Robustness Analysis of Deep Learning-Based Lung Cancer Classification Using Explainable Methods
Deep Learning (DL) based classification algorithms have been shown to achieve top results in clinical diagnosis, namely with lung cancer datasets. However, the complexity and opaqueness of the models together with the still scant training datasets call for the development of explainable modeling met...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9919875/ |
| _version_ | 1811256563393888256 |
|---|---|
| author | Mafalda Malafaia Francisco Silva Ines Neves Tania Pereira Helder P. Oliveira |
| author_facet | Mafalda Malafaia Francisco Silva Ines Neves Tania Pereira Helder P. Oliveira |
| author_sort | Mafalda Malafaia |
| collection | DOAJ |
| description | Deep Learning (DL) based classification algorithms have been shown to achieve top results in clinical diagnosis, namely with lung cancer datasets. However, the complexity and opaqueness of the models together with the still scant training datasets call for the development of explainable modeling methods enabling the interpretation of the results. To this end, in this paper we propose a novel interpretability approach and demonstrate how it can be used on a malignancy lung cancer DL classifier to assess its stability and congruence even when fed a low amount of image samples. Additionally, by disclosing the regions of the medical images most relevant to the resulting classification the approach provides important insights to the correspondent clinical meaning apprehended by the algorithm. Explanations of the results provided by ten different models against the same test sample are compared. These attest the stability of the approach and the algorithm focus on the same image regions. |
| first_indexed | 2024-04-12T17:43:16Z |
| format | Article |
| id | doaj.art-33dd000bf4364c9a9fc5c79c9cbfe45f |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-12T17:43:16Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-33dd000bf4364c9a9fc5c79c9cbfe45f2022-12-22T03:22:45ZengIEEEIEEE Access2169-35362022-01-011011273111274110.1109/ACCESS.2022.32148249919875Robustness Analysis of Deep Learning-Based Lung Cancer Classification Using Explainable MethodsMafalda Malafaia0https://orcid.org/0000-0002-8081-0454Francisco Silva1https://orcid.org/0000-0003-3069-2282Ines Neves2Tania Pereira3Helder P. Oliveira4https://orcid.org/0000-0002-6193-8540INESC TEC–Institute for Systems and Computer Engineering, Technology and Science, Porto, PortugalINESC TEC–Institute for Systems and Computer Engineering, Technology and Science, Porto, PortugalINESC TEC–Institute for Systems and Computer Engineering, Technology and Science, Porto, PortugalINESC TEC–Institute for Systems and Computer Engineering, Technology and Science, Porto, PortugalINESC TEC–Institute for Systems and Computer Engineering, Technology and Science, Porto, PortugalDeep Learning (DL) based classification algorithms have been shown to achieve top results in clinical diagnosis, namely with lung cancer datasets. However, the complexity and opaqueness of the models together with the still scant training datasets call for the development of explainable modeling methods enabling the interpretation of the results. To this end, in this paper we propose a novel interpretability approach and demonstrate how it can be used on a malignancy lung cancer DL classifier to assess its stability and congruence even when fed a low amount of image samples. Additionally, by disclosing the regions of the medical images most relevant to the resulting classification the approach provides important insights to the correspondent clinical meaning apprehended by the algorithm. Explanations of the results provided by ten different models against the same test sample are compared. These attest the stability of the approach and the algorithm focus on the same image regions.https://ieeexplore.ieee.org/document/9919875/CT scancongruencedeep learningdiagnostic imaginginterpretabilitymalignancy |
| spellingShingle | Mafalda Malafaia Francisco Silva Ines Neves Tania Pereira Helder P. Oliveira Robustness Analysis of Deep Learning-Based Lung Cancer Classification Using Explainable Methods IEEE Access CT scan congruence deep learning diagnostic imaging interpretability malignancy |
| title | Robustness Analysis of Deep Learning-Based Lung Cancer Classification Using Explainable Methods |
| title_full | Robustness Analysis of Deep Learning-Based Lung Cancer Classification Using Explainable Methods |
| title_fullStr | Robustness Analysis of Deep Learning-Based Lung Cancer Classification Using Explainable Methods |
| title_full_unstemmed | Robustness Analysis of Deep Learning-Based Lung Cancer Classification Using Explainable Methods |
| title_short | Robustness Analysis of Deep Learning-Based Lung Cancer Classification Using Explainable Methods |
| title_sort | robustness analysis of deep learning based lung cancer classification using explainable methods |
| topic | CT scan congruence deep learning diagnostic imaging interpretability malignancy |
| url | https://ieeexplore.ieee.org/document/9919875/ |
| work_keys_str_mv | AT mafaldamalafaia robustnessanalysisofdeeplearningbasedlungcancerclassificationusingexplainablemethods AT franciscosilva robustnessanalysisofdeeplearningbasedlungcancerclassificationusingexplainablemethods AT inesneves robustnessanalysisofdeeplearningbasedlungcancerclassificationusingexplainablemethods AT taniapereira robustnessanalysisofdeeplearningbasedlungcancerclassificationusingexplainablemethods AT helderpoliveira robustnessanalysisofdeeplearningbasedlungcancerclassificationusingexplainablemethods |