MedRDF : A Robust and Retrain-Less Diagnostic Framework for Medical Pretrained Models Against Adversarial Attack
Deep neural networks are discovered to be non-robust when attacked by imperceptible adversarial examples, which is dangerous for it applied into medical diagnostic system that requires high reliability. However, the defense methods that have good effect in natural images may not be suitable for medical diagnostic tasks. The pre-processing methods (e.g., random resizing, compression) may lead to the loss of the small lesions feature in the medical image. Retraining the network on the augmented data set is also not practical for medical models that have already been deployed online. Accordingly, it is necessary to design an easy-to-deploy and effective defense framework for medical diagnostic tasks. In this paper, we propose a Robust and Retrain-Less Diagnostic Framework for Medical pretrained models against adversarial attack (i.e., MedRDF). It acts on the inference time of the pretrained medical model. Specifically, for each test image, MedRDF firstly creates a large number of noisy copies of it, and obtains the output labels of these copies from the pretrained medical diagnostic model. Then, based on the labels of these copies, MedRDF outputs the final robust diagnostic result by majority voting. In addition to the diagnostic result, MedRDF produces the Robust Metric (RM) as the confidence of the result. Therefore, it is convenient and reliable to utilize MedRDF to convert pretrained non-robust diagnostic models into robust ones. The experimental results on COVID-19 and DermaMNIST datasets verify the effectiveness of our MedRDF in improving the robustness of medical diagnostic models.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2022 |
---|---|
Erschienen: |
2022 |
Enthalten in: |
Zur Gesamtaufnahme - volume:41 |
---|---|
Enthalten in: |
IEEE transactions on medical imaging - 41(2022), 8 vom: 29. Aug., Seite 2130-2143 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Xu, Mengting [VerfasserIn] |
---|
Links: |
---|
Themen: |
---|
Anmerkungen: |
Date Completed 03.08.2022 Date Revised 12.10.2022 published: Print-Electronic Citation Status MEDLINE |
---|
doi: |
10.1109/TMI.2022.3156268 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM337667578 |
---|
LEADER | 01000naa a22002652 4500 | ||
---|---|---|---|
001 | NLM337667578 | ||
003 | DE-627 | ||
005 | 20231225234934.0 | ||
007 | cr uuu---uuuuu | ||
008 | 231225s2022 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1109/TMI.2022.3156268 |2 doi | |
028 | 5 | 2 | |a pubmed24n1125.xml |
035 | |a (DE-627)NLM337667578 | ||
035 | |a (NLM)35235504 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Xu, Mengting |e verfasserin |4 aut | |
245 | 1 | 0 | |a MedRDF |b A Robust and Retrain-Less Diagnostic Framework for Medical Pretrained Models Against Adversarial Attack |
264 | 1 | |c 2022 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Completed 03.08.2022 | ||
500 | |a Date Revised 12.10.2022 | ||
500 | |a published: Print-Electronic | ||
500 | |a Citation Status MEDLINE | ||
520 | |a Deep neural networks are discovered to be non-robust when attacked by imperceptible adversarial examples, which is dangerous for it applied into medical diagnostic system that requires high reliability. However, the defense methods that have good effect in natural images may not be suitable for medical diagnostic tasks. The pre-processing methods (e.g., random resizing, compression) may lead to the loss of the small lesions feature in the medical image. Retraining the network on the augmented data set is also not practical for medical models that have already been deployed online. Accordingly, it is necessary to design an easy-to-deploy and effective defense framework for medical diagnostic tasks. In this paper, we propose a Robust and Retrain-Less Diagnostic Framework for Medical pretrained models against adversarial attack (i.e., MedRDF). It acts on the inference time of the pretrained medical model. Specifically, for each test image, MedRDF firstly creates a large number of noisy copies of it, and obtains the output labels of these copies from the pretrained medical diagnostic model. Then, based on the labels of these copies, MedRDF outputs the final robust diagnostic result by majority voting. In addition to the diagnostic result, MedRDF produces the Robust Metric (RM) as the confidence of the result. Therefore, it is convenient and reliable to utilize MedRDF to convert pretrained non-robust diagnostic models into robust ones. The experimental results on COVID-19 and DermaMNIST datasets verify the effectiveness of our MedRDF in improving the robustness of medical diagnostic models | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a Research Support, Non-U.S. Gov't | |
700 | 1 | |a Zhang, Tao |e verfasserin |4 aut | |
700 | 1 | |a Zhang, Daoqiang |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t IEEE transactions on medical imaging |d 1982 |g 41(2022), 8 vom: 29. Aug., Seite 2130-2143 |w (DE-627)NLM082855269 |x 1558-254X |7 nnns |
773 | 1 | 8 | |g volume:41 |g year:2022 |g number:8 |g day:29 |g month:08 |g pages:2130-2143 |
856 | 4 | 0 | |u http://dx.doi.org/10.1109/TMI.2022.3156268 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 41 |j 2022 |e 8 |b 29 |c 08 |h 2130-2143 |