Exemplars and Counterexemplars Explanations for Image Classifiers, Targeting Skin Lesion Labeling

Metta, Carlo; Guidotti, Riccardo; Yin, Yanru; Gallinari, Patrick; Rinzivillo, Salvatore

doi:10.1109/iscc53001.2021.9631485

Cited by 11 publications

(15 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section we briefly present the two main components of the methodology adopted to classify and explain the ISIC dataset. Details can be found in [15,14,3].…”

Section: Methodsmentioning

confidence: 99%

“…We summarize here the customization of abele we carried on in order to make it usable for the complex image classification task of the ISIC dataset. Details can be found in [3]. Generative Adversarial models are generally not easy to train as they are usually affected by a number of common failures.…”

Section: Progressive Growing Aaementioning

confidence: 99%

“…However, while in a GAN, the discriminator is linked to the generator output, in an AAE, the discriminator takes as input the encoded latent space instead of the full reconstructed image. Thus, in [3] we define Progressive Growing Adversarial Autoencoder (PGAAE) as follows.…”

Section: Progressive Growing Aaementioning

confidence: 99%

“…AI systems that are able to assist doctors in taking informative decisions, complementing their own knowledge with the information and suggestion yielded by the AI system [3,4]. However, if the logic for the decisions of AI systems are not available it would be impossible to accomplish this goal.…”

Section: Introductionmentioning

confidence: 99%

“…The aim of this paper is to extend and exploit the methodology illustrated in [14] and [3], and to study the usability of an explanation method into a real medical setting. In particular, we focus on skin lesion diagnosis from images.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Explainable Deep Image Classifiers for Skin Lesion Diagnosis

Metta¹,

Beretta²,

Guidotti³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

A key issue in critical contexts such as medical diagnosis is the interpretability of the deep learning models adopted in decision-making systems. Research in eXplainable Artificial Intelligence (XAI) is trying to solve this issue. However, often XAI approaches are only tested on generalist classifier and do not represent realistic problems such as those of medical diagnosis. In this paper, we analyze a case study on skin lesion images where we customize an existing XAI approach for explaining a deep learning model able to recognize different types of skin lesions. The explanation is formed by synthetic exemplar and counter-exemplar images of skin lesion and offers the practitioner a way to highlight the crucial traits responsible for the classification decision. A survey conducted with domain experts, beginners and unskilled people proof that the usage of explanations increases the trust and confidence in the automatic decision system. Also, an analysis of the latent space adopted by the explainer unveils that some of the most frequent skin lesion classes are distinctly separated. This phenomenon could derive from the intrinsic characteristics of each class and, hopefully, can provide support in the resolution of the most frequent misclassifications by human experts.

show abstract

“…In this section we briefly present the two main components of the methodology adopted to classify and explain the ISIC dataset. Details can be found in [15,14,3].…”

Section: Methodsmentioning

confidence: 99%

Section: Progressive Growing Aaementioning

confidence: 99%

Section: Progressive Growing Aaementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Explainable Deep Image Classifiers for Skin Lesion Diagnosis

Metta¹,

Beretta²,

Guidotti³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Evaluating perceptual and semantic interpretability of saliency methods: A case study of melanoma

Bokadia

Yang

et al. 2022

Applied AI Letters

View full text Add to dashboard Cite

In order to be useful, XAI explanations have to be faithful to the AI system they seek to elucidate and also interpretable to the people that engage with them. There exist multiple algorithmic methods for assessing faithfulness, but this is not so for interpretability, which is typically only assessed through expensive user studies. Here we propose two complementary metrics to algorithmically evaluate the interpretability of saliency map explanations. One metric assesses perceptual interpretability by quantifying the visual coherence of the saliency map. The second metric assesses semantic interpretability by capturing the degree of overlap between the saliency map and textbook features-features human experts use to make a classification. We use a melanoma dataset and a deep-neural network classifier as a case-study to explore how our two interpretability metrics relate to each other and a faithfulness metric. Across six commonly used saliency methods, we find that none achieves high scores across all three metrics for all test images, but that different methods perform well in different regions of the data distribution. This variation between methods can be leveraged to consistently achieve high interpretability and faithfulness by using our metrics to inform saliency mask selection on a case-by-case basis. Our interpretability metrics provide a new way to evaluate saliency-based explanations and allow for the adaptive combination of saliency-based explanation methods.

show abstract

Improving trust and confidence in medical skin lesion diagnosis through explainable deep learning

Metta

Beretta

Guidotti

et al. 2023

Int J Data Sci Anal

View full text Add to dashboard Cite

A key issue in critical contexts such as medical diagnosis is the interpretability of the deep learning models adopted in decision-making systems. Research in eXplainable Artificial Intelligence (XAI) is trying to solve this issue. However, often XAI approaches are only tested on generalist classifier and do not represent realistic problems such as those of medical diagnosis. In this paper, we aim at improving the trust and confidence of users towards automatic AI decision systems in the field of medical skin lesion diagnosis by customizing an existing XAI approach for explaining an AI model able to recognize different types of skin lesions. The explanation is generated through the use of synthetic exemplar and counter-exemplar images of skin lesions and our contribution offers the practitioner a way to highlight the crucial traits responsible for the classification decision. A validation survey with domain experts, beginners, and unskilled people shows that the use of explanations improves trust and confidence in the automatic decision system. Also, an analysis of the latent space adopted by the explainer unveils that some of the most frequent skin lesion classes are distinctly separated. This phenomenon may stem from the intrinsic characteristics of each class and may help resolve common misclassifications made by human experts.

show abstract

Exemplars and Counterexemplars Explanations for Image Classifiers, Targeting Skin Lesion Labeling

Cited by 11 publications

References 19 publications

Explainable Deep Image Classifiers for Skin Lesion Diagnosis

Explainable Deep Image Classifiers for Skin Lesion Diagnosis

Evaluating perceptual and semantic interpretability of saliency methods: A case study of melanoma

Improving trust and confidence in medical skin lesion diagnosis through explainable deep learning

Contact Info

Product

Resources

About