Cargando…
MAC-ResNet: Knowledge Distillation Based Lightweight Multiscale-Attention-Crop-ResNet for Eyelid Tumors Detection and Classification
Eyelid tumors are tumors that occur in the eye and its appendages, affecting vision and appearance, causing blindness and disability, and some having a high lethality rate. Pathological images of eyelid tumors are characterized by large pixels, multiple scales, and similar features. Solving the prob...
Autores principales: | , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9863962/ https://www.ncbi.nlm.nih.gov/pubmed/36675750 http://dx.doi.org/10.3390/jpm13010089 |
_version_ | 1784875465260924928 |
---|---|
author | Huang, Xingru Yao, Chunlei Xu, Feng Chen, Lingxiao Wang, Huaqiong Chen, Xiaodiao Ye, Juan Wang, Yaqi |
author_facet | Huang, Xingru Yao, Chunlei Xu, Feng Chen, Lingxiao Wang, Huaqiong Chen, Xiaodiao Ye, Juan Wang, Yaqi |
author_sort | Huang, Xingru |
collection | PubMed |
description | Eyelid tumors are tumors that occur in the eye and its appendages, affecting vision and appearance, causing blindness and disability, and some having a high lethality rate. Pathological images of eyelid tumors are characterized by large pixels, multiple scales, and similar features. Solving the problem of difficult and time-consuming fine-grained classification of pathological images is important to improve the efficiency and quality of pathological diagnosis. The morphology of Basal Cell Carcinoma (BCC), Meibomian Gland Carcinoma (MGC), and Cutaneous Melanoma (CM) in eyelid tumors are very similar, and it is easy to be misdiagnosed among each category. In addition, the diseased area, which is decisive for the diagnosis of the disease, usually occupies only a relatively minor portion of the entire pathology section, and screening the area of interest is a tedious and time-consuming task. In this paper, deep learning techniques to investigate the pathological images of eyelid tumors. Inspired by the knowledge distillation process, we propose the Multiscale-Attention-Crop-ResNet (MAC-ResNet) network model to achieve the automatic classification of three malignant tumors and the automatic localization of whole slide imaging (WSI) lesion regions using U-Net. The final accuracy rates of the three classification problems of eyelid tumors on MAC-ResNet were 96.8%, 94.6%, and 90.8%, respectively. |
format | Online Article Text |
id | pubmed-9863962 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2022 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-98639622023-01-22 MAC-ResNet: Knowledge Distillation Based Lightweight Multiscale-Attention-Crop-ResNet for Eyelid Tumors Detection and Classification Huang, Xingru Yao, Chunlei Xu, Feng Chen, Lingxiao Wang, Huaqiong Chen, Xiaodiao Ye, Juan Wang, Yaqi J Pers Med Article Eyelid tumors are tumors that occur in the eye and its appendages, affecting vision and appearance, causing blindness and disability, and some having a high lethality rate. Pathological images of eyelid tumors are characterized by large pixels, multiple scales, and similar features. Solving the problem of difficult and time-consuming fine-grained classification of pathological images is important to improve the efficiency and quality of pathological diagnosis. The morphology of Basal Cell Carcinoma (BCC), Meibomian Gland Carcinoma (MGC), and Cutaneous Melanoma (CM) in eyelid tumors are very similar, and it is easy to be misdiagnosed among each category. In addition, the diseased area, which is decisive for the diagnosis of the disease, usually occupies only a relatively minor portion of the entire pathology section, and screening the area of interest is a tedious and time-consuming task. In this paper, deep learning techniques to investigate the pathological images of eyelid tumors. Inspired by the knowledge distillation process, we propose the Multiscale-Attention-Crop-ResNet (MAC-ResNet) network model to achieve the automatic classification of three malignant tumors and the automatic localization of whole slide imaging (WSI) lesion regions using U-Net. The final accuracy rates of the three classification problems of eyelid tumors on MAC-ResNet were 96.8%, 94.6%, and 90.8%, respectively. MDPI 2022-12-29 /pmc/articles/PMC9863962/ /pubmed/36675750 http://dx.doi.org/10.3390/jpm13010089 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Huang, Xingru Yao, Chunlei Xu, Feng Chen, Lingxiao Wang, Huaqiong Chen, Xiaodiao Ye, Juan Wang, Yaqi MAC-ResNet: Knowledge Distillation Based Lightweight Multiscale-Attention-Crop-ResNet for Eyelid Tumors Detection and Classification |
title | MAC-ResNet: Knowledge Distillation Based Lightweight Multiscale-Attention-Crop-ResNet for Eyelid Tumors Detection and Classification |
title_full | MAC-ResNet: Knowledge Distillation Based Lightweight Multiscale-Attention-Crop-ResNet for Eyelid Tumors Detection and Classification |
title_fullStr | MAC-ResNet: Knowledge Distillation Based Lightweight Multiscale-Attention-Crop-ResNet for Eyelid Tumors Detection and Classification |
title_full_unstemmed | MAC-ResNet: Knowledge Distillation Based Lightweight Multiscale-Attention-Crop-ResNet for Eyelid Tumors Detection and Classification |
title_short | MAC-ResNet: Knowledge Distillation Based Lightweight Multiscale-Attention-Crop-ResNet for Eyelid Tumors Detection and Classification |
title_sort | mac-resnet: knowledge distillation based lightweight multiscale-attention-crop-resnet for eyelid tumors detection and classification |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9863962/ https://www.ncbi.nlm.nih.gov/pubmed/36675750 http://dx.doi.org/10.3390/jpm13010089 |
work_keys_str_mv | AT huangxingru macresnetknowledgedistillationbasedlightweightmultiscaleattentioncropresnetforeyelidtumorsdetectionandclassification AT yaochunlei macresnetknowledgedistillationbasedlightweightmultiscaleattentioncropresnetforeyelidtumorsdetectionandclassification AT xufeng macresnetknowledgedistillationbasedlightweightmultiscaleattentioncropresnetforeyelidtumorsdetectionandclassification AT chenlingxiao macresnetknowledgedistillationbasedlightweightmultiscaleattentioncropresnetforeyelidtumorsdetectionandclassification AT wanghuaqiong macresnetknowledgedistillationbasedlightweightmultiscaleattentioncropresnetforeyelidtumorsdetectionandclassification AT chenxiaodiao macresnetknowledgedistillationbasedlightweightmultiscaleattentioncropresnetforeyelidtumorsdetectionandclassification AT yejuan macresnetknowledgedistillationbasedlightweightmultiscaleattentioncropresnetforeyelidtumorsdetectionandclassification AT wangyaqi macresnetknowledgedistillationbasedlightweightmultiscaleattentioncropresnetforeyelidtumorsdetectionandclassification |