Risk Susceptibility of Brain Tumor Classification to Adversarial Attacks

Kotia, Jai; Kotwal, Adit; Bharti, Rishika

doi:10.1007/978-3-030-31964-9_17

Cited by 18 publications

(9 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the literature, there are also studies that used the same database for classification with pre-trained networks [23,[35][36][37][38][39][40] or, as input, they use only tumor region or some features that are extracted from the tumor region [7,21,23,41,42]. Similarly, in several papers, researchers have modified this database prior to classification [36,[43][44][45][46][47]. The designed networks are usually simpler than already-existing pre-trained networks and have faster execution speed.…”

Section: Comparison With State-of-the-art-methodsmentioning

confidence: 99%

Classification of Brain Tumors from MRI Images Using a Convolutional Neural Network

2020

View full text Add to dashboard Cite

The classification of brain tumors is performed by biopsy, which is not usually conducted before definitive brain surgery. The improvement of technology and machine learning can help radiologists in tumor diagnostics without invasive measures. A machine-learning algorithm that has achieved substantial results in image segmentation and classification is the convolutional neural network (CNN). We present a new CNN architecture for brain tumor classification of three tumor types. The developed network is simpler than already-existing pre-trained networks, and it was tested on T1-weighted contrast-enhanced magnetic resonance images. The performance of the network was evaluated using four approaches: combinations of two 10-fold cross-validation methods and two databases. The generalization capability of the network was tested with one of the 10-fold methods, subject-wise cross-validation, and the improvement was tested by using an augmented image database. The best result for the 10-fold cross-validation method was obtained for the record-wise cross-validation for the augmented data set, and, in that case, the accuracy was 96.56%. With good generalization capability and good execution speed, the new developed CNN architecture could be used as an effective decision-support tool for radiologists in medical diagnostics.

show abstract

Section: Comparison With State-of-the-art-methodsmentioning

confidence: 99%

Classification of Brain Tumors from MRI Images Using a Convolutional Neural Network

2020

View full text Add to dashboard Cite

show abstract

“…An unreliable MedAI output can be fatal in clinical settings, and erroneous results can misalign the overall cycle of future research and healthcare solutions towards the harmful direction. Adversarial attacks on neural networks can cause errors in identifying cancer tumors and damage the confidence in MedAI output (Kotia et al, 2019). Szegedy et al showed that very subtle adversarial inputs, which may not appear as pathological, can potentially change the output (Leung et al, 2015).…”

Section: Security and Integrity Of Medaimentioning

confidence: 99%

Predatory Medicine: Exploring and Measuring the Vulnerability of Medical AI to Predatory Science

Saini¹,

Saxena²

2022

Preprint

View full text Add to dashboard Cite

Medical Artificial Intelligence (MedAI) for diagnosis, treatment options, and drug development represents the new age of healthcare. The security, integrity, and credibility of MedAI tools are paramount issues because human lives are at stake. MedAI solutions are often heavily dependent on scientific medical research literature as a primary data source that draws the attacker's attention as a potential target. We present a first study of how the output of MedAI can be polluted with Predatory Publications Presence (PPP).We study two MedAI systems: mediKanren (Disease independent) and CancerMine (Disease-specific), which use research literature as primary data input from the research repository PubMed, PubMed derived database SemMedDB, and NIH translational Knowledge Graphs (KGs). Our study has a three-pronged focus: (1) identifying the PPP in PubMed; (2) verifying the PPP in SemMedDB and the KGs;(3) demonstrating the existing vulnerability of PPP traversing to the MedAI output.Our contribution lies in identifying the existing PPP in the MedAI inputs and demonstrating how predatory science can jeopardize the credibility of MedAI solutions, making their real-life deployment questionable.Data and Code Availability This paper uses the data and code from PubMed , NIH SemMedDB , NIH Translational KGs (provided by mediKanren team), CancerMine Zenodo repository, and mediKanren Github

show abstract

“…The vulnerability of brain tumor classification on adversarial attacks was studied by Kotia et al [82]. They applied three different white-box attacks, noise-based attack, FGSM and virtual adversarial training (VAT) [83].…”

Section: Existing Adversarial Attacks On Medical Imagesmentioning

confidence: 99%

A Survey on Adversarial Deep Learning Robustness in Medical Image Analysis

Apostolidis

Papakostas

2021

Electronics

View full text Add to dashboard Cite

In the past years, deep neural networks (DNN) have become popular in many disciplines such as computer vision (CV), natural language processing (NLP), etc. The evolution of hardware has helped researchers to develop many powerful Deep Learning (DL) models to face numerous challenging problems. One of the most important challenges in the CV area is Medical Image Analysis in which DL models process medical images—such as magnetic resonance imaging (MRI), X-ray, computed tomography (CT), etc.—using convolutional neural networks (CNN) for diagnosis or detection of several diseases. The proper function of these models can significantly upgrade the health systems. However, recent studies have shown that CNN models are vulnerable under adversarial attacks with imperceptible perturbations. In this paper, we summarize existing methods for adversarial attacks, detections and defenses on medical imaging. Finally, we show that many attacks, which are undetectable by the human eye, can degrade the performance of the models, significantly. Nevertheless, some effective defense and attack detection methods keep the models safe to an extent. We end with a discussion on the current state-of-the-art and future challenges.

show abstract

Risk Susceptibility of Brain Tumor Classification to Adversarial Attacks

Cited by 18 publications

References 7 publications

Classification of Brain Tumors from MRI Images Using a Convolutional Neural Network

Classification of Brain Tumors from MRI Images Using a Convolutional Neural Network

Predatory Medicine: Exploring and Measuring the Vulnerability of Medical AI to Predatory Science

A Survey on Adversarial Deep Learning Robustness in Medical Image Analysis

Contact Info

Product

Resources

About