DermX: An end-to-end framework for explainable automated dermatological diagnosis
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
DermX : An end-to-end framework for explainable automated dermatological diagnosis. / Jalaboi, Raluca; Faye, Frederik; Orbes-Arteaga, Mauricio; Jørgensen, Dan; Winther, Ole; Galimzianova, Alfiia.
In: Medical Image Analysis, Vol. 83, 102647, 2023.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - DermX
T2 - An end-to-end framework for explainable automated dermatological diagnosis
AU - Jalaboi, Raluca
AU - Faye, Frederik
AU - Orbes-Arteaga, Mauricio
AU - Jørgensen, Dan
AU - Winther, Ole
AU - Galimzianova, Alfiia
N1 - Publisher Copyright: © 2022 The Author(s)
PY - 2023
Y1 - 2023
N2 - Dermatological diagnosis automation is essential in addressing the high prevalence of skin diseases and critical shortage of dermatologists. Despite approaching expert-level diagnosis performance, convolutional neural network (ConvNet) adoption in clinical practice is impeded by their limited explainability, and by subjective, expensive explainability validations. We introduce DermX, an end-to-end framework for explainable automated dermatological diagnosis. DermX is a clinically-inspired explainable dermatological diagnosis ConvNet, trained using DermXDB, a 554 image dataset annotated by eight dermatologists with diagnoses, supporting explanations, and explanation attention maps. DermX+ extends DermX with guided attention training for explanation attention maps. Both methods achieve near-expert diagnosis performance, with DermX, DermX+, and dermatologist F1 scores of 0.79, 0.79, and 0.87, respectively. We assess the explanation performance in terms of identification and localization by comparing model-selected with dermatologist-selected explanations, and gradient-weighted class-activation maps with dermatologist explanation maps, respectively. DermX obtained an identification F1 score of 0.77, while DermX+ obtained 0.79. The localization F1 score is 0.39 for DermX and 0.35 for DermX+. These results show that explainability does not necessarily come at the expense of predictive power, as our high-performance models provide expert-inspired explanations for their diagnoses without lowering their diagnosis performance.
AB - Dermatological diagnosis automation is essential in addressing the high prevalence of skin diseases and critical shortage of dermatologists. Despite approaching expert-level diagnosis performance, convolutional neural network (ConvNet) adoption in clinical practice is impeded by their limited explainability, and by subjective, expensive explainability validations. We introduce DermX, an end-to-end framework for explainable automated dermatological diagnosis. DermX is a clinically-inspired explainable dermatological diagnosis ConvNet, trained using DermXDB, a 554 image dataset annotated by eight dermatologists with diagnoses, supporting explanations, and explanation attention maps. DermX+ extends DermX with guided attention training for explanation attention maps. Both methods achieve near-expert diagnosis performance, with DermX, DermX+, and dermatologist F1 scores of 0.79, 0.79, and 0.87, respectively. We assess the explanation performance in terms of identification and localization by comparing model-selected with dermatologist-selected explanations, and gradient-weighted class-activation maps with dermatologist explanation maps, respectively. DermX obtained an identification F1 score of 0.77, while DermX+ obtained 0.79. The localization F1 score is 0.39 for DermX and 0.35 for DermX+. These results show that explainability does not necessarily come at the expense of predictive power, as our high-performance models provide expert-inspired explanations for their diagnoses without lowering their diagnosis performance.
KW - Convolutional neural networks
KW - Dataset
KW - Dermatology
KW - Explainability
U2 - 10.1016/j.media.2022.102647
DO - 10.1016/j.media.2022.102647
M3 - Journal article
C2 - 36272237
AN - SCOPUS:85140802997
VL - 83
JO - Medical Image Analysis
JF - Medical Image Analysis
SN - 1361-8415
M1 - 102647
ER -
ID: 327786747