Qutrit-Inspired Fully Self-Supervised Shallow Quantum Learning Network for Brain Tumor Segmentation

Abstract

Classical self-supervised networks suffer from convergence problems and reduced segmentation accuracy due to forceful termination. Qubits or bi-level quantum bits often describe quantum neural network models. In this article, a novel self-supervised shallow learning network model exploiting the sophisticated three-level qutrit-inspired quantum information system referred to as Quantum Fully Self-Supervised Neural Network (QFS-Net) is presented for automated segmentation of brain MR images. The QFS-Net model comprises a trinity of a layered structure of qutrits inter-connected through parametric Hadamard gates using an 8-connected second-order neighborhood-based topology. The non-linear transformation of the qutrit states allows the underlying quantum neural network model to encode the quantum states, thereby enabling a faster self-organized counter-propagation of these states between the layers without supervision. The suggested QFS-Net model is tailored and extensively validated on Cancer Imaging Archive (TCIA) data set collected from Nature repository and also compared with state of the art supervised (U-Net and URes-Net architectures) and the self-supervised QIS-Net model. Results shed promising segmented outcome in detecting tumors in terms of dice similarity and accuracy with minimum human intervention and computational resources.

Random Quantum Neural Networks (RQNN) for Noisy Image Recognition

“…where, |X (ω) is a quantum state in the RQNN model's dressed quantum layer corresponding to the classical information |x from the classical layers of RNNs. The |X (ω) quantum state or qubit characteristics are sent to a Hadamard gate, H, for equal superposition of the qubit states 50 .…”

Quantum Neurobiology

“…Qutrits (three-level quantum states) may be conducive to brain tumor analysis since many quantum states are not binary. One proposal suggests that a qutrit model might better correspond to grayscale imaging data, using a quantum neural network model to segment brain lesions [64]. Whereas qubits are a relatively simple system, expanding to higher dimension qudits (quantum information digits) is non-trivial as it is difficult to quantify the quantum correlations in the system (using the diagonalization of correlation matrices for bipartite systems) [65].…”

3D Quantum-inspired Self-supervised Tensor Network for Volumetric Segmentation of Medical Images

“…The quantum versions of the classical self-supervised neural network architectures [10], [11], [37], [38] offer a potential candidate for faster and efficient image segmentation and surpasses the classical counterparts. Konar et al recently developed quantum-inspired neural network models referred to as QIS-Net [12], QFS-Net [13] and QIBDS-Net [39] suitable for brain MR image segmentation. These networks have been found to attain promising outcome in complete brain tumor segmentation and serves the motivation behind assimilation of quantum-inspired computing in the current 3D-QNet architecture.…”

“…where, ρij and κj refer to the learning rates for the adjustments of weights and activation, respectively and are evaluated as The sequences of {ω l,d } and {ϑ l,d } converge super-linearly subject to the following conditions [13]. The convergence of the sequence {ω l,d } according to L-Lipschitz continuity is illustrated as [48] ζ(ω…”

“…The unified concept of quantum-inspired neural networks (QINN) enhances the approximation and generalization capabilities of classical neural networks and has emerged to process information faster in the field computer science [8], [9]. Of late, quantum-inspired neural networks are becoming popular in solving problems in the domain of pattern recognition and classification [10]- [13] employing the inherent characteristics of quantum computation. However, the complex and timeintensive quantum back-propagation algorithm involved in the aforementioned QINN models suffer from slow convergence problems.…”