Identifying gross post-mortem organ images using a pre-trained convolutional neural network
Standard
Identifying gross post-mortem organ images using a pre-trained convolutional neural network. / Garland, Jack; Hu, Mindy; Kesha, Kilak; Glenn, Charley; Morrow, Paul; Stables, Simon; Ondruschka, Benjamin; Tse, Rexson.
In: J FORENSIC SCI, Vol. 66, No. 2, 03.2021, p. 630-635.Research output: SCORING: Contribution to journal › SCORING: Journal article › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - JOUR
T1 - Identifying gross post-mortem organ images using a pre-trained convolutional neural network
AU - Garland, Jack
AU - Hu, Mindy
AU - Kesha, Kilak
AU - Glenn, Charley
AU - Morrow, Paul
AU - Stables, Simon
AU - Ondruschka, Benjamin
AU - Tse, Rexson
N1 - © 2020 American Academy of Forensic Sciences.
PY - 2021/3
Y1 - 2021/3
N2 - Identifying organs/tissue and pathology on radiological and microscopic images can be performed using convolutional neural networks (CNN). However, there are scant studies on applying CNN to post-mortem gross images of visceral organs. This proof-of-concept study used 537 gross post-mortem images of dissected brain, heart, lung, liver, spleen, and kidney, which were randomly divided into a training and teaching datasets for the pre-trained CNN Xception. The CNN was trained using the training dataset and subsequently tested on the testing dataset. The overall accuracies were >95% percent for both training and testing datasets and have an F1 score of >0.95 for all dissected organs. This study showed that small datasets of post-mortem images can be classified with a very high accuracy using a pre-trained CNN. This novel area has the potential for future application in data mining, education and teaching, case review, research, quality assurance, auditing purposes, and identifying pathology.
AB - Identifying organs/tissue and pathology on radiological and microscopic images can be performed using convolutional neural networks (CNN). However, there are scant studies on applying CNN to post-mortem gross images of visceral organs. This proof-of-concept study used 537 gross post-mortem images of dissected brain, heart, lung, liver, spleen, and kidney, which were randomly divided into a training and teaching datasets for the pre-trained CNN Xception. The CNN was trained using the training dataset and subsequently tested on the testing dataset. The overall accuracies were >95% percent for both training and testing datasets and have an F1 score of >0.95 for all dissected organs. This study showed that small datasets of post-mortem images can be classified with a very high accuracy using a pre-trained CNN. This novel area has the potential for future application in data mining, education and teaching, case review, research, quality assurance, auditing purposes, and identifying pathology.
U2 - 10.1111/1556-4029.14608
DO - 10.1111/1556-4029.14608
M3 - SCORING: Journal article
C2 - 33105027
VL - 66
SP - 630
EP - 635
JO - J FORENSIC SCI
JF - J FORENSIC SCI
SN - 0022-1198
IS - 2
ER -