Deep Neural Network for the Prediction of KRAS Genotype in Rectal Cancer
(2022) In Journal of the American College of Surgeons 235(3). p.482-493- Abstract
BACKGROUND: KRAS mutation can alter the treatment plan after resection of colorectal cancer. Despite its importance, the KRAS status of several patients remains unchecked because of the high cost and limited resources. This study developed a deep neural network (DNN) to predict the KRAS genotype using hematoxylin and eosin (H&E)-stained histopathological images. STUDY DESIGN: Three DNNs were created (KRAS_Mob, KRAS_Shuff, and KRAS_Ince) using the structural backbone of the MobileNet, ShuffleNet, and Inception networks, respectively. The Cancer Genome Atlas was screened to extract 49,684 image tiles that were used for deep learning and internal validation. An independent cohort of 43,032 image tiles was used for external validation.... (More)
BACKGROUND: KRAS mutation can alter the treatment plan after resection of colorectal cancer. Despite its importance, the KRAS status of several patients remains unchecked because of the high cost and limited resources. This study developed a deep neural network (DNN) to predict the KRAS genotype using hematoxylin and eosin (H&E)-stained histopathological images. STUDY DESIGN: Three DNNs were created (KRAS_Mob, KRAS_Shuff, and KRAS_Ince) using the structural backbone of the MobileNet, ShuffleNet, and Inception networks, respectively. The Cancer Genome Atlas was screened to extract 49,684 image tiles that were used for deep learning and internal validation. An independent cohort of 43,032 image tiles was used for external validation. The performance was compared with humans, and a virtual cost-saving analysis was done. RESULTS: The KRAS_Mob network (area under the receiver operating curve [AUC] 0.8, 95% CI 0.71 to 0.89) was the best-performing model for predicting the KRAS genotype, followed by the KRAS_Shuff (AUC 0.73, 95% CI 0.62 to 0.84) and KRAS_Ince (AUC 0.71, 95% CI 0.6 to 0.82) networks. Combing the KRAS_Mob and KRAS_Shuff networks as a double prediction approach showed improved performance. KRAS_Mob network accuracy surpassed that of two independent pathologists (AUC 0.79 [95% CI 0.64 to 0.93], 0.51 [95% CI 0.34 to 0.69], and 0.51 (95% CI 0.34 to 0.69]; p < 0.001 for all comparisons). CONCLUSION: The DNN has the potential to predict the KRAS genotype directly from H&E-stained histopathological slide images. As an algorithmic screening method to prioritize patients for laboratory confirmation, such a model might possibly reduce the number of patients screened, resulting in significant test-related time and economic savings.
(Less)
- author
- Ghareeb, Waleed M. ; Draz, Eman ; Madbouly, Khaled ; Hussein, Ahmed H. ; Faisal, Mohammed ; Elkashef, Wagdi ; Emile, Mona Hany ; Edelhamre, Marcus LU ; Kim, Seon Hahn and Emile, Sameh Hany
- author collaboration
- organization
- publishing date
- 2022-09
- type
- Contribution to journal
- publication status
- published
- subject
- in
- Journal of the American College of Surgeons
- volume
- 235
- issue
- 3
- pages
- 12 pages
- publisher
- Elsevier
- external identifiers
-
- scopus:85132334856
- pmid:35972169
- ISSN
- 1879-1190
- DOI
- 10.1097/XCS.0000000000000277
- language
- English
- LU publication?
- yes
- id
- 3f486550-0ace-4e95-8803-1d663f8fab9b
- date added to LUP
- 2023-01-02 13:27:19
- date last changed
- 2024-07-25 18:51:43
@article{3f486550-0ace-4e95-8803-1d663f8fab9b,
abstract = {{<p>BACKGROUND: KRAS mutation can alter the treatment plan after resection of colorectal cancer. Despite its importance, the KRAS status of several patients remains unchecked because of the high cost and limited resources. This study developed a deep neural network (DNN) to predict the KRAS genotype using hematoxylin and eosin (H&E)-stained histopathological images. STUDY DESIGN: Three DNNs were created (KRAS_Mob, KRAS_Shuff, and KRAS_Ince) using the structural backbone of the MobileNet, ShuffleNet, and Inception networks, respectively. The Cancer Genome Atlas was screened to extract 49,684 image tiles that were used for deep learning and internal validation. An independent cohort of 43,032 image tiles was used for external validation. The performance was compared with humans, and a virtual cost-saving analysis was done. RESULTS: The KRAS_Mob network (area under the receiver operating curve [AUC] 0.8, 95% CI 0.71 to 0.89) was the best-performing model for predicting the KRAS genotype, followed by the KRAS_Shuff (AUC 0.73, 95% CI 0.62 to 0.84) and KRAS_Ince (AUC 0.71, 95% CI 0.6 to 0.82) networks. Combing the KRAS_Mob and KRAS_Shuff networks as a double prediction approach showed improved performance. KRAS_Mob network accuracy surpassed that of two independent pathologists (AUC 0.79 [95% CI 0.64 to 0.93], 0.51 [95% CI 0.34 to 0.69], and 0.51 (95% CI 0.34 to 0.69]; p < 0.001 for all comparisons). CONCLUSION: The DNN has the potential to predict the KRAS genotype directly from H&E-stained histopathological slide images. As an algorithmic screening method to prioritize patients for laboratory confirmation, such a model might possibly reduce the number of patients screened, resulting in significant test-related time and economic savings.</p>}},
author = {{Ghareeb, Waleed M. and Draz, Eman and Madbouly, Khaled and Hussein, Ahmed H. and Faisal, Mohammed and Elkashef, Wagdi and Emile, Mona Hany and Edelhamre, Marcus and Kim, Seon Hahn and Emile, Sameh Hany}},
issn = {{1879-1190}},
language = {{eng}},
number = {{3}},
pages = {{482--493}},
publisher = {{Elsevier}},
series = {{Journal of the American College of Surgeons}},
title = {{Deep Neural Network for the Prediction of KRAS Genotype in Rectal Cancer}},
url = {{http://dx.doi.org/10.1097/XCS.0000000000000277}},
doi = {{10.1097/XCS.0000000000000277}},
volume = {{235}},
year = {{2022}},
}