Intelligent diagnosis system and method for early liver cancer based on multi-phase enhanced CT

Abstract

The present invention relates to the technical field of medical information systems, and more specifically, to an intelligent early liver cancer diagnosis system and method based on multi-phase enhanced CT, comprising: a CT image module, for acquiring multi-phase enhanced CT images of patients; an image super-resolution reconstruction module, connected to the CT image module; a radiomics feature module, connected to the image super-resolution reconstruction module, and a deep feature module, connected to the image super-resolution reconstruction module, for constructing a deep convolutional neural network model; a knowledge distillation module, connected to the deep feature module; an early diagnosis module, connected to the radiomics feature module, the deep feature module, and the knowledge distillation module, for integrating high-throughput radiomics features and deep features; and generating early liver cancer diagnosis results based on an interpretable diagnostic strategy, thereby retaining the prior knowledge of human experts and making full use of the powerful feature learning capabilities of deep learning.

Description

Technical Field

[0001] The present invention relates to the technical field of medical information systems, and more particularly to an intelligent diagnosis system and method for early liver cancer based on multi-phase enhanced CT. Background Art

[0002] With the increasing number of patients with chronic kidney disease, blood purification therapy is increasingly being used in clinical practice. Liver cancer is one of the most common malignancies worldwide, and early diagnosis is crucial for improving patient survival. In recent years, with the continuous advancement of computed tomography (CT) technology, multi-phase contrast-enhanced CT has become an important tool for liver cancer diagnosis. However, traditional diagnostic methods rely primarily on the experience of radiologists, which can be highly subjective and inefficient.

[0003] To improve the accuracy and efficiency of diagnosis, researchers have begun to try to apply artificial intelligence technology to liver cancer diagnosis. Currently, the closest existing technologies focus on the following aspects:

[0004] First, some researchers have proposed radiomics-based methods. These methods extract quantitative features from CT images, such as shape and texture, to aid diagnosis. While radiomics methods have improved diagnostic objectivity to a certain extent, they still face challenges such as difficulty in feature selection and high image quality requirements.

[0005] Secondly, with the development of deep learning technology, some researchers have attempted to use models such as convolutional neural networks to learn features directly from CT images. Although these methods have achieved good results on some datasets, they face challenges such as poor interpretability and limited generalization ability.

[0006] In addition, some researchers have attempted to combine radiomics with deep learning methods. However, these methods often involve simple feature concatenation or model fusion, failing to fully leverage the advantages of both methods.

[0007] Although existing technologies have made certain progress, the following problems still exist in the diagnosis of early liver cancer:

[0008] 1. The temporal information of multi-phase CT images is not fully utilized, and the dynamic enhancement characteristics of tumors cannot be fully captured.

[0009] 2. Image quality limits diagnostic accuracy, especially for the detection of early small lesions.

[0010] 3. The model lacks interpretability, making it difficult to gain the trust of clinicians and be widely used.

[0011] 4. Lack of effective integration of expert experience and failure to fully utilize existing medical knowledge.

[0012] 5. The diagnostic results lack foresight and cannot provide sufficient support for clinical decision-making.

[0013] In view of the above problems, there is an urgent need for an intelligent diagnostic system that can fully utilize multi-phase enhanced CT information, improve image quality, enhance model interpretability, effectively integrate expert experience, and have predictive capabilities. Summary of the Invention

[0014] The present invention aims to address the aforementioned technical issues by providing an intelligent early-stage liver cancer diagnosis system and method based on multi-phase contrast-enhanced CT. This system, through the collaborative operation of multiple modules, achieves intelligent management of the entire process, from image acquisition, preprocessing, feature extraction, to final diagnosis.

[0015] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0016] The intelligent diagnosis system for early liver cancer based on multi-phase enhanced CT includes:

[0017] CT imaging module for:

[0018] Acquire multi-phase enhanced CT images of patients;

[0019] Preprocessing the multi-phase enhanced CT image;

[0020] An image super-resolution reconstruction module, connected to the CT image module, is used to:

[0021] receiving the preprocessed multi-phase enhanced CT image sent by the CT image module;

[0022] Performing super-resolution reconstruction based on the preprocessed multi-phase enhanced CT image to obtain a reconstructed high-resolution CT image;

[0023] The radiomics feature module is connected to the image super-resolution reconstruction module and is used to:

[0024] extracting high-throughput radiomics features from the reconstructed high-resolution CT image;

[0025] A deep feature module, connected to the image super-resolution reconstruction module, is used to:

[0026] Build a deep convolutional neural network model;

[0027] Extracting depth features from the reconstructed high-resolution CT image using the deep convolutional neural network model;

[0028] The knowledge distillation module is connected to the deep feature module and is used to:

[0029] Form a training set based on expert annotations;

[0030] Training the deep feature module to form an interpretable diagnostic strategy;

[0031] An early diagnosis module, connected to the radiomics feature module, the deep feature module, and the knowledge distillation module, is used to:

[0032] integrating the high-throughput radiomics features and the deep features;

[0033] Based on the explainable diagnostic strategy, early liver cancer diagnosis results are generated.

[0034] Preferably, the CT image module includes:

[0035] An image acquisition unit, used for acquiring multi-phase enhanced CT images of the patient;

[0036] An image preprocessing unit, configured to perform normalization and standardization on the multi-phase enhanced CT image;

[0037] The image alignment unit is used to register the multi-phase enhanced CT images to eliminate the problem of uneven distribution of CT values.

[0038] Preferably, the image super-resolution reconstruction module includes:

[0039] The encoding-decoding module is built based on the U-Net architecture;

[0040] Multi-level residual structure, which combines global residual and local residual;

[0041] Atrous convolution unit, used to reduce the number of parameters;

[0042] The encoding-decoding module includes alternating convolutional layers and residual convolutional layers.

[0043] Preferably, the radiomics feature module is used to:

[0044] Extract the dynamic enhancement curve characteristics of the lesion;

[0045] Extract sequence features;

[0046] Extract texture features;

[0047] Extract shape features;

[0048] Among them, the radiomics feature module uses a multi-channel convolutional neural network layer to extract high-throughput image features, and uses a deep convolutional neural network layer to fuse high-throughput image features and high-throughput image features.

[0049] Preferably, the depth feature module includes:

[0050] Temporal attention model, used to model dynamic temporal attention;

[0051] Convolutional neural network model for feature extraction and fusion;

[0052] Among them, the deep feature module adopts void convolution and attention mechanism to build a deep convolutional network model, and uses image pyramid and residual structure to optimize the network model.

[0053] Preferably, the temporal attention model includes:

[0054] Encoding module, including convolutional layer, fully connected layer and pooling layer;

[0055] Decoding module, including dense attention layer;

[0056] The encoding module is used to extract CT image features and lesion temporal patterns, and the decoding module is used to extract correlations between features.

[0057] Preferably, the knowledge distillation module is used to:

[0058] Use expert experience as new samples;

[0059] Converting the expert experience into a feature vector through a feature extraction network;

[0060] Matching the feature vector with the prediction result of the deep feature module;

[0061] Optimizing the prediction results of the deep feature module.

[0062] Preferably, the early diagnosis module includes:

[0063] a multi-feature fusion unit, configured to integrate the high-throughput radiomics features and the deep features;

[0064] A prediction model building unit, configured to build a prediction model for identifying early-stage liver cancer based on the interpretable diagnostic strategy;

[0065] The diagnosis result generating unit is used to generate an early liver cancer diagnosis result using the early liver cancer recognition and prediction model.

[0066] As an advantage, it also includes:

[0067] The lesion growth prediction module is connected to the early diagnosis module and is used to:

[0068] Predicting the growth of the lesion based on the early liver cancer diagnosis results and the patient's historical medical data;

[0069] Generate early diagnosis recommendations.

[0070] The intelligent diagnosis method for early liver cancer based on multi-phase enhanced CT includes the following steps:

[0071] S1: Acquire a multi-phase enhanced CT image of a patient and preprocess the multi-phase enhanced CT image;

[0072] S2: performing super-resolution reconstruction on the preprocessed multi-phase enhanced CT image to obtain a reconstructed high-resolution CT image;

[0073] S3: extracting high-throughput radiomics features from the reconstructed high-resolution CT image;

[0074] S4: extracting deep features from the reconstructed high-resolution CT image using a deep convolutional neural network model;

[0075] S5: Forming a training set based on expert annotations, training the deep convolutional neural network model to form an interpretable diagnostic strategy;

[0076] S6: Integrating the high-throughput radiomics features and the deep features to generate early liver cancer diagnosis results based on the interpretable diagnostic strategy;

[0077] S7: Based on the early liver cancer diagnosis results and the patient's historical medical data, predict the growth status of the lesion and generate early diagnosis recommendations.

[0078] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0079] First, the present invention constructs a complete diagnostic framework, with each module working closely together to form a closed-loop diagnostic process. The CT image module provides high-quality raw data for subsequent analysis; the image super-resolution reconstruction module significantly improves image quality, creating conditions for the detection of early small lesions; the radiomics feature module and the deep feature module extract image features from different angles, respectively, retaining the prior knowledge of human experts and fully utilizing the powerful feature learning capabilities of deep learning; the knowledge distillation module cleverly integrates expert experience into the model, improving the interpretability and credibility of the system; and the early diagnosis module integrates multi-source information to provide the final diagnosis result.

[0080] Secondly, the present invention effectively resolves several contradictions in existing technologies. For example, by introducing super-resolution reconstruction technology, image quality is improved without increasing radiation dose; by combining radiomics and deep learning features, feature interpretability is ensured while improving model performance; and through knowledge distillation technology, a balance is achieved between model complexity and interpretability.

[0081] Furthermore, the effects of the various modules of the present invention are highly complementary and synergistic. For example, super-resolution reconstruction not only improves image quality but also provides richer information for subsequent feature extraction. The combination of radiomics features and deep learning features enables the model to capture both human-interpretable features and underlying complex patterns. The introduction of knowledge distillation not only improves the interpretability of the model but also, to a certain extent, compensates for the problem of insufficient training data.

[0082] Finally, the overall design of this invention takes into account the practical needs of clinical application. By providing interpretable diagnostic results and lesion growth predictions, it provides comprehensive support for clinicians' decision-making. This design not only improves diagnostic accuracy but also enhances the system's usability and acceptability in real-world clinical settings.

[0083] In summary, this invention achieves high precision, high efficiency and high interpretability in the diagnosis of early liver cancer through the organic combination of multiple innovative technologies, providing a powerful tool for improving the detection rate of early liver cancer and improving patient prognosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 It is a logic block diagram of the overall system structure of the present invention.

[0085] Figure 2 This is a logic block diagram of the CT image module of the present invention.

[0086] Figure 3 This is a logic block diagram of the image super-resolution reconstruction module of the present invention.

[0087] Figure 4 This is a logical block diagram of the deep feature module of the present invention.

[0088] Figure 5 This is a logic block diagram of the early diagnosis module of the present invention. DETAILED DESCRIPTION

[0089] like Figure 1-5 As shown, the present invention provides an intelligent system and method for diagnosing early-stage liver cancer based on multi-phase contrast-enhanced CT. This system achieves high-precision diagnosis of early-stage liver cancer through multimodal data fusion and deep learning technology. The specific embodiments of the present invention are described in detail below.

[0090] First, the intelligent early liver cancer diagnosis system based on multi-phase contrast-enhanced CT of the present invention includes the following modules: CT image module 1, image super-resolution reconstruction module 2, radiomics feature module 3, deep feature module 4, knowledge distillation module 5, and early diagnosis module 6. These modules work together to complete the intelligent diagnosis task of early liver cancer.

[0091] The CT imaging module 1 is used to acquire multi-phase contrast-enhanced CT images of the patient and preprocess these images. In practical applications, the CT imaging module 1 is typically connected directly to a hospital's CT equipment. Preferably, the present invention utilizes a 3.0T superconducting magnetic resonance imaging scanner (e.g., the Toshiba Aquilion ONE 320-row) for image acquisition. The acquired multi-phase contrast-enhanced CT images typically include images from the arterial phase, portal venous phase, and delayed phase.

[0092] The preprocessing steps in CT image module 1 include image normalization and standardization. Normalization scales pixel values ​​to the range [0, 1], while standardization sets the image to a mean of 0 and a standard deviation of 1. These preprocessing steps help improve the stability and accuracy of subsequent processing.

[0093] The image super-resolution reconstruction module 2 is connected to the CT image module 1 and is used to receive the pre-processed multi-phase enhanced CT image and perform super-resolution reconstruction to obtain a reconstructed high-resolution CT image. The present invention uses a deep learning model based on the U-Net architecture for super-resolution reconstruction. The mathematical expression of this model is as follows:

[0094]

[0095] Among them, X is the input low-resolution image, Y is the reconstructed high-resolution image, and F i represents the i-th layer transformation of the network.

[0096] Preferably, the super-resolution reconstruction model of the present invention adopts a multi-level residual structure combining global residual and local residual. The global residual can be expressed as:

[0097]

[0098] The local residual is applied inside each convolutional block. This structure helps alleviate the gradient vanishing problem and improves the convergence speed and performance of the model.

[0099] In addition, the present invention also introduces dilated convolution to expand the receptive field while reducing the number of parameters. The mathematical expression of dilated convolution is:

[0100]

[0101] Among them, l is the expansion rate, F is the input feature map, and k is the convolution kernel.

[0102] Radiomics feature module 3 is connected to image super-resolution reconstruction module 2 to extract high-throughput radiomics features from the reconstructed high-resolution CT images. These features include, but are not limited to, shape features, texture features, intensity features, and wavelet features. For example, shape features can include volume, surface area, and sphericity; texture features can include features derived from gray-level co-occurrence matrices (GLCMs) and run-length matrices (RLGLs); and intensity features can include statistics such as mean, variance, skewness, and kurtosis.

[0103] A significant feature of the present invention is the introduction of dynamic enhancement features. Specifically, the radiomics feature module 3 extracts the dynamic enhancement curve features of the lesion. This can be achieved by the following steps:

[0104] 1. Register multi-temporal images.

[0105] 2. Segment the region of interest (ROI) at each time phase.

[0106] 3. Calculate the average CT value within the ROI and obtain the time-density curve (TDC).

[0107] 4. Extract features from TDC, such as peak enhancement, peak time, washout rate, etc.

[0108] These dynamic features are important for distinguishing early-stage liver cancer from benign lesions.

[0109] The deep feature module 4 is also connected to the image super-resolution reconstruction module 2 to build a deep convolutional neural network model and extract deep features from the reconstructed high-resolution CT image. The present invention adopts an architecture based on the residual network (ResNet) and introduces an attention mechanism. The mathematical expression of the residual block is as follows:

[0110]

[0111] Among them, x and y are the input and output of the residual block respectively, represents the residual map.

[0112] The attention mechanism can be expressed as follows:

[0113]

[0114] Among them, Q, K, and V represent query, key, and value respectively. is the dimension of the key.

[0115] This structural design enables the model to automatically learn more effective feature representations, which is particularly advantageous for detecting small lesions such as early-stage liver cancer.

[0116] The knowledge distillation module 5 is connected to the deep feature module 4 to generate a training set based on expert annotations. This module then trains the deep feature module to develop an interpretable diagnostic strategy. The core concept of knowledge distillation is to use a complex "teacher model" to guide the learning of a relatively simple "student model." In this paper, the diagnostic experience of radiologists is considered the "teacher model," while the deep learning model serves as the "student model."

[0117] The loss function of knowledge distillation can be expressed as:

[0118]

[0119] in, is the cross entropy loss, is the KL divergence, and are the output logits of the teacher model and the student model respectively, T is the temperature parameter, is the balance coefficient.

[0120] In this way, the expert's diagnostic experience can be effectively integrated into the deep learning model, improving the model's interpretability and credibility.

[0121] Early Diagnosis Module 6 connects to Radiomics Feature Module 3, Deep Feature Module 4, and Knowledge Distillation Module 5 to integrate high-throughput radiomics and deep features and generate early liver cancer diagnostic results based on an interpretable diagnostic strategy. This module employs ensemble learning to fuse features from different sources to achieve more robust and accurate diagnostic results.

[0122] Specifically, the early diagnosis module 6 first normalizes the radiomics features and deep features, and then performs feature fusion by weighted summation:

[0123]

[0124] in, and , respectively represent radiomic features and deep features, and is the corresponding weight.

[0125] The fused features are then fed into a final classifier (such as a support vector machine or random forest) to produce a diagnosis. To improve the interpretability of the model, SHAP (SHapley Additive exPlanations) values ​​can be used to explain the contribution of each feature to the final decision.

[0126] The method of the present invention achieves high-precision diagnosis of early-stage liver cancer through multimodal data fusion and deep learning technology. Compared with traditional methods, the present invention has the following advantages:

[0127] 1. Utilizing multi-phase CT images, the dynamic enhancement characteristics of tumors are fully captured, improving the sensitivity to early lesions.

[0128] 2. Through super-resolution reconstruction technology, the image quality is improved without increasing the radiation dose, which is conducive to the detection of tiny lesions.

[0129] 3. Combining manually designed radiomics features and automatically learned deep features, it not only retains the prior knowledge of human experts but also fully utilizes the powerful feature learning ability of deep learning.

[0130] 4. Introducing knowledge distillation technology to incorporate expert experience into the model, improving the interpretability and credibility of the model.

[0131] 5. Adopting ensemble learning strategy to integrate multiple features and diagnostic strategies improves the stability and accuracy of diagnosis.

[0132] In summary, the present invention provides a comprehensive, efficient, and accurate intelligent diagnosis system and method for early liver cancer, which is expected to play an important role in clinical practice and improve the detection rate and diagnostic accuracy of early liver cancer.

[0133] In one embodiment of the present invention, the structure of the CT image module 1 is further clarified. The module includes an image acquisition unit 11, an image preprocessing unit 12, and an image alignment unit 13. These three units work together to ensure the quality of the CT images input to subsequent modules.

[0134] Image acquisition unit 11 is responsible for acquiring multi-phase contrast-enhanced CT images of the patient. Preferably, the present invention utilizes dynamic contrast-enhanced scanning technology. After intravenous injection of an iodinated contrast agent, scans are performed at different time points to obtain CT images in the arterial, portal, and delayed phases. This multi-phase acquisition method fully captures the blood supply characteristics of the tumor and is of great significance for the detection of early-stage liver cancer.

[0135] The image preprocessing unit 12 performs normalization and standardization on the acquired CT images. Normalization typically uses the minimum-maximum normalization method to scale pixel values ​​to the range [0, 1]. Standardization uses the Z-score normalization method to set the image mean to 0 and the standard deviation to 1. These preprocessing steps help eliminate differences caused by different equipment and scanning parameters, improving the stability of subsequent processing.

[0136] The image alignment unit 13 is responsible for registering the multi-phase CT images to eliminate the problem of uneven CT value distribution. The present invention preferably uses a non-rigid registration algorithm, such as B-spline registration or optical flow registration. Taking B-spline registration as an example, its deformation field can be expressed as:

[0137]

[0138] Where x is the image coordinate, is the control point, is the cubic B-spline basis function, and h is the control point spacing.

[0139] By minimizing the mutual information or mean square error between images, the optimal deformation field parameters can be obtained to achieve image alignment.

[0140] In one embodiment of the present invention, the structure of image super-resolution reconstruction module 2 is further refined. This module includes an encoding-decoding module 21, a multi-level residual structure 22, and a dilated convolution unit 23. This design fully considers the characteristics of medical images and can effectively improve image resolution without introducing artifacts.

[0141] The encoding-decoding module 21 is built on the U-Net architecture. An important feature of U-Net is the skip connection, which can directly pass the low-level features of the encoder to the corresponding layer of the decoder, helping to recover the detailed information of the image. The skip connection can be expressed as:

[0142]

[0143] in, represents the identity mapping, represents the residual function.

[0144] The multi-level residual structure 22 uses a combination of global residual and local residual. The global residual helps learn the residual mapping between input and output, while the local residual is applied within each convolutional block, helping the network learn more fine-grained features. This structure can be expressed as:

[0145]

[0146] Among them, F(X) represents the global residual map, ) represents the local residual map.

[0147] The dilated convolution unit 23 is used to reduce the number of parameters while expanding the receptive field. An important parameter of dilated convolution is the dilation rate, which determines the spacing between convolution kernel elements. Preferably, the present invention adopts multi-scale dilated convolution, that is, using dilated convolutions with different dilation rates in the same layer and then splicing the results. This can be expressed as:

[0148]

[0149] in, The expansion rate is The dilated convolution.

[0150] In one embodiment of the present invention, the function of the radiomics feature module 3 is further clarified. This module is responsible for extracting various types of radiomics features, including lesion dynamic enhancement curve features, sequence features, texture features, and shape features.

[0151] The dynamic enhancement curve characteristics of the lesion reflect the blood supply characteristics of the tumor and are an important indicator for distinguishing benign and malignant lesions. The present invention extracts such characteristics through the following steps:

[0152] 1. Segment the region of interest (ROI) on each phase image.

[0153] 2. Calculate the average CT value within the ROI and obtain the time-density curve (TDC).

[0154] 3. Extract features from TDC, such as peak enhancement (PE), time to peak (TTP), washout rate (WOR), etc.

[0155] These features can be expressed as:

[0156]

[0157]

[0158]

[0159] in, is the baseline CT value.

[0160] Sequential features reflect the changing patterns of lesions at different time phases. This paper uses a method based on sequential pattern mining to extract these features. Specifically, the changing pattern of CT values ​​is encoded as a sequence, and then a sequential pattern mining algorithm (such as PrefixSpan) is used to discover frequent patterns.

[0161] Texture features reflect the heterogeneity within the lesion. This paper uses methods such as gray-level co-occurrence matrix (GLCM) and run-length matrix (RLGL) to extract texture features. Taking GLCM as an example, the following features can be calculated:

[0162] Texture features reflect the heterogeneity within the lesion. The present invention uses methods such as gray-level co-occurrence matrix (GLCM) and run-length matrix (RLGL) to extract texture features. Taking GLCM as an example, the following features can be calculated:

[0163] energy:

[0164] Contrast ratio:

[0165] Dependencies:

[0166] in, are elements in GLCM.

[0167] Shape features reflect the geometric properties of the lesion. The shape features extracted by the present invention include but are not limited to: volume:

[0168] Surface area:

[0169] Sphericity:

[0170] in, is the binary image after segmentation.

[0171] In one embodiment of the present invention, the structure of the deep feature module 4 is further refined. The module includes a temporal attention model 41 and a convolutional neural network model 42, which work together to capture complex features in CT images.

[0172] The temporal attention model 41 is used to model dynamic temporal attention. The present invention adopts a temporal modeling method based on the self-attention mechanism. Specifically, for the input sequence , first calculate the self-attention:

[0173]

[0174] Among them, Q, K, and V are query, key, and value matrices respectively, which are obtained by linear transformation of X.

[0175] Then, a gating mechanism is used to fuse the original features and the attention features:

[0176]

[0177]

[0178] in, The attention characteristics of is the gate value, Represents element-wise multiplication.

[0179] The convolutional neural network model 42 is used to implement feature extraction and fusion. The present invention adopts an architecture based on a residual network and introduces a dilated convolution and attention mechanism. The mathematical expression of the residual block is as follows:

[0180]

[0181] in, represents the residual map.

[0182] The use of dilated convolution can expand the receptive field without increasing the number of parameters. The present invention adopts multi-scale dilated convolution, that is, using dilated convolutions with different dilation rates in the same layer and then splicing the results.

[0183] In addition, the present invention also uses spatial attention and channel attention mechanisms. Spatial attention can help the model focus on important areas in the image, while channel attention can highlight important feature channels. These two attention mechanisms can be expressed as:

[0184]

[0185]

[0186] Among them, F is the input feature map, represents 7X7 convolution, and MLP represents multi-layer perceptron.

[0187] The structure of the temporal attention model 41 is further clarified. The model includes an encoding module 411 and a decoding module 412, which together extract and model the temporal features of CT images.

[0188] The encoding module 411 includes a convolutional layer, a fully connected layer, and a pooling layer. The convolutional layer is used to extract local features and can be expressed as:

[0189]

[0190] Where * represents the convolution operation and f is the activation function (such as ReLU).

[0191] The fully connected layer is used to integrate global information and can be expressed as:

[0192]

[0193] The pooling layer is used to reduce the feature dimension and increase the translation invariance of the model. The maximum pooling can be expressed as:

[0194]

[0195] in, Represents the pooling area centered at (i, j).

[0196] The decoding module 412 includes a dense attention layer. The core idea of ​​the dense attention layer is to calculate attention at different scales and then fuse the results. This can be expressed as:

[0197]

[0198]

[0199] in, represents the feature transformation of the kth scale, and W is the weight matrix used for feature fusion.

[0200] Through this design, the temporal attention model 41 can effectively capture the changing characteristics of CT images in the time dimension, providing an important basis for the diagnosis of early liver cancer.

[0201] Overall, this invention achieves comprehensive analysis of multi-phase contrast-enhanced CT images by refining the structure and functionality of each module. From image acquisition, preprocessing, and super-resolution reconstruction to radiomic feature extraction, deep feature learning, and time series modeling, each step has been carefully designed to maximize the accuracy of early liver cancer diagnosis. This multimodal, multi-scale analysis method is expected to significantly improve the detection rate of early liver cancer and provide important support for timely treatment of patients.

[0202] In one embodiment of the present invention, the function of the knowledge distillation module 5 is further clarified. This module is designed to effectively integrate expert experience into the deep learning model to improve the interpretability and credibility of the model. Specifically, the knowledge distillation module 5 performs the following steps:

[0203] First, expert experience is used as a new sample. This expert experience can include radiologists' diagnostic results on CT images, interpretations of specific imaging features, etc. Preferably, the present invention records expert experience in a structured manner, such as using a predefined feature dictionary and scoring criteria.

[0204] Secondly, the expert experience is converted into a feature vector through the feature extraction network. The purpose of this step is to convert qualitative expert experience into quantifiable feature representations. The feature extraction network can use a pre-trained natural language processing model such as BERT. For the input expert experience text T, the feature extraction process can be expressed as:

[0205] v=BERT(T)

[0206] Where v is the extracted feature vector.

[0207] Then, the feature vector is matched with the prediction result of the deep feature module. The core of this step is to calculate the similarity between the expert experience feature vector and the model prediction result. The present invention uses cosine similarity as the matching metric:

[0208]

[0209] Here, p is the prediction vector of the model.

[0210] Finally, the prediction results of the deep feature module are optimized. This step is achieved by introducing knowledge distillation loss to "distill" the expert knowledge into the model. The knowledge distillation loss can be expressed as:

[0211]

[0212] in, and are the output logits of the teacher model (expert) and the student model (deep learning model), T is the temperature parameter, KL is the Kullback-Leibler divergence, CE is the cross entropy loss, is the balance coefficient.

[0213] In this way, the present invention achieves effective integration of expert knowledge and improves the interpretability and clinical acceptability of the model.

[0214] The structure of the early diagnosis module 6 is further refined. The module includes a multi-feature fusion unit 61, a prediction model construction unit 62 and a diagnosis result generation unit 63, which work together to generate the final diagnosis result.

[0215] The multi-feature fusion unit 61 is responsible for integrating high-throughput radiomics features and deep features. The present invention adopts a feature fusion method based on the attention mechanism. Specifically, for radiomics features and deep features , the fusion process can be expressed as:

[0216]

[0217]

[0218] in, and is a learnable weight matrix.

[0219] The prediction model construction unit 62 constructs an early-stage liver cancer identification prediction model based on an interpretable diagnostic strategy. The present invention employs the concept of ensemble learning, combining multiple basic classifiers to improve the model's generalization capabilities. Preferably, this method uses a random forest as the ensemble model. For the input feature x, the prediction of the random forest can be expressed as:

[0220]

[0221] Where M is the number of decision trees, Represents the mth decision tree.

[0222] The diagnosis result generating unit 63 generates an early liver cancer diagnosis result using the early liver cancer recognition prediction model. To improve the interpretability of the result, the present invention also introduces the SHAP value to explain the contribution of each feature to the final decision. The SHAP value can be calculated using the following formula:

[0223]

[0224] Among them, F is the set of all features, S is the subset that does not contain feature i, is the model prediction using only the feature subset S.

[0225] In one embodiment of the present invention, the present invention introduces a lesion growth prediction module 7, which further enhances the clinical value of the system. This module is connected to the early diagnosis module 6 and is used to predict the growth of the lesion and generate early diagnosis suggestions.

[0226] The lesion growth prediction module 7 first constructs a time series model based on the early liver cancer diagnosis results and the patient's historical medical data. The present invention preferably uses a long short-term memory network (LSTM) to capture the long-term dependencies of lesion growth.

[0227] Based on the LSTM prediction results, the lesion growth prediction module 7 also generates personalized early diagnosis recommendations. These recommendations take into account factors such as lesion growth rate, patient age, and liver function status, providing comprehensive support for clinical decision-making.

[0228] Finally, the invention provides a complete intelligent diagnosis method for early liver cancer based on multi-phase enhanced CT. The method includes the following steps:

[0229] S1: Acquire and preprocess the patient's multi-phase contrast-enhanced CT images. This step is typically performed in a hospital's radiology department using high-end CT equipment (such as 64-slice or higher spiral CT). Preprocessing includes image denoising and normalization.

[0230] S2: Perform super-resolution reconstruction on the pre-processed multi-phase enhanced CT image to obtain a reconstructed high-resolution CT image. This step is completed using the aforementioned image super-resolution reconstruction module 2, which can significantly improve the image details.

[0231] S3: Extract high-throughput radiomic features from the reconstructed high-resolution CT images. This step is completed using the radiomic feature module 3. The extracted features include multiple dimensions such as shape, texture, and intensity.

[0232] S4: Extract deep features from the reconstructed high-resolution CT images using a deep convolutional neural network model. This step is performed using the Deep Feature Module 4, which automatically learns complex image features.

[0233] S5: Based on the expert annotations, a training set is formed and the deep convolutional neural network model is trained to form an interpretable diagnostic strategy. This step is completed using the knowledge distillation module 5, effectively integrating expert experience into the model.

[0234] S6: Integrate high-throughput radiomics features and deep features to generate early liver cancer diagnosis results based on an interpretable diagnostic strategy. This step is completed using the Early Diagnosis Module 6, which comprehensively considers multiple features to improve diagnostic accuracy.

[0235] S7: Based on the early liver cancer diagnosis results and the patient's historical medical data, the lesion growth is predicted and early diagnosis recommendations are generated. This step is completed using the Lesion Growth Prediction Module 7, providing support for subsequent clinical decision-making.

[0236] Through this series of steps, the method of the present invention realizes the intelligentization of the entire process from image acquisition to diagnosis suggestion generation, providing a powerful tool for accurate diagnosis of early liver cancer. Compared with traditional methods, this method has the following advantages:

[0237] 1. Multimodal data fusion: Combining radiomics features and deep learning features to comprehensively capture the imaging characteristics of tumors.

[0238] 2. Utilization of temporal information: Through multi-phase CT images and temporal attention models, the dynamic enhancement characteristics of tumors are fully utilized.

[0239] 3. Knowledge integration: Through knowledge distillation technology, expert experience is effectively integrated into the model, improving interpretability.

[0240] 4. Predictive analysis: It not only provides current diagnostic results but also predicts the future growth of lesions, providing more comprehensive information for clinical decision-making.

[0241] In summary, the intelligent early-stage liver cancer diagnosis system and method based on multi-phase contrast-enhanced CT provided by this invention achieves high accuracy, efficiency, and interpretability in early-stage liver cancer diagnosis through the organic integration of multiple innovative technologies. This method is expected to play an important role in clinical practice, improving the detection rate of early-stage liver cancer and contributing to timely treatment and improved prognosis for patients.

[0242] The above description is only a preferred specific embodiment of the present invention; however, the protection scope of the present invention is not limited thereto; any person familiar with the art who, within the scope disclosed by the present invention, makes substitutions or changes based on the scheme of the present invention and its improved concepts shall be covered within the protection scope of the present invention.

Claims

1. An intelligent diagnosis system for early liver cancer based on multi-phase enhanced CT, characterized by: include: CT imaging module for: Acquire multi-phase enhanced CT images of patients; Preprocessing the multi-phase enhanced CT image; An image super-resolution reconstruction module, connected to the CT image module, is used to: receiving the preprocessed multi-phase enhanced CT image sent by the CT image module; Performing super-resolution reconstruction based on the preprocessed multi-phase enhanced CT image to obtain a reconstructed high-resolution CT image; The radiomics feature module is connected to the image super-resolution reconstruction module and is used to: extracting high-throughput radiomics features from the reconstructed high-resolution CT image; A deep feature module, connected to the image super-resolution reconstruction module, is used to: Build a deep convolutional neural network model; Extracting depth features from the reconstructed high-resolution CT image using the deep convolutional neural network model; The knowledge distillation module is connected to the deep feature module and is used to: Form a training set based on expert annotations; Training the deep feature module to form an interpretable diagnostic strategy; An early diagnosis module, connected to the radiomics feature module, the deep feature module, and the knowledge distillation module, is used to: integrating the high-throughput radiomics features and the deep features; Based on the interpretable diagnostic strategy, early liver cancer diagnosis results are generated; The image super-resolution reconstruction module includes: The encoding-decoding module is built based on the U-Net architecture; Multi-level residual structure, which combines global residual and local residual; Atrous convolution unit, used to reduce the number of parameters; The encoding-decoding module includes alternating convolutional layers and residual convolutional layers; The deep feature module includes: Temporal attention model, used to model dynamic temporal attention; Convolutional neural network model for feature extraction and fusion; The deep feature module uses dilated convolution and attention mechanisms to build a deep convolutional network model, and uses image pyramid and residual structure to optimize the network model; The temporal attention model includes: Encoding module, including convolutional layer, fully connected layer and pooling layer; Decoding module, including dense attention layer; The encoding module is used to extract CT image features and lesion temporal patterns, and the decoding module is used to extract correlations between features; The knowledge distillation module is used to: Use expert experience as new samples; Converting the expert experience into a feature vector through a feature extraction network; Matching the feature vector with the prediction result of the deep feature module; Optimizing the prediction results of the deep feature module; Also includes: The lesion growth prediction module is connected to the early diagnosis module and is used to: Predicting the growth of the lesion based on the early liver cancer diagnosis results and the patient's historical medical data; Generate early diagnosis recommendations.

2. The system according to claim 1, wherein: The CT image module includes: An image acquisition unit, used for acquiring multi-phase enhanced CT images of the patient; An image preprocessing unit, configured to perform normalization and standardization on the multi-phase enhanced CT image; The image alignment unit is used to register the multi-phase enhanced CT images to eliminate the problem of uneven distribution of CT values.

3. The system according to claim 1, wherein: The radiomics feature module is used to: Extract the dynamic enhancement curve characteristics of the lesion; Extract sequence features; Extract texture features; Extract shape features; Among them, the radiomics feature module uses a multi-channel convolutional neural network layer to extract high-throughput image features, and uses a deep convolutional neural network layer to fuse high-throughput image features and high-throughput image features.

4. The system according to claim 1, wherein: The early diagnosis module includes: a multi-feature fusion unit, configured to integrate the high-throughput radiomics features and the deep features; A prediction model building unit, configured to build a prediction model for identifying early-stage liver cancer based on the interpretable diagnostic strategy; The diagnosis result generating unit is used to generate an early liver cancer diagnosis result using the early liver cancer recognition and prediction model.

Images