Design and Analysis Methods for Trials with AI-Based Diagnostic Devices for Breast Cancer
title_sort	design and analysis methods for trials with ai-based diagnostic devices for breast cancer
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/c35788de3bda478abde59c4b49110efe
work_keys_str_mv	AT luliu designandanalysismethodsfortrialswithaibaseddiagnosticdevicesforbreastcancer AT kevinjparker designandanalysismethodsfortrialswithaibaseddiagnosticdevicesforbreastcancer AT sinhojung designandanalysismethodsfortrialswithaibaseddiagnosticdevicesforbreastcancer
_version_	1718411642106019840

Design and Analysis Methods for Trials with AI-Based Diagnostic Devices for Breast Cancer

Ejemplares similares