Article Text

Download PDFPDF
Automated identification of malignancy in whole-slide pathological images: identification of eyelid malignant melanoma in gigapixel pathological slides using deep learning
  1. Linyan Wang1,
  2. Longqian Ding2,
  3. Zhifang Liu1,
  4. Lingling Sun2,
  5. Lirong Chen3,
  6. Renbing Jia4,
  7. Xizhe Dai1,
  8. Jing Cao1,
  9. Juan Ye1
  1. 1 Department of Ophthalmology, The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
  2. 2 Key Laboratory of RF Circuits and Systems, Ministry of Education, Hangzhou Dianzi University, Hangzhou, China
  3. 3 Department of Pathology, The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
  4. 4 Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
  1. Correspondence to Juan Ye, Department of Ophthalmology, The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310009, China; yejuan{at}zju.edu.cn

Abstract

Background/Aims To develop a deep learning system (DLS) that can automatically detect malignant melanoma (MM) in the eyelid from histopathological sections with colossal information density.

Methods Setting: Double institutional study.

Study population: We retrospectively reviewed 225 230 pathological patches (small section cut from pathologist-labelled area from an H&E image), cut from 155 H&E-stained whole-slide images (WSI).

Observation procedures: Labelled gigapixel pathological WSIs were used to train and test a model designed to assign patch-level classification. Using malignant probability from a convolutional neural network, the patches were embedded back into each WSI to generate a visualisation heatmap and leveraged a random forest model to establish a WSI-level diagnosis.

Main outcome measure(s): For classification, the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity and specificity were used to evaluate the efficacy of the DLS in detecting MM.

Results For patch diagnosis, the model achieved an AUC of 0.989 (95% CI 0.989 to 0.991), with an accuracy, sensitivity and specificity of 94.9%, 94.7% and 95.3%, respectively. We displayed the lesion area on the WSIs as graded by malignant potential. For WSI, the obtained sensitivity, specificity and accuracy were 100%, 96.5% and 98.2%, respectively, with an AUC of 0.998 (95% CI 0.994 to 1.000).

Conclusion Our DLS, which uses artificial intelligence, can automatically detect MM in histopathological slides and highlight the lesion area on WSIs using a probabilistic heatmap. In addition, our approach has the potential to be applied to the histopathological sections of other tumour types.

  • eyelids
  • pathology
  • telemedicine

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Footnotes

  • Contributors JY contributed to the design and took part in the whole study. LW contributed to the literature review, data collection and paper writing. LD and LS contributed to algorithm development and figure designs. LC and RJ provided data. JC and XD contributed to data collection and analysis.

  • Funding The study was supported by the National Natural Science Foundation of China (grant 81670888).

  • Competing interests None declared.

  • Ethics approval The study was approved by The Second Affiliated Hospital, Zhejiang University School of Medicine (ZJU-2) Ethics Committee (No 2018-347) and adhered to the tenets of the Declaration of Helsinki.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement Data may be obtained from a third party and are not publicly available.

Linked Articles

  • At a glance
    Keith Barton James Chodosh Jost B Jonas