Deep learning saliency maps do not accurately highlight diagnostically relevant regions for medical image interpretation

Adriel Saporta; Xiaotong Gui; Ashwin Agrawal; Anuj Pareek; Steven Q.H. Truong; Chanh D.T. Nguyen; Van Doan Ngo; Jayne Seekins; Francis G. Blankenberg; Andrew Y. Ng; Matthew P Lundgren; Pranav Rajpurkar

Deep learning saliency maps do not accurately highlight diagnostically relevant regions for medical image interpretation

Adriel Saporta, Xiaotong Gui, Ashwin Agrawal, Anuj Pareek, Steven Q.H. Truong, Chanh D.T. Nguyen, Van Doan Ngo, Jayne Seekins, Francis G. Blankenberg, Andrew Y. Ng, Matthew P Lundgren, Pranav Rajpurkar

Original language	English
Publisher	medRxiv
Publication status	Published - 2021
Externally published	Yes

Cite this

@techreport{976678d7a20e4d9ca8192648d096548c,

title = "Deep learning saliency maps do not accurately highlight diagnostically relevant regions for medical image interpretation",

author = "Adriel Saporta and Xiaotong Gui and Ashwin Agrawal and Anuj Pareek and Truong, {Steven Q.H.} and Nguyen, {Chanh D.T.} and Ngo, {Van Doan} and Jayne Seekins and Blankenberg, {Francis G.} and Ng, {Andrew Y.} and Lundgren, {Matthew P} and Pranav Rajpurkar",

year = "2021",

language = "English",

publisher = "medRxiv",

type = "WorkingPaper",

institution = "medRxiv",

}

TY - UNPB

T1 - Deep learning saliency maps do not accurately highlight diagnostically relevant regions for medical image interpretation

AU - Saporta, Adriel

AU - Gui, Xiaotong

AU - Agrawal, Ashwin

AU - Pareek, Anuj

AU - Truong, Steven Q.H.

AU - Nguyen, Chanh D.T.

AU - Ngo, Van Doan

AU - Seekins, Jayne

AU - Blankenberg, Francis G.

AU - Ng, Andrew Y.

AU - Lundgren, Matthew P

AU - Rajpurkar, Pranav

PY - 2021

Y1 - 2021

M3 - Preprint

BT - Deep learning saliency maps do not accurately highlight diagnostically relevant regions for medical image interpretation

PB - medRxiv

ER -