Artificial neural networks and pathologists recognize basal cell carcinomas based on different histological patterns

Recent advances in artificial intelligence, particularly in the field of deep learning, have enabled researchers to create compelling algorithms for medical image analysis. Histological slides of basal cell carcinomas (BCCs), the most frequent skin tumor, are accessed by pathologists on a daily basis and are therefore well suited for automated prescreening by neural networks for the identification of cancerous regions and swift tumor classification. In this proof-of-concept study, we implemented an accurate and intuitively interpretable artificial neural network (ANN) for the detection of BCCs in histological whole-slide images (WSIs). Furthermore, we identified and compared differences in the diagnostic histological features and recognition patterns relevant for machine learning algorithms vs. expert pathologists. An attention-ANN was trained with WSIs of BCCs to identify tumor regions (n = 820). The diagnosis-relevant regions used by the ANN were compared to regions of interest for pathologists, detected by eye-tracking techniques. This ANN accurately identified BCC tumor regions on images of histologic slides (area under the ROC curve: 0.993, 95% CI: 0.990–0.995; sensitivity: 0.965, 95% CI: 0.951–0.979; specificity: 0.910, 95% CI: 0.859–0.960). The ANN implicitly calculated a weight matrix, indicating the regions of a histological image that are important for the prediction of the network. Interestingly, compared to pathologists’ eye-tracking results, machine learning algorithms rely on significantly different recognition patterns for tumor identification (p < 10−4). To conclude, we found on the example of BCC WSIs, that histopathological images can be efficiently and interpretably analyzed by state-of-the-art machine learning techniques. Neural networks and machine learning algorithms can potentially enhance diagnostic precision in digital pathology and uncover hitherto unused classification patterns.

[1]  F. Mohs.  CHEMOSURGERY: A MICROSCOPICALLY CONTROLLED METHOD OF CANCER EXCISION , 1941 .

[2]  Tobias Pincock.  Fitzpatrick's Dermatology in General Medicine , 2003 .

[3]  U. Bertheim,et al.  The stromal reaction in basal cell carcinomas. A prerequisite for tumour progression and treatment strategy. , 2004, British journal of plastic surgery.

[4]  W. C. Quevedo,et al.  General Biology of Mammalian Pigmentation , 2007 .

[5]  Fabio A. González,et al.  An unsupervised feature learning framework for basal cell carcinoma image analysis , 2015, Artif. Intell. Medicine.

[6]  P. Saldanha,et al.  Cutaneous basal cell carcinoma: A morphological spectrum , 2015 .

[7]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[8]  Dhruv Batra,et al.  Human Attention in Visual Question Answering: Do Humans and Deep Networks look at the same regions? , 2016, EMNLP.

[9]  Linda G. Shapiro,et al.  Multi-instance multi-label learning for whole slide breast histopathology , 2016, SPIE Medical Imaging.

[10]  Dhruv Batra,et al.  Human Attention in Visual Question Answering: Do Humans and Deep Networks look at the same regions? , 2016, EMNLP.

[11]  Sepp Hochreiter,et al.  Self-Normalizing Neural Networks , 2017, NIPS.

[12]  Jon Griffin,et al.  Digital pathology in clinical use: where are we now and what is holding us back? , 2017, Histopathology.

[13]  D. W. D. Cruz,et al.  Enhanced larval supply and recruitment can replenish reef corals on degraded reefs , 2017, Scientific Reports.

[14]  Anant Madabhushi,et al.  Accurate and reproducible invasive breast cancer detection in whole-slide images: A Deep Learning approach for quantifying tumor extent , 2017, Scientific Reports.

[15]  Sebastian Thrun,et al.  Dermatologist-level classification of skin cancer with deep neural networks , 2017, Nature.

[16]  Max Welling,et al.  Attention-based Deep Multiple Instance Learning , 2018, ICML.

[17]  Clive R. Taylor,et al.  Whole Slide Imaging Versus Microscopy for Primary Diagnosis in Surgical Pathology , 2017, The American journal of surgical pathology.

[18]  S. Han,et al.  Keratinocytic Skin Cancer Detection on the Face Using Region-Based Convolutional Neural Network. , 2019, JAMA dermatology.

[19]  Rajath E. Soans,et al.  Augmenting the Pathology Lab: An Intelligent Whole Slide Image Classification System for the Real World , 2019, ArXiv.

[20]  Thomas J. Fuchs,et al.  Whole slide imaging equivalency and efficiency study: experience at a large academic center , 2019, Modern Pathology.

[21]  Detecting cutaneous basal cell carcinomas in ultra-high resolution and weakly labelled histopathological images , 2019, ArXiv.

[22]  R. Hofmann-Wellenhof,et al.  Comparison of the accuracy of human readers versus machine-learning algorithms for pigmented skin lesion classification: an open, web-based, international, diagnostic study. , 2019, The Lancet. Oncology.

[23]  A. Parwani.  Next generation diagnostic pathology: use of digital pathology and artificial intelligence tools to augment a pathological diagnosis , 2019, Diagnostic Pathology.

[24]  A. Enk,et al.  Deep learning outperformed 11 pathologists in the classification of histopathological melanoma images. , 2019, European journal of cancer.

[25]  H. Hemker,et al.  Heparins: A Shift of Paradigm , 2019, Front. Med..

[26]  Luiz Eduardo Soares de Oliveira,et al.  Multiple instance learning for histopathological breast cancer image classification , 2019, Expert Syst. Appl..

[27]  Neofytos Dimitriou,et al.  Deep Learning for Whole Slide Image Analysis: An Overview , 2019, Front. Med..

[28]  Thomas J. Fuchs,et al.  Clinical-grade computational pathology using weakly supervised deep learning on whole slide images , 2019, Nature Medicine.

[29]  Hanqiu Sun,et al.  Human vs Machine Attention in Neural Networks: A Comparative Study , 2019, ArXiv.

[30]  J. Elder,et al.  Histologic findings associated with laser interstitial thermotherapy for glioblastoma multiforme , 2019, Diagnostic Pathology.

[31]  M. Gurcan,et al.  Digital pathology and artificial intelligence. , 2019, The Lancet. Oncology.

[32]  Julianna D. Ianni,et al.  Tailored for Real-World: A Whole Slide Image Classification System Validated on Uncurated Multi-Site Data Emulating the Prospective Pathology Workload , 2019, Scientific Reports.

[33]  Matteo Porro,et al.  Early detection of general movements trajectories in very low birth weight infants , 2020, Scientific Reports.

[34]  Alexandre Cadrin-Chênevert,et al.  Deep learning workflow in radiology: a primer , 2020, Insights into Imaging.

[35]  Hao Chen,et al.  Automatic lesion detection with three-dimensional convolutional neural networks , 2020 .

[36]  John Paoli,et al.  Human–computer collaboration for skin cancer recognition , 2020, Nature Medicine.