Facial Recognition Intensity in Disease Diagnosis Using Automatic Facial Recognition

Artificial intelligence (AI) technology is widely applied in different medical fields, including the diagnosis of various diseases on the basis of facial phenotypes, but there is no evaluation or quantitative synthesis regarding the performance of artificial intelligence. Here, for the first time, w...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Danning Wu, Shi Chen, Yuelun Zhang, Huabing Zhang, Qing Wang, Jianqiang Li, Yibo Fu, Shirui Wang, Hongbo Yang, Hanze Du, Huijuan Zhu, Hui Pan, Zhen Shen
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	artificial intelligence computer-aided diagnosis facial phenotypes machine learning complexity theory Medicine R
Acceso en línea:	https://doaj.org/article/eca1070c912b4a4788f1460befe4be09
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:eca1070c912b4a4788f1460befe4be09
record_format	dspace
spelling	oai:doaj.org-article:eca1070c912b4a4788f1460befe4be092021-11-25T18:07:46ZFacial Recognition Intensity in Disease Diagnosis Using Automatic Facial Recognition10.3390/jpm111111722075-4426https://doaj.org/article/eca1070c912b4a4788f1460befe4be092021-11-01T00:00:00Zhttps://www.mdpi.com/2075-4426/11/11/1172https://doaj.org/toc/2075-4426Artificial intelligence (AI) technology is widely applied in different medical fields, including the diagnosis of various diseases on the basis of facial phenotypes, but there is no evaluation or quantitative synthesis regarding the performance of artificial intelligence. Here, for the first time, we summarized and quantitatively analyzed studies on the diagnosis of heterogeneous diseases on the basis on facial features. In pooled data from 20 systematically identified studies involving 7 single diseases and 12,557 subjects, quantitative random-effects models revealed a pooled sensitivity of 89% (95% CI 82% to 93%) and a pooled specificity of 92% (95% CI 87% to 95%). A new index, the facial recognition intensity (FRI), was established to describe the complexity of the association of diseases with facial phenotypes. Meta-regression revealed the important contribution of FRI to heterogeneous diagnostic accuracy (<i>p</i> = 0.021), and a similar result was found in subgroup analyses (<i>p</i> = 0.003). An appropriate increase in the training size and the use of deep learning models helped to improve the diagnostic accuracy for diseases with low FRI, although no statistically significant association was found between accuracy and photographic resolution, training size, AI architecture, and number of diseases. In addition, a novel hypothesis is proposed for universal rules in AI performance, providing a new idea that could be explored in other AI applications.Danning WuShi ChenYuelun ZhangHuabing ZhangQing WangJianqiang LiYibo FuShirui WangHongbo YangHanze DuHuijuan ZhuHui PanZhen ShenMDPI AGarticleartificial intelligencecomputer-aided diagnosisfacial phenotypesmachine learningcomplexity theoryMedicineRENJournal of Personalized Medicine, Vol 11, Iss 1172, p 1172 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	artificial intelligence computer-aided diagnosis facial phenotypes machine learning complexity theory Medicine R
spellingShingle	artificial intelligence computer-aided diagnosis facial phenotypes machine learning complexity theory Medicine R Danning Wu Shi Chen Yuelun Zhang Huabing Zhang Qing Wang Jianqiang Li Yibo Fu Shirui Wang Hongbo Yang Hanze Du Huijuan Zhu Hui Pan Zhen Shen Facial Recognition Intensity in Disease Diagnosis Using Automatic Facial Recognition
description	Artificial intelligence (AI) technology is widely applied in different medical fields, including the diagnosis of various diseases on the basis of facial phenotypes, but there is no evaluation or quantitative synthesis regarding the performance of artificial intelligence. Here, for the first time, we summarized and quantitatively analyzed studies on the diagnosis of heterogeneous diseases on the basis on facial features. In pooled data from 20 systematically identified studies involving 7 single diseases and 12,557 subjects, quantitative random-effects models revealed a pooled sensitivity of 89% (95% CI 82% to 93%) and a pooled specificity of 92% (95% CI 87% to 95%). A new index, the facial recognition intensity (FRI), was established to describe the complexity of the association of diseases with facial phenotypes. Meta-regression revealed the important contribution of FRI to heterogeneous diagnostic accuracy (<i>p</i> = 0.021), and a similar result was found in subgroup analyses (<i>p</i> = 0.003). An appropriate increase in the training size and the use of deep learning models helped to improve the diagnostic accuracy for diseases with low FRI, although no statistically significant association was found between accuracy and photographic resolution, training size, AI architecture, and number of diseases. In addition, a novel hypothesis is proposed for universal rules in AI performance, providing a new idea that could be explored in other AI applications.
format	article
author	Danning Wu Shi Chen Yuelun Zhang Huabing Zhang Qing Wang Jianqiang Li Yibo Fu Shirui Wang Hongbo Yang Hanze Du Huijuan Zhu Hui Pan Zhen Shen
author_facet	Danning Wu Shi Chen Yuelun Zhang Huabing Zhang Qing Wang Jianqiang Li Yibo Fu Shirui Wang Hongbo Yang Hanze Du Huijuan Zhu Hui Pan Zhen Shen
author_sort	Danning Wu
title	Facial Recognition Intensity in Disease Diagnosis Using Automatic Facial Recognition
title_short	Facial Recognition Intensity in Disease Diagnosis Using Automatic Facial Recognition
title_full	Facial Recognition Intensity in Disease Diagnosis Using Automatic Facial Recognition
title_fullStr	Facial Recognition Intensity in Disease Diagnosis Using Automatic Facial Recognition
title_full_unstemmed	Facial Recognition Intensity in Disease Diagnosis Using Automatic Facial Recognition
title_sort	facial recognition intensity in disease diagnosis using automatic facial recognition
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/eca1070c912b4a4788f1460befe4be09
work_keys_str_mv	AT danningwu facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition AT shichen facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition AT yuelunzhang facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition AT huabingzhang facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition AT qingwang facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition AT jianqiangli facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition AT yibofu facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition AT shiruiwang facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition AT hongboyang facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition AT hanzedu facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition AT huijuanzhu facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition AT huipan facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition AT zhenshen facialrecognitionintensityindiseasediagnosisusingautomaticfacialrecognition
_version_	1718411654742409216

Facial Recognition Intensity in Disease Diagnosis Using Automatic Facial Recognition

Ejemplares similares