Secure and Robust Machine Learning for Healthcare: A Survey (2020-01-21T00:00:00.000000Z)

TL;DR

An overview of various application areas in healthcare that leverage ML techniques from security and privacy point of view and present associated challenges and potential methods to ensure secure and privacy-preserving ML for healthcare applications is presented.

Abstract

Recent years have witnessed widespread adoption of machine learning (ML)/deep learning (DL) techniques due to their superior performance for a variety of healthcare applications ranging from the prediction of cardiac arrest from one-dimensional heart signals to computer-aided diagnosis (CADx) using multi-dimensional medical images. Notwithstanding the impressive performance of ML/DL, there are still lingering doubts regarding the robustness of ML/DL in healthcare settings (which is traditionally considered quite challenging due to the myriad security and privacy issues involved), especially in light of recent results that have shown that ML/DL are vulnerable to adversarial attacks. In this paper, we present an overview of various application areas in healthcare that leverage such techniques from security and privacy point of view and present associated challenges. In addition, we present potential methods to ensure secure and privacy-preserving ML for healthcare applications. Finally, we provide insight into the current research challenges and promising directions for future research.

Authors

A. Qayyum

4 papers

Ala I. Al-Fuqaha

5 papers

Junaid Qadir

5 papers

TL;DR

Abstract

Authors

References217 items

Homomorphic Encryption for Machine Learning in Medicine and Bioinformatics

Evaluating saliency map explanations for convolutional neural networks: a user study

Data Augmentation for Histopathological Images Based on Gaussian-Laplacian Pyramid Blending

Security and Privacy for the Internet of Medical Things Enabled Healthcare Systems: A Survey

A Gaussian Data Augmentation Technique on Highly Dimensional, Limited Labeled Data for Multiclass Classification Using Deep Learning

Differential Privacy for Data and Model Publishing of Medical Data

Differential Privacy-enabled Federated Learning for Sensitive Health Data

An Efficient Differential Privacy Logistic Classification Mechanism

Automated deep learning design for medical image classification by health-care professionals with no coding experience: a feasibility study.

Turning the crank for machine learning: ease, at what expense?

Do no harm: a roadmap for responsible machine learning for health care

The “inconvenient truth” about AI in healthcare

Feature Robustness in Non-stationary Health Records: Caveats to Deployable Model Performance in Common Clinical Machine Learning Tasks

Practical guidance on artificial intelligence for health-care data.

Self-Supervised Domain Adaptation for Computer Vision Tasks

Reinforcement learning in clinical medicine: a method to optimize dynamic treatment regime over time.

Emotion recognition using secure edge and cloud computing

Reproducibility in Machine Learning for Health

Review of Causal Discovery Methods Based on Graphical Models

The Adversarial Machine Learning Conundrum: Can the Insecurity of ML Become the Achilles' Heel of Cognitive Networks?

Learning-Based Privacy-Aware Offloading for Healthcare IoT With Energy Harvesting

Securing Connected & Autonomous Vehicles: Challenges Posed by Adversarial Machine Learning and the Way Forward

Deep Learning Techniques for Medical Image Segmentation: Achievements and Challenges

Sensor-Based Human Activity Recognition for Smart Healthcare: A Semi-supervised Machine Learning

Reinforcement Learning for Sepsis Treatment: Baselines and Analysis

A Euclidean Group Assessment on Semi-Supervised Clustering for Healthcare Clinical Implications Based on Real-Life Data

Unlock Your Heart: Next Generation Biometric in Resource-Constrained Healthcare Systems and IoT

A Machine Learning Framework for Biometric Authentication Using Electrocardiogram

Fairness in Algorithmic Decision Making: An Excursion Through the Lens of Causality

Caveat Emptor: The Risks of Using Big Data for Human Development

Adversarial attacks on medical machine learning

Medical Diagnostics Based on Encrypted Medical Data

Exploiting the Vulnerability of Deep Learning-Based Artificial Intelligence Models in Medical Imaging: Adversarial Attacks

Retrospective Motion Correction in Multishot MRI using Generative Adversarial Network

TensorSCONE: A Secure TensorFlow Framework using Intel SGX

Is AmI (Attacks Meet Interpretability) Robust to Adversarial Examples?

The need for uncertainty quantification in machine-assisted medical decision making

Hybrid Reasoning-based Privacy-Aware Disease Prediction Support System

A General Approach to Adding Differential Privacy to Iterative Training Procedures

A Survey of Unsupervised Deep Domain Adaptation

Privacy-Preserving Distributed Deep Learning for Clinical Data

Split learning for health: Distributed deep learning without sharing raw patient data

FADL: Federated-Autonomous Deep Learning for Distributed Electronic Health Record

What is Interpretable? Using Machine Learning to Design Interpretable Decision-Support Systems

Deep learning for healthcare: review, opportunities and challenges

Automating Motion Correction in Multishot MRI Using Generative Adversarial Networks

A Spectral View of Adversarially Robust Features

Unsupervised domain adaptation for medical imaging segmentation with self-ensembling

Mortality prediction in patients with isolated moderate and severe traumatic brain injury using machine learning models

Real-Time Medical Systems Based on Human Biometric Steganography: a Systematic Review

Attacks Meet Interpretability: Attribute-steered Detection of Adversarial Samples

Multimodal Recurrent Model with Attention for Automated Radiology Report Generation

Toward an Understanding of Adversarial Examples in Clinical Trials

Implementation and Evaluation of an Algorithm for Cryptographically Private Principal Component Analysis on Genomic Data

Subsampled Rényi Differential Privacy and Analytical Moments Accountant

A Review of Challenges and Opportunities in Machine Learning for Health.

Predicting and Understanding Unexpected Respiratory Decompensation in Critical Care Using Sparse and Heterogeneous Clinical Data

Opportunities in Machine Learning for Healthcare

Why Is My Classifier Discriminatory?

Defending Against Adversarial Attacks by Leveraging an Entire GAN

Towards the first adversarially robust neural network model on MNIST

Artificial intelligence will soon change the landscape of medical physics research and practice.

Privacy-Preserving Deep Learning via Additively Homomorphic Encryption

Context-Aware Symptom Checking for Disease Diagnosis Using Hierarchical Reinforcement Learning

Fairness in Decision-Making - The Causal Explanation Formula

Secure Logistic Regression Based on Homomorphic Encryption: Design and Evaluation

Adversarial Attacks Against Medical Deep Learning Systems

Semi-supervised learning with generative adversarial networks for chest X-ray classification with ability of data domain adaptation

Manipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learning

Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural Networks

Federated learning of predictive models from federated Electronic Health Records

Clinical Natural Language Processing in languages other than English: opportunities and challenges

Privacy-Preserving Machine Learning: Threats and Solutions

Brain Tumor Type Classification via Capsule Networks

Defense-GAN: Protecting Classifiers Against Adversarial Attacks Using Generative Models

Scalable Private Learning with PATE