Journal Article
. 2020 Oct; 10:1559.
doi: 10.3389/fonc.2020.01559.

Recognition of Cutaneous Melanoma on Digitized Histopathological Slides via Artificial Intelligence Algorithm

Francesco De Logu 1 Filippo Ugolini 2 Vincenza Maio 3 Sara Simi 2 Antonio Cossu 4 Daniela Massi 2 Italian Association for Cancer Research (AIRC) Study Group  Romina Nassini 1 Marco Laurino 5 
Affiliations
  • PMID: 33014803
  •     24 References
  •     1 citations

Abstract

Increasing incidence of skin cancer combined with a shortage of dermatopathologists has increased the workload of pathology departments worldwide. In addition, the high intraobserver and interobserver variability in the assessment of melanocytic skin lesions can result in underestimated or overestimated diagnosis of melanoma. Thus, the development of new techniques for skin tumor diagnosis is essential to assist pathologists to standardize diagnoses and plan accurate patient treatment. Here, we describe the development of an artificial intelligence (AI) system that recognizes cutaneous melanoma from histopathological digitalized slides with clinically acceptable accuracy. Whole-slide digital images from 100 formalin-fixed paraffin-embedded primary cutaneous melanoma were used to train a convolutional neural network (CNN) based on a pretrained Inception-ResNet-v2 to accurately and automatically differentiate tumoral areas from healthy tissue. The CNN was trained by using 60 digital slides in which regions of interest (ROIs) of tumoral and healthy tissue were extracted by experienced dermatopathologists, while the other 40 slides were used as test datasets. A total of 1377 patches of healthy tissue and 2141 patches of melanoma were assessed in the training/validation set, while 791 patches of healthy tissue and 1122 patches of pathological tissue were evaluated in the test dataset. Considering the classification by expert dermatopathologists as reference, the trained deep net showed high accuracy (96.5%), sensitivity (95.7%), specificity (97.7%), F1 score (96.5%), and a Cohen's kappa of 0.929. Our data show that a deep learning system can be trained to recognize melanoma samples, achieving accuracies comparable to experienced dermatopathologists. Such an approach can offer a valuable aid in improving diagnostic efficiency when expert consultation is not available, as well as reducing interobserver variability. Further studies in larger data sets are necessary to verify whether the deep learning algorithm allows subclassification of different melanoma subtypes.

Keywords: artificial intelligence; convolutional neural network; cutaneous melanoma; diagnosis; image analysis.

Convolutional neural networks for computer-aided detection or diagnosis in medical image analysis: An overview.
Jun Gao, Qian Jiang, Bo Zhou, Daozheng Chen.
Math Biosci Eng, 2019 Nov 09; 16(6). PMID: 31698575
Review.
Learning transferable features in deep convolutional neural networks for diagnosing unseen machine conditions.
Te Han, Chao Liu, Wenguang Yang, Dongxiang Jiang.
ISA Trans, 2019 Apr 03; 93. PMID: 30935654
Pathologist-level classification of histopathological melanoma images with deep neural networks.
Achim Hekler, Jochen Sven Utikal, +10 authors, Titus Josef Brinker.
Eur J Cancer, 2019 May 28; 115. PMID: 31129383
Deep Learning for Whole Slide Image Analysis: An Overview.
Neofytos Dimitriou, Ognjen Arandjelović, Peter D Caie.
Front Med (Lausanne), 2019 Dec 12; 6. PMID: 31824952    Free PMC article.
Review.
Deep learning outperformed 11 pathologists in the classification of histopathological melanoma images.
Achim Hekler, Jochen S Utikal, +12 authors, Titus J Brinker.
Eur J Cancer, 2019 Jul 22; 118. PMID: 31325876
Not Just Digital Pathology, Intelligent Digital Pathology.
Balazs Acs, David L Rimm.
JAMA Oncol, 2018 Feb 03; 4(3). PMID: 29392271
The 2018 World Health Organization Classification of Cutaneous, Mucosal, and Uveal Melanoma: Detailed Analysis of 9 Distinct Subtypes Defined by Their Evolutionary Pathway.
David E Elder, Boris C Bastian, +2 authors, Richard A Scolyer.
Arch Pathol Lab Med, 2020 Feb 15; 144(4). PMID: 32057276
Review.
Impact of JPEG 2000 compression on deep convolutional neural networks for metastatic cancer detection in histopathological images.
Farhad Ghazvinian Zanjani, Svitlana Zinger, +3 authors, Peter H N de With.
J Med Imaging (Bellingham), 2019 May 01; 6(2). PMID: 31037247    Free PMC article.
Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning.
Hoo-Chang Shin, Holger R Roth, +6 authors, Ronald M Summers.
IEEE Trans Med Imaging, 2016 Feb 18; 35(5). PMID: 26886976    Free PMC article.
Highly Cited.
Novel Approaches for Diagnosing Melanoma Skin Lesions Through Supervised and Deep Learning Algorithms.
J Premaladha, K S Ravichandran.
J Med Syst, 2016 Feb 15; 40(4). PMID: 26872778
Medical Image Analysis using Convolutional Neural Networks: A Review.
Syed Muhammad Anwar, Muhammad Majid, +3 authors, Muhammad Khurram Khan.
J Med Syst, 2018 Oct 10; 42(11). PMID: 30298337
Review.
Image analysis and machine learning in digital pathology: Challenges and opportunities.
Anant Madabhushi, George Lee.
Med Image Anal, 2016 Jul 18; 33. PMID: 27423409    Free PMC article.
Highly Cited. Review.
Cutaneous malignant melanoma: update on diagnostic and prognostic biomarkers.
Ossama Abbas, Daniel D Miller, Jag Bhawan.
Am J Dermatopathol, 2014 May 08; 36(5). PMID: 24803061
Review.
Artificial Intelligence in Medicine: Where Are We Now?
Sagar Kulkarni, Nuran Seneviratne, Mirza Shaheer Baig, Ameer Hamid Ahmed Khan.
Acad Radiol, 2019 Oct 23; 27(1). PMID: 31636002
Review.
Dermatologist-level classification of skin cancer with deep neural networks.
Andre Esteva, Brett Kuprel, +4 authors, Sebastian Thrun.
Nature, 2017 Jan 25; 542(7639). PMID: 28117445    Free PMC article.
Highly Cited.
A comparison of machine learning methods for the diagnosis of pigmented skin lesions.
S Dreiseitl, L Ohno-Machado, +3 authors, M Binder.
J Biomed Inform, 2001 May 30; 34(1). PMID: 11376540
Discordance in the histopathologic diagnosis of melanoma at a melanoma referral center.
B Aika Shoo, Richard W Sagebiel, Mohammed Kashani-Sabet.
J Am Acad Dermatol, 2010 Mar 23; 62(5). PMID: 20303612
Pathologists' diagnosis of invasive melanoma and melanocytic proliferations: observer accuracy and reproducibility study.
Joann G Elmore, Raymond L Barnhill, +11 authors, Michael W Piepkorn.
BMJ, 2017 Jul 01; 357. PMID: 28659278    Free PMC article.
Highly Cited.
Deep Convolutional Neural Networks Enable Discrimination of Heterogeneous Digital Pathology Images.
Pegah Khosravi, Ehsan Kazemi, +2 authors, Iman Hajirasouliha.
EBioMedicine, 2018 Jan 03; 27. PMID: 29292031    Free PMC article.
Highly Cited.
An open source automated tumor infiltrating lymphocyte algorithm for prognosis in melanoma.
Balazs Acs, Fahad Shabbir Ahmed, +7 authors, David L Rimm.
Nat Commun, 2019 Dec 01; 10(1). PMID: 31784511    Free PMC article.
Machine learning-based diagnosis of melanoma using macro images.
Diwakar Gautam, Mushtaq Ahmed, Yogesh Kumar Meena, Ahtesham Ul Haq.
Int J Numer Method Biomed Eng, 2017 Dec 22; 34(5). PMID: 29266819
Cancer Statistics, 2017.
Rebecca L Siegel, Kimberly D Miller, Ahmedin Jemal.
CA Cancer J Clin, 2017 Jan 06; 67(1). PMID: 28055103
Highly Cited.
Automated identification of malignancy in whole-slide pathological images: identification of eyelid malignant melanoma in gigapixel pathological slides using deep learning.
Linyan Wang, Longqian Ding, +6 authors, Juan Ye.
Br J Ophthalmol, 2019 Jul 16; 104(3). PMID: 31302629
Machine learning approaches for pathologic diagnosis.
Daisuke Komura, Shumpei Ishikawa.
Virchows Arch, 2019 Jun 22; 475(2). PMID: 31222375
Review.
Spatial architecture of the immune microenvironment orchestrates tumor immunity and therapeutic response.
Tong Fu, Lei-Jie Dai, +4 authors, Zhi-Ming Shao.
J Hematol Oncol, 2021 Jun 27; 14(1). PMID: 34172088    Free PMC article.
Review.