Clinically Relevant Vulnerabilities of Deep Machine Learning Systems for Skin Cancer Diagnosis.
作者信息
Du-Harpur Xinyi, Arthurs Callum, Ganier Clarisse, Woolf Rick, Laftah Zainab, Lakhan Manpreet, Salam Amr, Wan Bo, Watt Fiona M, Luscombe Nicholas M, Lynch Magnus D
机构信息
Centre for Stem Cells and Regenerative Medicine, King's College London, London, United Kingdom; Bioinformatics and Computational Biology Laboratory, The Francis Crick Institute, London, United Kingdom; St John's Institute of Dermatology, Guys Hospital, London, United Kingdom.
Centre for Stem Cells and Regenerative Medicine, King's College London, London, United Kingdom.
出版信息
J Invest Dermatol. 2021 Apr;141(4):916-920. doi: 10.1016/j.jid.2020.07.034. Epub 2020 Sep 12.
Abstract
摘要
相似文献
1
Clinically Relevant Vulnerabilities of Deep Machine Learning Systems for Skin Cancer Diagnosis.用于皮肤癌诊断的深度机器学习系统的临床相关漏洞
J Invest Dermatol. 2021 Apr;141(4):916-920. doi: 10.1016/j.jid.2020.07.034. Epub 2020 Sep 12.
2
Dermoscopy: basic concepts.皮肤镜检查：基本概念。
Int J Dermatol. 2008 Jul;47(7):712-9. doi: 10.1111/j.1365-4632.2008.03556.x.
3
Dermoscopy and the negative pigment network.皮肤镜检查与阴性色素网
J Am Acad Dermatol. 2014 Feb;70(2):e29-30. doi: 10.1016/j.jaad.2013.09.003.
4
Site-specific acral nevus histologically reminiscent of melanoma: Recognition of the utility of the Fontana-Masson stain.组织学上类似黑色素瘤的特定部位肢端痣：认识Fontana-Masson染色的效用
J Dermatol. 2019 May;46(5):e183-e185. doi: 10.1111/1346-8138.14735. Epub 2018 Dec 14.
5
Failure to compare dermoscopy findings of pigmented lesions on your patient: Comment on "Dermoscopy of patients with multiple nevi".未对你患者色素性皮损的皮肤镜检查结果进行比较：评《多发性痣患者的皮肤镜检查》
Arch Dermatol. 2011 Jan;147(1):50. doi: 10.1001/archdermatol.2010.390.
6
Image Gallery: In vivo fluorescence-advanced videodermatoscopy for the characterization of skin melanocytic pigmented lesions.图片库：用于皮肤黑素细胞色素沉着病变特征描述的体内荧光-高级视频皮肤镜检查
Br J Dermatol. 2019 Apr;180(4):e104. doi: 10.1111/bjd.17457.
7
Computer algorithms show potential for improving dermatologists' accuracy to diagnose cutaneous melanoma: Results of the International Skin Imaging Collaboration 2017.计算机算法显示出提高皮肤科医生诊断皮肤黑色素瘤准确性的潜力：国际皮肤成像协作 2017 年的研究结果。
J Am Acad Dermatol. 2020 Mar;82(3):622-627. doi: 10.1016/j.jaad.2019.07.016. Epub 2019 Jul 12.
8
Three dermoscopic signs of growth of pigmented lesions.色素性皮损生长的三种皮肤镜征象。
J Am Acad Dermatol. 2015 Oct;73(4):e133-5. doi: 10.1016/j.jaad.2015.06.032.
9
Tape stripping: A very short-term follow-up procedure for suspicious black lesions.胶带剥离法：针对可疑黑色病变的一种短期随访程序。
J Am Acad Dermatol. 2015 Jun;72(6):e151-2. doi: 10.1016/j.jaad.2015.02.1098.
10
Congenital nevus of the nail apparatus--diagnostic approach of a case through dermoscopy.甲器先天性痣——通过皮肤镜检查对一例病例的诊断方法
Pediatr Dermatol. 2013 Nov-Dec;30(6):e293-4. doi: 10.1111/j.1525-1470.2012.01813.x. Epub 2012 Jul 24.
引用本文的文献
1
So You've Got a High AUC, Now What? An Overview of Important Considerations when Bringing Machine-Learning Models from Computer to Bedside.你已经获得了较高的曲线下面积（AUC），接下来怎么办？将机器学习模型从计算机应用到临床床边时的重要考量概述。
Med Decis Making. 2025 Aug;45(6):640-653. doi: 10.1177/0272989X251343082. Epub 2025 May 29.
2
A Framework for Interpretability in Machine Learning for Medical Imaging.医学成像机器学习中的可解释性框架。
IEEE Access. 2024;12:53277-53292. doi: 10.1109/access.2024.3387702. Epub 2024 Apr 11.
3
The utility and reliability of a deep learning algorithm as a diagnosis support tool in head & neck non-melanoma skin malignancies.一种深度学习算法作为头颈部非黑色素瘤皮肤恶性肿瘤诊断支持工具的实用性和可靠性。
Eur Arch Otorhinolaryngol. 2025 Mar;282(3):1585-1592. doi: 10.1007/s00405-024-08951-z. Epub 2024 Sep 6.
4
Non-Melanoma Skin Cancer Detection in the Age of Advanced Technology: A Review.先进技术时代的非黑色素瘤皮肤癌检测：综述
Cancers (Basel). 2023 Jun 7;15(12):3094. doi: 10.3390/cancers15123094.
5
Recent Advances in Melanoma Diagnosis and Prognosis Using Machine Learning Methods.利用机器学习方法在黑色素瘤诊断和预后方面的最新进展。
Curr Oncol Rep. 2023 Jun;25(6):635-645. doi: 10.1007/s11912-023-01407-3. Epub 2023 Mar 31.
6
Artificial Intelligence for the Classification of Pigmented Skin Lesions in Populations with Skin of Color: A Systematic Review.人工智能在有色人种人群中色素性皮肤病变分类中的应用：系统评价。
Dermatology. 2023;239(4):499-513. doi: 10.1159/000530225. Epub 2023 Mar 21.
7
Open-Source Clinical Machine Learning Models: Critical Appraisal of Feasibility, Advantages, and Challenges.开源临床机器学习模型：对可行性、优势及挑战的批判性评估
JMIR Form Res. 2022 Apr 11;6(4):e33970. doi: 10.2196/33970.
8
Checklist for Evaluation of Image-Based Artificial Intelligence Reports in Dermatology: CLEAR Derm Consensus Guidelines From the International Skin Imaging Collaboration Artificial Intelligence Working Group.皮肤科基于图像人工智能报告评估清单：来自国际皮肤成像协作人工智能工作组的 CLEAR Derm 共识指南。
JAMA Dermatol. 2022 Jan 1;158(1):90-96. doi: 10.1001/jamadermatol.2021.4915.
9
Lack of Transparency and Potential Bias in Artificial Intelligence Data Sets and Algorithms: A Scoping Review.人工智能数据集和算法中缺乏透明度和潜在偏见：范围综述。
JAMA Dermatol. 2021 Nov 1;157(11):1362-1369. doi: 10.1001/jamadermatol.2021.3129.
本文引用的文献
1
Assessment of Accuracy of an Artificial Intelligence Algorithm to Detect Melanoma in Images of Skin Lesions.评估人工智能算法在皮肤损伤图像中检测黑色素瘤的准确性。
JAMA Netw Open. 2019 Oct 2;2(10):e1913436. doi: 10.1001/jamanetworkopen.2019.13436.
2
Association Between Surgical Skin Markings in Dermoscopic Images and Diagnostic Performance of a Deep Learning Convolutional Neural Network for Melanoma Recognition.皮肤镜图像中的手术皮肤标记与用于黑色素瘤识别的深度学习卷积神经网络诊断性能之间的关联
JAMA Dermatol. 2019 Oct 1;155(10):1135-1141. doi: 10.1001/jamadermatol.2019.1735.
3
A convolutional neural network trained with dermoscopic images performed on par with 145 dermatologists in a clinical melanoma image classification task.在一项临床黑色素瘤图像分类任务中，经过皮肤镜图像训练的卷积神经网络在性能上可与 145 名皮肤科医生相媲美。
Eur J Cancer. 2019 Apr;111:148-154. doi: 10.1016/j.ejca.2019.02.005. Epub 2019 Mar 8.
4
Automated Classification of Skin Lesions: From Pixels to Practice.皮肤损伤自动分类：从像素到实践。
J Invest Dermatol. 2018 Oct;138(10):2108-2110. doi: 10.1016/j.jid.2018.06.175.
5
The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions.HAM10000 数据集，一个大型的常见色素性皮肤病变多源皮肤镜图像集合。
Sci Data. 2018 Aug 14;5:180161. doi: 10.1038/sdata.2018.161.
6
Classification of the Clinical Images for Benign and Malignant Cutaneous Tumors Using a Deep Learning Algorithm.基于深度学习算法的良性和恶性皮肤肿瘤临床图像分类。
J Invest Dermatol. 2018 Jul;138(7):1529-1538. doi: 10.1016/j.jid.2018.01.028. Epub 2018 Feb 8.
7
Dermatologist-level classification of skin cancer with deep neural networks.基于深度神经网络的皮肤癌皮肤科医生级分类。
Nature. 2017 Feb 2;542(7639):115-118. doi: 10.1038/nature21056. Epub 2017 Jan 25.
8
Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs.深度学习算法在视网膜眼底照片糖尿病视网膜病变检测中的开发与验证。
JAMA. 2016 Dec 13;316(22):2402-2410. doi: 10.1001/jama.2016.17216.
Zotero 插件