Quantum-classical convolutional neural networks in radiological image classification

  title={Quantum-classical convolutional neural networks in radiological image classification},
  author={Andrea Matic and Maureen Monnet and Jeanette Lorenz and Balthasar Schachtner and Thomas Messerer},
—Quantum machine learning is receiving significant attention currently, but its usefulness in comparison to classical machine learning techniques for practical applications remains unclear. However, there are indications that certain quantum machine learning algorithms might result in improved training capabilities with respect to their classical counterparts - which might be particularly beneficial in situations with little train- ing data available. Such situations naturally arise in medical… 

Figures and Tables from this paper

Incremental Data-Uploading for Full-Quantum Classification
This work proposes ‘incremental data-uploading’, a novel encoding pattern for high dimensional data that tackles challenges of encoding into a quantum circuit for a NISQ device without any loss of information.


Quantum convolutional neural networks
A quantum circuit-based algorithm inspired by convolutional neural networks is shown to successfully perform quantum phase recognition and devise quantum error correcting codes when applied to arbitrary input quantum states.
Quantum convolutional neural network for classical data classification
This work benchmarks fully parameterized quantum convolutional neural networks (QCNNs) for classical data classification and proposes a quantum neural network model inspired by CNN that only uses two-qubit interactions throughout the entire algorithm.
Quantum Algorithms for Deep Convolutional Neural Networks
A new quantum tomography algorithm with $\ell_{\infty}$ norm guarantees, and new applications of probabilistic sampling in the context of information processing and numerical simulations for the classification of the MNIST dataset are presented.
Hybrid quantum-classical convolutional neural networks
QCCNN is friendly to currently noisy intermediate-scale quantum computers, in terms of both number of qubit as well as circuit's depths, while retaining important features of classical CNN, such as nonlinearity and scalability.
Comparing concepts of quantum and classical neural network models for image classification task
Results of experiments on training and performance of a hybrid quantum-classical neural network developed for the problem of classification of handwritten digits from the MNIST data set indicate that the quantum network, although its simulation is time-consuming, overcomes the classical network.
Quanvolutional neural networks: powering image recognition with quantum circuits
A new type of transformational layer called a quantum convolution, or quanvolutional layer is introduced, which operates on input data by locally transforming the data using a number of random quantum circuits, in a way that is similar to the transformations performed by random convolutional filter layers.
Quantum Convolutional Neural Network For Image Classification
The results on the breast cancer image dataset show that the proposed QCNN with the proposed feature extraction outperformed the classic CNN in terms of recognition rate 90%–93% over 83%–85% on performance.
A quantum convolutional neural network on NISQ devices
This machine learning model can provide implementable quantum circuits that accurately corresponds to a specific classical convolutional kernel and opens up the prospect of exploiting quantum power to process information in the era of big data.
Variational Quanvolutional Neural Networks with enhanced image encoding
There is not one best image encoding, but that the choice of the encoding depends on the specific constraints of the application, and the experiments show that some image encodings are better suited for variational circuits.
Supervised learning with quantum-enhanced feature spaces
Two classification algorithms that use the quantum state space to produce feature maps are demonstrated on a superconducting processor, enabling the solution of problems when the feature space is large and the kernel functions are computationally expensive to estimate.