4.7 Review

Explainable artificial intelligence in skin cancer recognition: A systematic review

Related references

Note: Only part of the references are listed.
Article Computer Science, Information Systems

An explainable stacked ensemble of deep learning models for improved melanoma skin cancer detection

Mohammad Shorfuzzaman

Summary: This study proposes an explainable CNN-based stacked ensemble framework for early detection of melanoma skin cancer, which is evaluated using a dataset of melanoma images and shows high accuracy and sensitivity.

MULTIMEDIA SYSTEMS (2022)

Article Computer Science, Artificial Intelligence

Explainable skin lesion diagnosis using taxonomies

Catarina Barata et al.

Summary: This study aims to address the lack of explainability in deep learning methods in medical diagnostic systems, improving the safety of skin cancer diagnostic systems. By combining medical knowledge and recent advances, a diagnostic system with competitive results was successfully developed.

PATTERN RECOGNITION (2021)

Article Computer Science, Artificial Intelligence

Predictive and discriminative localization of pathology using high resolution class activation maps with CNNs

Sumeet Shinde et al.

Summary: Current class activation mapping techniques may not be able to capture subtle abnormalities in medical images effectively. To address this issue, a new technique called high resolution class activation mapping (HR-CAMs) is proposed, which fuses feature maps from multiple layers of a CNN to provide detailed localization of abnormalities in original image resolution.

PEERJ COMPUTER SCIENCE (2021)

Review Oncology

Original Research Skin cancer classification via convolutional neural networks: systematic review of studies involving human experts

Sarah Haggenmueller et al.

Summary: The study systematically analyzed the research status of reader studies involving melanoma and found that CNN-based classifiers demonstrated superior or at least equivalent performance compared with clinicians. However, it was noted that most studies were conducted in highly artificial settings, with test sets not representing the full range of patient populations and melanoma subtypes encountered in clinical practice.

EUROPEAN JOURNAL OF CANCER (2021)

Article Biochemistry & Molecular Biology

CellBox: Interpretable Machine Learning for Perturbation Biology with Application to the Design of Cancer Combination Therapy

Bo Yuan et al.

Summary: Systematic perturbation of cells followed by comprehensive measurements of molecular and phenotypic responses can provide informative data resources for constructing computational models of cell biology. Machine learning faces challenges in finding global optima in complex multidimensional spaces and mechanistically interpreting solutions on large biological datasets. The hybrid approach introduced in this study combines mathematical models of cell dynamics with a machine-learning framework to efficiently describe cellular behavior and recapitulate known interactions.

CELL SYSTEMS (2021)

Article Oncology

Robustness of convolutional neural networks in recognition of pigmented skin lesions

Roman C. Maron et al.

Summary: A high robustness is a basic requirement for AI-based image analysis systems in clinical practice. This study quantified the impact of minor image perturbations on CNN-mediated skin lesion classification and evaluated three possible solutions. Results showed that minor image changes may affect CNNs' robustness, but existing methods can reduce brittleness to some extent.

EUROPEAN JOURNAL OF CANCER (2021)

Article Computer Science, Artificial Intelligence

Interpretable deep learning systems for multi-class segmentation and classification of non-melanoma skin cancer

Simon M. Thomas et al.

Summary: This study applies interpretable deep learning methods to analyze the most common skin cancers in a histological setting, demonstrating the potential for automatic machine analysis of dermatopathology work. By characterizing tissue into meaningful dermatological classes, the research aims to pave the way for future computer aided diagnosis systems with human interpretable outcomes.

MEDICAL IMAGE ANALYSIS (2021)

Article Multidisciplinary Sciences

Human-interpretable image features derived from densely mapped cancer pathology slides predict diverse molecular phenotypes

James A. Diao et al.

Summary: Computational methods have made substantial progress in improving the accuracy and throughput of pathology workflows, but lack of interpretability remains a significant barrier to clinical integration. The authors present an approach for predicting clinically-relevant molecular phenotypes from whole-slide histopathology images using human-interpretable image features.

NATURE COMMUNICATIONS (2021)

Article Computer Science, Information Systems

A Visually Interpretable Deep Learning Framework for Histopathological Image-Based Skin Cancer Diagnosis

Shancheng Jiang et al.

Summary: This study proposed a deep learning framework DRANet for diagnosing skin diseases, which can accurately differentiate 11 types of skin diseases and generate visualized diagnostic reports. Experimental results showed that DRANet outperformed other models, highlighting the significance of its role in decision-making for skin disease diagnosis.

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS (2021)

Article Oncology

Original Research A benchmark for neural network robustness in skin cancer classification

Roman C. Maron et al.

Summary: The study aimed to establish a benchmark for skin cancer classification on dermoscopic images to measure classifier robustness to out-of-distribution (OOD) data. By creating three datasets and evaluating the robustness of four different CNN architectures, it was found that all CNNs showed susceptibility to corruptions and perturbations.

EUROPEAN JOURNAL OF CANCER (2021)

Editorial Material Biochemistry & Molecular Biology

AI in medicine must be explainable

Shinjini Kundu

NATURE MEDICINE (2021)

Article Biochemistry & Molecular Biology

Interpretable deep learning uncovers cellular properties in label-free live cell images that are predictive of highly metastatic melanoma

Assaf Zaritsky et al.

Summary: The study utilized a generative neural network and supervised machine learning to classify the metastatic efficiency of melanoma cells, revealing characteristic features of metastatic cells and validating these findings with live cell observations. It demonstrates how artificial intelligence can aid in identifying cell properties predictive of complex phenotypes and integrated cell functions that may be subtle and difficult for human experts to identify in raw imagery.

CELL SYSTEMS (2021)

Article Biochemistry & Molecular Biology

A deep learning system for differential diagnosis of skin diseases

Yuan Liu et al.

NATURE MEDICINE (2020)

Article Engineering, Biomedical

Melanoma detection using adversarial training and deep transfer learning

Hasib Zunair et al.

PHYSICS IN MEDICINE AND BIOLOGY (2020)

Article Computer Science, Interdisciplinary Applications

A Mutual Bootstrapping Model for Automated Skin Lesion Segmentation and Classification

Yutong Xie et al.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2020)

Article Biochemistry & Molecular Biology

Human-computer collaboration for skin cancer recognition

Philipp Tschandl et al.

NATURE MEDICINE (2020)

Article Health Care Sciences & Services

Artificial Intelligence and Its Effect on Dermatologists' Accuracy in Dermoscopic Melanoma Image Classification: Web-Based Survey Study

Roman C. Maron et al.

JOURNAL OF MEDICAL INTERNET RESEARCH (2020)

Article Computer Science, Artificial Intelligence

Concept whitening for interpretable image recognition

Zhi Chen et al.

NATURE MACHINE INTELLIGENCE (2020)

Article Computer Science, Information Systems

Automatic Skin Cancer Detection in Dermoscopy Images Based on Ensemble Lightweight Deep Learning Network

Lisheng Wei et al.

IEEE ACCESS (2020)

Article Biochemistry & Molecular Biology

Clinical-grade computational pathology using weakly supervised deep learning on whole slide images

Gabriele Campanella et al.

NATURE MEDICINE (2019)

Article Oncology

Deep neural networks are superior to dermatologists in melanoma image classification

Titus J. Brinker et al.

EUROPEAN JOURNAL OF CANCER (2019)

Article Computer Science, Interdisciplinary Applications

Attention Residual Learning for Skin Lesion Classification

Jianpeng Zhang et al.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2019)

Article Computer Science, Artificial Intelligence

Pathologist-level interpretable whole-slide cancer diagnosis with deep learning

Zizhao Zhang et al.

NATURE MACHINE INTELLIGENCE (2019)

Article Computer Science, Artificial Intelligence

Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead

Cynthia Rudin

NATURE MACHINE INTELLIGENCE (2019)

Article Dermatology

Classification of the Clinical Images for Benign and Malignant Cutaneous Tumors Using a Deep Learning Algorithm

Seung Seog Han et al.

JOURNAL OF INVESTIGATIVE DERMATOLOGY (2018)

Article Multidisciplinary Sciences

Dermatologist-level classification of skin cancer with deep neural networks

Andre Esteva et al.

NATURE (2017)

Article Computer Science, Artificial Intelligence

MED-NODE: A computer-assisted melanoma diagnosis system using non-dermoscopic images

Ioannis Giotis et al.

EXPERT SYSTEMS WITH APPLICATIONS (2015)

Review Computer Science, Artificial Intelligence

Content-based image retrieval at the end of the early years

AWM Smeulders et al.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2000)