Deep Transfer Learning-Based Approach for Glucose Transporter-1 (GLUT1) Expression Assessment
-
Published:2023-09-05
Issue:6
Volume:36
Page:2367-2381
-
ISSN:0897-1889
-
Container-title:Journal of Digital Imaging
-
language:en
-
Short-container-title:J Digit Imaging
Author:
Al Zorgani Maisun MohamedORCID, Ugail Hassan, Pors Klaus, Dauda Abdullahi Magaji
Abstract
Abstract
Glucose transporter-1 (GLUT-1) expression level is a biomarker of tumour hypoxia condition in immunohistochemistry (IHC)-stained images. Thus, the GLUT-1 scoring is a routine procedure currently employed for predicting tumour hypoxia markers in clinical practice. However, visual assessment of GLUT-1 scores is subjective and consequently prone to inter-pathologist variability. Therefore, this study proposes an automated method for assessing GLUT-1 scores in IHC colorectal carcinoma images. For this purpose, we leverage deep transfer learning methodologies for evaluating the performance of six different pre-trained convolutional neural network (CNN) architectures: AlexNet, VGG16, GoogleNet, ResNet50, DenseNet-201 and ShuffleNet. The target CNNs are fine-tuned as classifiers or adapted as feature extractors with support vector machine (SVM) to classify GLUT-1 scores in IHC images. Our experimental results show that the winning model is the trained SVM classifier on the extracted deep features fusion Feat-Concat from DenseNet201, ResNet50 and GoogLeNet extractors. It yields the highest prediction accuracy of 98.86%, thus outperforming the other classifiers on our dataset. We also conclude, from comparing the methodologies, that the off-the-shelf feature extraction is better than the fine-tuning model in terms of time and resources required for training.
Publisher
Springer Science and Business Media LLC
Subject
Computer Science Applications,Radiology, Nuclear Medicine and imaging,Radiological and Ultrasound Technology
Reference44 articles.
1. Brown, J. M., Wilson, W.R. Exploiting tumour hypoxia in cancer treatment, Nature Reviews Cancer, vol.4(6), pp.437,(2004). 2. Mirabello, V., Cortezon-Tamarit, F. and Pascu, S.I., 2018. Oxygen sensing, hypoxia tracing and in vivo imaging with functional metalloprobes for the early detection of non-communicable diseases, Frontiers in chemistry, vol.6, p.27, (2018). 3. Lepp{\"a}nen, J., Helminen, O., Huhta, H., Kauppila, J.H., Isohookana, J., Haapasaari, K.M., Karihtala, P., Parkkila, S., Saarnio, J., Lehenkari, P.P. and Karttunen, T.J., 2018. Toll‐like receptors 2, 4 and 9 and hypoxia markers HIF‐1alpha and CAIX in pancreatic intraepithelial neoplasia, Wiley Online Library, Apmis, vol.126(11), pp.852–863, (2018). 4. Bhandari, V., Hoey, C., Liu, L.Y., Lalonde, E., Ray, J., Livingstone, J., Lesurf, R., Shiah, Y.J., Vujcic, T., Huang, X. and Espiritu, S.M. Molecular landmarks of tumor hypoxia across cancer types, Nature Publishing Group, Nature genetics, vol.51(2), pp.308-318, (2019). 5. Godet, I., Shin, Y.J., Ju, J.A., Ye, I.C., Wang, G. and Gilkes, D.M. Fate-mapping post-hypoxic tumor cells reveals a ROS-resistant phenotype that promotes metastasis, Nature communications, Nature Publishing Group, Nature communications, vol.10(1), pp.1-18, (2019).
|
|