Recognition of Cutaneous Melanoma on Digitized Histopathological Slides via Artificial Intelligence Algorithm
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 overestim...
Main Authors: | , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2020-08-01
|
Series: | Frontiers in Oncology |
Subjects: | |
Online Access: | https://www.frontiersin.org/article/10.3389/fonc.2020.01559/full |
id |
doaj-3ffd9a2a6c3141f78ffb0dcf65f84105 |
---|---|
record_format |
Article |
spelling |
doaj-3ffd9a2a6c3141f78ffb0dcf65f841052020-11-25T02:53:01ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2020-08-011010.3389/fonc.2020.01559565026Recognition of Cutaneous Melanoma on Digitized Histopathological Slides via Artificial Intelligence AlgorithmFrancesco De Logu0Filippo Ugolini1Vincenza Maio2Sara Simi3Antonio Cossu4Daniela Massi5Italian Association for Cancer Research (AIRC) Study GroupRomina Nassini6Marco Laurino7Section of Clinical Pharmacology and Oncology, Department of Health Sciences, University of Florence, Florence, ItalySection of Pathological Anatomy, Department of Health Sciences, University of Florence, Florence, ItalyHistopathology and Molecular Diagnostics, Careggi University Hospital, Florence, ItalySection of Pathological Anatomy, Department of Health Sciences, University of Florence, Florence, ItalyDepartment of Medical, Surgical, and Experimental Sciences, University of Sassari, Sassari, ItalySection of Pathological Anatomy, Department of Health Sciences, University of Florence, Florence, ItalySection of Clinical Pharmacology and Oncology, Department of Health Sciences, University of Florence, Florence, ItalyInstitute of Clinical Physiology, National Research Council, Pisa, ItalyIncreasing 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.https://www.frontiersin.org/article/10.3389/fonc.2020.01559/fullcutaneous melanomaartificial intelligenceconvolutional neural networkimage analysisdiagnosis |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Francesco De Logu Filippo Ugolini Vincenza Maio Sara Simi Antonio Cossu Daniela Massi Italian Association for Cancer Research (AIRC) Study Group Romina Nassini Marco Laurino |
spellingShingle |
Francesco De Logu Filippo Ugolini Vincenza Maio Sara Simi Antonio Cossu Daniela Massi Italian Association for Cancer Research (AIRC) Study Group Romina Nassini Marco Laurino Recognition of Cutaneous Melanoma on Digitized Histopathological Slides via Artificial Intelligence Algorithm Frontiers in Oncology cutaneous melanoma artificial intelligence convolutional neural network image analysis diagnosis |
author_facet |
Francesco De Logu Filippo Ugolini Vincenza Maio Sara Simi Antonio Cossu Daniela Massi Italian Association for Cancer Research (AIRC) Study Group Romina Nassini Marco Laurino |
author_sort |
Francesco De Logu |
title |
Recognition of Cutaneous Melanoma on Digitized Histopathological Slides via Artificial Intelligence Algorithm |
title_short |
Recognition of Cutaneous Melanoma on Digitized Histopathological Slides via Artificial Intelligence Algorithm |
title_full |
Recognition of Cutaneous Melanoma on Digitized Histopathological Slides via Artificial Intelligence Algorithm |
title_fullStr |
Recognition of Cutaneous Melanoma on Digitized Histopathological Slides via Artificial Intelligence Algorithm |
title_full_unstemmed |
Recognition of Cutaneous Melanoma on Digitized Histopathological Slides via Artificial Intelligence Algorithm |
title_sort |
recognition of cutaneous melanoma on digitized histopathological slides via artificial intelligence algorithm |
publisher |
Frontiers Media S.A. |
series |
Frontiers in Oncology |
issn |
2234-943X |
publishDate |
2020-08-01 |
description |
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. |
topic |
cutaneous melanoma artificial intelligence convolutional neural network image analysis diagnosis |
url |
https://www.frontiersin.org/article/10.3389/fonc.2020.01559/full |
work_keys_str_mv |
AT francescodelogu recognitionofcutaneousmelanomaondigitizedhistopathologicalslidesviaartificialintelligencealgorithm AT filippougolini recognitionofcutaneousmelanomaondigitizedhistopathologicalslidesviaartificialintelligencealgorithm AT vincenzamaio recognitionofcutaneousmelanomaondigitizedhistopathologicalslidesviaartificialintelligencealgorithm AT sarasimi recognitionofcutaneousmelanomaondigitizedhistopathologicalslidesviaartificialintelligencealgorithm AT antoniocossu recognitionofcutaneousmelanomaondigitizedhistopathologicalslidesviaartificialintelligencealgorithm AT danielamassi recognitionofcutaneousmelanomaondigitizedhistopathologicalslidesviaartificialintelligencealgorithm AT italianassociationforcancerresearchaircstudygroup recognitionofcutaneousmelanomaondigitizedhistopathologicalslidesviaartificialintelligencealgorithm AT rominanassini recognitionofcutaneousmelanomaondigitizedhistopathologicalslidesviaartificialintelligencealgorithm AT marcolaurino recognitionofcutaneousmelanomaondigitizedhistopathologicalslidesviaartificialintelligencealgorithm |
_version_ |
1724727213114261504 |