Hybrid, feature-based, quantum-enhanced computing system for highly precise multimodal detection and classification of breast cancer

Abstract

A hybrid quantum computing system for the multimodal detection and classification of breast cancer, comprising: an input module configured to receive multimodal medical breast images containing at least one of the following features: mammography images, ultrasound images, and / or histopathological images; a preprocessing module configured to perform noise reduction, normalization, and segmentation of relevant areas on the received multimodal medical breast images; a feature extraction module comprising a hybrid quantum-classical convolutional neural network-based deep feature extractor configured to extract deep features from the preprocessed images, and a handcrafted feature extractor configured to extract handcrafted features encompassing texture features, shape features, and statistical features from the preprocessed images;a quantum-optimized feature fusion module configured to encode the deep features and hand-crafted features into a quantum-inspired representation, apply superposition-based transformations to the encoded features, and apply entanglement-based transformations to the encoded features to generate fused features; a classification module consisting of a hybrid neural classifier configured to process the fused features to predict a diagnostic classification of benign or malignant tissue; and an output module connected to a user interface and configured to provide a diagnostic decision and confidence score based on the predicted diagnostic classification, which are then displayed via the user interface.

Description

AREA OF INVENTION The present disclosure relates to the field of machine learning and disease classification, in particular a hybrid quantum computing system for the multimodal detection and classification of breast cancer. More specifically, the invention relates to a hybrid computing system that combines deep neural networks, feature fusion techniques, and quantum-inspired techniques to enable precise detection and classification of breast cancer with improved diagnostic performance while simultaneously reducing computational complexity. BACKGROUND OF THE INVENTION The precise evaluation of medical imaging, including mammography, ultrasound, and histology, is essential for the diagnosis of breast cancer. Traditional deep learning models often suffer from class imbalance, a lack of generalizability, and the loss of local texture information. The exclusive use of CNN models has its own drawbacks, which can be addressed by combining local and global feature extraction methods. Local feature extraction techniques such as HOG, GLCM, and LBP focus on identifying characteristic patterns in specific image areas. Texture, color variations, edges, curves, contours, and critical points are just some of the many components that make up these patterns. A state-of-the-art approach proposes a hybrid architecture that integrates both local and global information by combining deep and handcrafted features, thereby improving classification performance compared to the previous method. Another state of the art proposes a hybrid quantum-classical convolutional neural network (HQCNN) model to increase the accuracy of image categorization with fewer training parameters. To extract complex visual and textual features independently, a state of the art proposes QMNN, which introduces highly efficient unimodal feature extraction blocks. To integrate these features and improve the interactions between text and images, a highly entangled multimodal feature fusion block is created. Another state of the art proposes a hybrid quantum-classical neural network model (HQCNN) that retains the nonlinear and scalability properties of neural networks by combining the superposition and entanglement properties of quantum computing with convolutional networks. Classical images are transformed into quantum states by combining PQC with classical neural networks and used in quantum convolution layers. Amplitude encoding allows the quantum convolution layer proposed in a prior art to reduce the number of qubits and optimize the model structure for classification. The preceding discussion indicates that the classification accuracy of medical images and data needs improvement. Therefore, an optimized system is required. The present invention provides a hybrid quantum computing system for the multimodal detection and classification of breast cancer. By combining deep neural networks, feature fusion techniques, and quantum-inspired methods, it enables precise breast cancer detection and classification with improved diagnostic performance while simultaneously reducing computational complexity. SUMMARY OF THE INVENTION The present disclosure relates to a hybrid quantum computing system for the multimodal detection and classification of breast cancer. To significantly increase the accuracy of breast cancer detection and classification, the present invention presents a hybrid computing system that combines deep neural networks, feature fusion techniques, and quantum-inspired technologies. Compared to conventional methods, the three interconnected innovative components of the system work together to improve diagnostic performance while simultaneously reducing computational complexity. The present disclosure relates to a hybrid quantum computing system for the multimodal detection and classification of breast cancer.The system comprises: an input module for receiving multimodal medical breast images, including mammography, ultrasound, and / or histopathological images; a preprocessing module for noise reduction, normalization, and segmentation of relevant areas of the received multimodal medical breast images; a feature extraction module with a hybrid quantum-based classical convolutional network for extracting deep features from the preprocessed images, as well as a manual feature extractor for extracting manual features (texture, shape, and statistical features) from the preprocessed images; and a quantum-optimized feature fusion module for encoding the deep and manual features into a quantum-inspired representation, applying superposition and entanglement transformations to the encoded features to generate fused features.a classification module consisting of a hybrid neural classifier configured to process the fused features to predict a diagnostic classification of benign or malignant tissue; and an output module connected to a user interface and configured to provide a diagnostic decision and confidence score based on the predicted diagnostic classification, which are then displayed via the user interface. The subject of the present disclosure is the provision of a hybrid quantum computing system for the multimodal detection and classification of breast cancer. Another objective of the present disclosure is the development of a quantum-optimized framework for breast cancer detection that improves feature representation using quantum and quantum-inspired principles. Another objective of the present disclosure is to provide a fusion-based neural network model that integrates global deep features and locally handcrafted descriptors for robust cancer classification. Another objective of the present disclosure is the development of a hybrid, feature-based, quantum-inspired model that increases classification accuracy while reducing computational complexity and data dependency. Another objective of the present disclosure is to provide a hybrid computing system that combines deep neural networks, feature fusion techniques and quantum-inspired techniques to enable accurate detection and classification of breast cancer with improved diagnostic performance while reducing computational complexity. However, another objective of the present disclosure is to provide a system for the accurate evaluation of medical imaging. To further clarify the advantages and features of the present disclosure, the invention is described in more detail with reference to specific embodiments illustrated in the accompanying drawings. It is understood that these drawings merely show typical embodiments of the invention and are therefore not to be understood as limiting its scope of protection. The invention is described and explained in more detail and with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE IMAGES These and other features, aspects, and advantages of the present disclosure will be better understood if the following detailed description is read with reference to the accompanying drawings, in which identical symbols represent identical parts, wherein: Fig. 1 shows a block diagram of a hybrid quantum computing system for the multimodal detection and classification of breast cancer according to an embodiment of the present disclosure; Fig. 2 shows a diagram illustrating the operation of the quantum feature extractor and the classical feature extractor according to an embodiment of the present disclosure; Fig. 3 shows a diagram illustrating the operation of the hybrid neural classifier according to an embodiment of the present disclosure; and Fig.Figure 4 shows a diagram illustrating the functioning of the proposed system for the detection and classification of breast cancer according to an embodiment of the present disclosure. Furthermore, those skilled in the art will recognize that the elements in the drawings are simplified and not necessarily drawn to scale. For example, the flowcharts illustrate the process by highlighting the main steps to facilitate understanding of this disclosure. With regard to the construction of the device, one or more components may be represented in the drawings by conventional symbols. The drawings may show only those specific details relevant to understanding the embodiments of this disclosure, so as not to clutter the drawings with details that are already apparent to those skilled in the art from the description contained herein. DETAILED DESCRIPTION: To facilitate understanding of the principles of the invention, reference is made below to the embodiment illustrated in the drawings, which is described using specific terms. It is understood, however, that this does not limit the scope of protection of the invention. Rather, modifications and further developments of the illustrated system, as well as further applications of the inventive principles depicted therein, are conceivable, insofar as they would typically occur to a person skilled in the art in the field of the invention. It will be clear to those skilled in the art that the foregoing general description and the following detailed description are exemplary and explanatory of the invention and are not to be understood as a limitation of it. References to “an aspect”, “another aspect”, or similar phrases in this description mean that a particular feature, structure, or property described in connection with the embodiment is included in at least one embodiment of the present disclosure. Therefore, phrases such as “in one embodiment”, “in another embodiment”, and similar expressions in this description may, but do not necessarily, all refer to the same embodiment. The terms "includes," "comprehensive," or similar expressions denote non-exclusive inclusion. Thus, a procedure or method containing a list of steps does not only include those steps but may also include further steps not explicitly listed or inherent in the procedure or method. Likewise, the statement "includes..." for one or more devices, subsystems, elements, structures, or components, without further limitations, does not preclude the existence of other devices, subsystems, elements, structures, or components. Unless otherwise defined, all technical and scientific terms used herein have the same meanings generally known to those skilled in the art in the field to which this invention belongs. The systems, methods, and examples described herein serve only for illustration and are not to be understood as limiting. Embodiments of the present disclosure are described in detail below with reference to the accompanying drawings. The functional units described in this specification are referred to as devices. A device may be implemented in programmable hardware such as processors, digital signal processors, central processing units, FPGAs, PALs, PLDs, cloud processing systems, or similar. Devices may also be implemented in software for execution by various processor types. An identified device may contain executable code and, for example, comprise one or more physical or logical blocks of computer instructions, which may be organized as an object, procedure, function, or other construct. However, the executable files of an identified device need not be physically related; they may consist of different instructions stored in different locations that, when logically combined, constitute the device and fulfill its purpose. The executable code of a device or module can consist of a single instruction or multiple instructions and can even extend across different code sections, applications, and storage media. Similarly, operational data within the device can be identified and represented, and can exist in any suitable form and be organized in any data structure. The operational data can be captured as a single data record or distributed across various storage media and may exist, at least partially, as electronic signals within a system or network. References to “a selected embodiment”, “an embodiment”, or “an embodiment” in this description mean that a particular feature, structure, or property described in connection with the embodiment is included in at least one embodiment of the disclosed subject matter. Therefore, the phrases “a selected embodiment”, “in an embodiment”, or “in an embodiment” appearing at different points in this description do not necessarily refer to the same embodiment. Furthermore, the described features, structures, or properties can be combined in one or more embodiments in any suitable manner. The following description contains numerous specific details to enable a comprehensive understanding of the embodiments of the disclosed subject matter. However, a person skilled in the art will recognize that the disclosed subject matter can also be realized without one or more of the specific details or with other methods, components, materials, etc. In other cases, known structures, materials, or processes are not presented or described in detail so as not to obscure aspects of the disclosed subject matter. According to the exemplary embodiments, the disclosed computer programs or modules can be executed in a variety of ways, for example, as an application running in the memory of a device or as a hosted application running on a server and communicating with the device application or browser via various standard protocols such as TCP / IP, HTTP, XML, SOAP, REST, JSON, and other suitable protocols. The disclosed computer programs can be written in programming languages ​​that run either in the device's memory or on a hosted server, such as BASIC, COBOL, C, C++, Java, Pascal, or scripting languages ​​such as JavaScript, Python, Ruby, PHP, Perl, or other suitable programming languages. Some of the described embodiments involve data transmission over a network, such as the transmission of various inputs or files. The network may include, for example, the internet, wide area networks (WANs), local area networks (LANs), analog or digital wired and wireless telephone networks (e.g., PSTN, ISDN, cellular networks, and xDSL), radio, television, cable, satellite, and / or other data transmission or tunneling mechanisms. It may include multiple networks or subnetworks, each of which may, for example, have a wired or wireless data path. The network may include a circuit-switched voice network, a packet-switched data network, or another network for transmitting electronic data. For example, it may be based on the Internet Protocol (IP) or the asynchronous transmission mode (ATM) and support voice communication using VoIP, Voice over ATM, or similar protocols.In one embodiment, the network comprises a mobile network configured for the exchange of text or SMS messages. Examples of networks include Personal Area Networks (PAN), Storage Area Networks (SAN), Home Area Networks (HAN), Campus Area Networks (CAN), Local Area Networks (LAN), Wide Area Networks (WAN), Metropolitan Area Networks (MAN), Virtual Private Networks (VPN), Enterprise Private Networks (EPN), the Internet, Global Area Networks (GAN), and so on. Fig. 1 shows a block diagram of a hybrid quantum computing system for multimodal detection and classification of breast cancer according to an embodiment of the present disclosure. With reference to Fig. 1, the hybrid quantum computing system (100) comprises: an input module (102) for receiving multimodal medical breast images, including at least one of the following: mammography, ultrasound, and histopathology images; a preprocessing module (104) for noise reduction, normalization, and segmentation of regions of interest in the received multimodal medical breast images; a feature extraction module (106) with a hybrid quantum feature extractor (106a) based on a classical convolutional neural network (CNN) for extracting deep features from the preprocessed images and a feature extractor (106b) for extracting hand-drawn features consisting of texture, shape, and statistical features from the preprocessed images;A quantum-optimized feature fusion module (108) configured to encode the deep features and the manually created features into a quantum-inspired representation, apply superposition-based transformations to the encoded features, and apply entanglement-based transformations to the encoded features to generate fused features; a classification module (110) comprising a hybrid neural classifier (110a) configured to process the fused features to predict a diagnostic classification of benign or malignant tissue; and an output module (112) connected to a user interface (114) configured to provide a diagnostic decision and confidence score based on the predicted diagnostic classification, which is then displayed via the user interface. In one embodiment, the preprocessing module (104) is further configured to improve the contrast in the multimodal medical breast images prior to segmentation of the relevant area and to remove artifacts. In one embodiment, the feature extraction module (106) is configured to process the multimodal medical breast images in parallel to extract the deep features and the manually created features simultaneously. In one embodiment, the hybrid quantum classical convolutional neural network-based deep feature extractor (106a) is configured to extract global image features from the preprocessed images. In one embodiment, the handcrafted feature extractor (106b) is configured to extract local image features that include at least one of the following features: edge information, intensity distributions, and morphological patterns. In one embodiment, the quantum-enhanced feature fusion module (108) is configured to process the deep features and the hand-crafted features through parallel coding paths before applying the superposition-based transformations and the entanglement-based transformations. In one embodiment, the quantum-enhanced feature fusion module (108) comprises parameterized quantum gates configured to transform classical feature vectors into the quantum-enhanced representation. In one embodiment, the hybrid neural classifier (110a) is configured to distinguish between benign and malignant tissue based on the fused features, which integrate both local and global image features. In one embodiment, the system (100) is configured to operate on classical computer hardware, near-future quantum computer hardware, or a combination of both. In one embodiment, the output module (112) is further configured to generate a diagnostic report that includes the diagnostic decision, the confidence value and a feature contribution analysis. In one embodiment, the input module (102), the preprocessing module (104), the feature extraction module (106), the quantum-optimized feature fusion module (108), the classification module (110), and the output module (112) with user interface (114) can be implemented in programmable hardware devices such as processors, digital signal processors, central processing units, FPGAs, PALs, PLDs, cloud processing systems, or the like. Alternatively, the devices can also be implemented in software for execution by various processor types. Figure 2 illustrates the operation of the quantum and classical feature extractors according to one embodiment of the present disclosure. The architectural design of the feature extraction module (200) represents a sophisticated hybrid approach to medical image analysis, splitting the data flow into two separate but complementary pipelines: the quantum feature extractor (202) and the classical feature extractor (204). This dual-track system was designed to address the inherent complexities of multimodal breast imaging, such as mammography and histopathology, by capturing both high-level abstract representations and detailed, manually generated morphological features. By utilizing a parallel processing framework, the module ensures that the subsequent fusion stage receives a multidimensional dataset, significantly improving the sensitivity and specificity of diagnostic outcomes. The quantum feature extractor (202) utilizes a parameterized quantum circuit (202a) and a quantum convolution filter (202b) to overcome the limitations of traditional linear data processing. In this submodule, classically preprocessed data is mapped into a high-dimensional Hilbert space. This allows the system to leverage quantum-inspired mathematical principles such as superposition and entanglement. This approach is particularly effective in identifying nonlinear correlations and global image features that are often difficult to detect in the early stages of cancer. The quantum convolution filters act as an advanced level of abstraction, refining the input data into quantum-inspired representations that are more robust against the noise typically found in raw medical scans. In parallel to the quantum pathway, the classical feature extractor (204) focuses on established deterministic methods to ensure that the system is based on observable clinical markers. It integrates the histogram-of-gradients (HoG) method (204a) applied to the image mask of the region of interest (ROI). This is essential for capturing the orientation and distribution of gradients within the tumor margin. This manual approach is crucial for extracting texture and shape features—such as spiculations or density variations—that radiologists rely on for manual diagnosis. By integrating these features with a parallel feature extraction method (204b), the module remains compatible with current classical hardware while providing a robust foundation of statistical and structural data. The synergy between these two extractors forms the basis for the improved classification accuracy of the invention. After feature extraction, the features are fed to the quantum-enhanced feature fusion module, where the classical, manually created features and the quantum-based features are integrated using entanglement transformations. This comprehensive feature set enables the hybrid neural classifier to make more nuanced predictions between benign and malignant findings. By reducing data dependency through this hybrid extraction, the system achieves better generalization across different patient datasets and ultimately provides the user with a clinical risk assessment and a diagnostic confusion matrix via a professional user interface. Fig. 3 shows a diagram illustrating the operation of the hybrid neural classifier according to an embodiment of the present disclosure. The depicted architectural setup describes the hybrid neural classifier (110a), which serves as the primary classification module of the invention for predicting benign / malignant tissues. This module receives integrated data from the quantum-optimized feature fusion module, in which encoded classical features, transformed by superposition and entanglement, are processed to improve tissue differentiation. The classifier utilizes a combination of quantum-inspired methods and classical decision models to ensure high diagnostic accuracy. The core of the classification logic consists of the quantum kernel classifier (110b) and the quantum kernel circuit (110c). These components utilize quantum-inspired representations for the actual classification, enabling the system to process complex, nonlinear relationships in multimodal breast images such as mammography and ultrasound. By employing a kernel-based approach in the circuit, the system can effectively project fused feature data into a higher-dimensional space to more precisely separate clinical categories. This high-dimensional analysis is supported by the quantum Hilbert feature space (110d), which provides the mathematical environment necessary for quantum-inspired coding. Within this space, the system uses transformations based on quantum principles to improve the fusion of CNN-based deep features and manually created features. This robust processing of local and global image features allows the classifier to maintain high accuracy even with reduced data dependency, thus enabling better generalization across different diagnostic scenarios. The final stage of this hierarchy is the output module (110e), which delivers the diagnostic decision and a corresponding confidence score. This module translates the internal classification results into a user-friendly format, such as a clinical risk assessment and a diagnostic output with a confusion matrix. This information is then provided to the user via a dedicated interface and supports applications ranging from computer-aided diagnosis to telemedicine and remote diagnostics. Figure 4 illustrates the operation of the proposed breast cancer detection and classification system according to an embodiment of the present disclosure. Input data is fed into the system, with the input images undergoing preprocessing. This includes image scaling and cropping, noise filtering, histogram normalization, and segmentation of relevant areas. The preprocessed images are then subjected to feature extraction. The feature extraction module, consisting of a quantum and a classical feature extractor, is configured for parallel feature extraction and combines CNN-based with manual feature extraction. The extracted features are fed to the quantum-optimized feature fusion module, where they are encoded and fused. Subsequently, they are used for hybrid neural classification and diagnostic output.Classification is performed by a quantum-inspired classifier, a decision module conducts a clinical risk assessment, and the diagnostic output with confusion matrix is ​​provided to the user via a user interface. In one embodiment, the system architecture comprises: an input module for providing multimodal breast images (mammography, ultrasound, histopathology); a preprocessing module for noise reduction, normalization, and ROI segmentation; a feature extraction module with a) a CNN-based deep feature extractor and b) a manually created feature extractor for texture, shape, and statistical features; a quantum-optimized feature fusion module for encoding classical features into a quantum-inspired representation using superposition and entanglement transformations; a classification module with a hybrid neural classifier for predicting benign / malignant findings; and an output module for diagnostic decision-making and confidence score calculation. The present invention offers several advantages over existing systems, including: improved classification accuracy through optimized feature fusion; robust processing of local and global image features; reduced data dependency and improved generalization; and compatibility with current quantum and classical hardware. The proposed invention presents a hybrid, classical-quantum-inspired system that combines convolutional neural networks (CNNs), manual feature extraction, and parameterized, quantum-inspired feature fusion. The system processes medical breast images using parallel feature extraction pipelines and integrates them using quantum-inspired coding, superposition, and entanglement concepts to improve the differentiation between benign and malignant tissue. The proposed system can be used for: computer-aided diagnostic systems; clinical decision support; telemedicine and remote diagnostics; and research in the field of quantum-inspired medical AI. The drawings and the preceding description illustrate embodiments. Those skilled in the art will recognize that one or more of the described elements can be combined to form a single functional element. Alternatively, certain elements can be divided into several functional elements. Elements of one embodiment can be added to another. For example, the process flows described here can be modified and are not limited to the manner described herein. Furthermore, the actions of a flowchart need not be performed in the sequence shown; nor do all actions necessarily need to be carried out. Actions that do not depend on other actions can be performed in parallel with the other actions. The scope of protection of the embodiments is in no way limited by these specific examples. Numerous variations, whether explicitly stated in the description or not, such as...Differences in structure, dimensions, and materials are possible. The scope of protection of the embodiments is at least as comprehensive as described by the following claims. The advantages, other benefits, and problem solutions have been described above with reference to specific embodiments. However, the advantages, benefits, problem solutions, and any components that can effect or enhance an advantage, benefit, or solution are not to be construed as critical, necessary, or essential features or components of the claims. REFERENCE 100 A quantum hybrid computing system for multimodal detection and classification of breast cancer. 102 Input Module 104 Preprocessing Module 106 Feature Extraction Module 106a Hybrid Quantum Classical Convolution Neural Network-Based Deep Feature Extractor 106b Handcrafted Feature Extractor 108 Quantum Optimized Feature Fusion Module 110 Classification Module 110a Hybrid Neural Classifier 110b Quantum Kernel Classical Classifier 110c Quantum Kernel Circuit 110d Quantum Hilbert Feature Space 110e Output Module 112 Output Module 114 User Interface 202 Quantum Feature Extractor 202a Parameterized Quantum Circuit 202b Quantum Convolution Filter 204 Classical Feature Extractor 204a Gradient Method ROI Image Mask Histogram 204b Parallel Feature Extraction Method 402 Patient Data Input 402a Mammography 402b Histopathology 402c Ultrasound 404 Data preprocessing and normalization 406 Quantum feature extractor 408 Classical feature extractor 410Quantum-inspired classifier 412 Class inequality technique 414 Fusion layer 416 Decision module 418 Clinical risk measurement 420 Diagnostic output with confusion matrix 422 Benign 424 Malignant

Claims

A hybrid quantum computing system for the multimodal detection and classification of breast cancer, comprising: an input module configured to receive multimodal medical breast images containing at least one of the following features: mammography images, ultrasound images, and / or histopathological images; a preprocessing module configured to perform noise reduction, normalization, and segmentation of relevant areas on the received multimodal medical breast images; a feature extraction module comprising a hybrid quantum-classical convolutional neural network-based deep feature extractor configured to extract deep features from the preprocessed images, and a handcrafted feature extractor configured to extract handcrafted features encompassing texture features, shape features, and statistical features from the preprocessed images;a quantum-optimized feature fusion module configured to encode the deep features and hand-crafted features into a quantum-inspired representation, apply superposition-based transformations to the encoded features, and apply entanglement-based transformations to the encoded features to generate fused features; a classification module consisting of a hybrid neural classifier configured to process the fused features to predict a diagnostic classification of benign or malignant tissue; and an output module connected to a user interface and configured to provide a diagnostic decision and confidence score based on the predicted diagnostic classification, which are then displayed via the user interface. System according to claim 1, wherein the preprocessing module is further configured to improve the contrast of the multimodal medical breast images prior to segmentation of the relevant area and to remove artifacts from them. System according to claim 1, wherein the feature extraction module is configured to process the multimodal medical breast images in parallel to extract the deep features and the manually created features simultaneously. System according to claim 1, wherein the hybrid quantum-classical convolutional neural network-based deep feature extractor is configured to extract global image features from the preprocessed images. System according to claim 1, wherein the handcrafted feature extractor is configured to extract local image features comprising at least one of the following features: edge information, intensity distributions, and morphological patterns. System according to claim 1, wherein the quantum-enhanced feature fusion module is configured to process the deep features and the hand-crafted features through parallel coding paths before applying the superposition-based transformations and the entanglement-based transformations. System according to claim 1, wherein the quantum-enhanced feature fusion module comprises parameterized quantum gates configured to transform classical feature vectors into the quantum-enhanced representation. System according to claim 1, wherein the hybrid neuronal classifier is configured to distinguish between benign and malignant tissue based on the fused features integrating both local and global image features. System according to claim 1, wherein the system is configured to operate on classical computer hardware, near-future quantum computer hardware or a combination of both. System according to claim 1, wherein the output module is further configured to generate a diagnostic report that includes the diagnostic decision, the confidence value and the feature contribution analysis.

Images