2019
DOI: 10.3390/s19132969
|View full text |Cite
|
Sign up to set email alerts
|

Local Interpretable Model-Agnostic Explanations for Classification of Lymph Node Metastases

Abstract: An application of explainable artificial intelligence on medical data is presented. There is an increasing demand in machine learning literature for such explainable models in health-related applications. This work aims to generate explanations on how a Convolutional Neural Network (CNN) detects tumor tissue in patches extracted from histology whole slide images. This is achieved using the “locally-interpretable model-agnostic explanations” methodology. Two publicly-available convolutional neural networks trai… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
36
0
6

Year Published

2020
2020
2024
2024

Publication Types

Select...
4
1
1

Relationship

0
6

Authors

Journals

citations
Cited by 98 publications
(42 citation statements)
references
References 12 publications
(25 reference statements)
0
36
0
6
Order By: Relevance
“…Palatnik de Sousa et al [83] explored an AI algorithm that classifies lymph node metastases by creating heatmaps of the area in the input patch that contributed most to the prediction. The authors found that deep learning algorithms trained for this task have underlying 'reasoning' behind their predictions similar to human logic.…”
Section: Xai Methods In Medical Imagingmentioning
confidence: 99%
“…Palatnik de Sousa et al [83] explored an AI algorithm that classifies lymph node metastases by creating heatmaps of the area in the input patch that contributed most to the prediction. The authors found that deep learning algorithms trained for this task have underlying 'reasoning' behind their predictions similar to human logic.…”
Section: Xai Methods In Medical Imagingmentioning
confidence: 99%
“…A segmentação FHA já foi aplicada com sucesso, na literatura, para uma tarefa similar de classificação de imagens histopatológicas [Palatnik de Sousa et al 2019]. Esse algoritmo consiste numa oversegmentation da imagem feita através de clustering das regiões com base no método de minimum spanning trees [Felzenszwalb and Huttenlocher 2004].…”
Section: Algoritmo De Segmentaçãounclassified
“…Uma varredura mais detalhada de parâmetros e a análise de outros métodos de segmentação certamente poderiam ser proveitosos. No entanto, isso fica fora do escopo deste trabalho, dado o tempo e custo computacional requerido para todos os testes, e dado que a heurística encontrada, com este algoritmo já usado para aplicação similar [Palatnik de Sousa et al 2019], parece ser suficiente para segmentar as regiões de maior interesse das imagens pertencentesà base de dados aqui analisada.…”
Section: Algoritmo De Segmentaçãounclassified
See 2 more Smart Citations