Multi-stage outcome prediction method based on multi-classification-head cascading and inter-stage fusion loss

Abstract

The invention relates to the technical field of artificial intelligence and medical image processing, in particular to a multi-stage outcome prediction method based on multi-classification head cascading and inter-stage fusion loss, which comprises the following steps of: establishing a multi-stage outcome prediction model consisting of an encoder and a multi-classification head cascading module, a cross-period fusion loss function is designed based on Focal loss and is used for constraining the consistency of multi-period prediction; performing end-to-end training on the multi-period outcome prediction model by adopting an Adam optimizer to obtain an optimal multi-period outcome prediction model; and outputting short-term, middle-term and long-term outcome probabilities of the patient by using the optimal multi-stage outcome prediction model. According to the method, the multi-classification head and the cross-period fusion loss are combined, and the classification intermediate features of the previous period are fused in a cascade mode to serve as the auxiliary input of the next period, so that the constructed multi-period outcome prediction model can output the high-precision outcome probabilities of three time points through the single acute-period CT image.

Description

Technical Field

[0001] This invention relates to the fields of artificial intelligence and medical image processing technology, specifically to a multi-stage outcome prediction method based on multi-class head cascade and inter-stage fusion loss. Background Technology

[0002] In the clinical management of lung diseases (such as severe pneumonia and pulmonary fibrosis), physicians often rely on imaging information to determine the disease progression and develop follow-up plans. Traditional prediction methods are usually based on assessments at a single point in time and lack in-depth modeling of the changing trends of the patient's condition across time scales, making it difficult to meet the needs of personalized follow-up and intervention.

[0003] In recent years, deep learning has been widely used in medical image diagnosis. However, most current models focus on single-stage tasks. At the same time, the outcome states at different time points have a certain temporal continuity and logical consistency. Currently, there is a lack of introduction of temporal information, which means that the current prediction results cannot be used to guide the prediction of previous and later stages, resulting in insufficient ability to predict the outcome of patients in multiple stages. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-stage outcome prediction method based on multi-class head cascade and inter-stage fusion loss, in order to solve the technical problem that existing multi-stage medical image prediction models cannot effectively model the disease progression pattern across time points and have insufficient multi-stage outcome prediction capabilities.

[0005] To solve the above-mentioned technical problems, the present invention specifically provides the following technical solution: A multi-period outcome prediction method based on multi-class head concatenation and inter-period fusion loss includes the following steps: Acquire chest CT images of acute lung diseases and perform standardized resampling, cropping and intensity normalization on the chest CT images; A multi-period outcome prediction model consisting of an encoder and a multi-classification head cascade module was established, and a cross-period fusion loss function based on Focal loss was designed to constrain the consistency of multi-period predictions. The Adam optimizer is used for end-to-end training of the multi-period outcome prediction model. With the goal of minimizing the fusion loss, the multi-period outcome prediction model is iteratively optimized to obtain the optimal multi-period outcome prediction model. Using the optimal multi-stage outcome prediction model, the short-term, medium-term, and long-term outcomes of patients are output based on their chest CT images.

[0006] As a preferred embodiment of the present invention, the encoder is used to extract high-dimensional image features from chest CT images. .

[0007] As a preferred embodiment of the present invention, the multi-classification head cascade module includes three cascaded classification heads, each of which consists of two linear layers and a softmax activation layer. The first classification head is used to classify high-dimensional image features. Given the input, output the short-term outcome probability; The structural expression of the first classification head is: In the formula, The short-term regression probability is the output of the first classifier. , These are the first and second linear layers in the classification head, respectively. High-dimensional image features Intermediate classification features generated through the first linear layer of the first classification head. This is the softmax activation layer.

[0008] As a preferred embodiment of the present invention, the second classification head is used to classify high-dimensional image features. and the intermediate classification features Compositional fusion characteristics Given the input, output the intermediate outcome probability; The structural expression for the second classification head is: In the formula, This represents the intermediate regression probability output by the second classification head. , These are the first and second linear layers in the classification head, respectively. For fusion features Intermediate classification features generated through the first linear layer of the second classification head. This is a softmax activation layer, and Concat is the feature concatenation operator.

[0009] As a preferred embodiment of the present invention, the third classification head is used to classify high-dimensional image features. The intermediate classification features and the intermediate classification features Compositional fusion characteristics Given the input, output the long-term outcome probability; The structural expression of the third classification head is: In the formula, This represents the long-term regress probability output by the third classification head. , These are the first and second linear layers in the classification head, respectively. This is a softmax activation layer, and Concat is the feature concatenation operator.

[0010] As a preferred embodiment of the present invention, the fusion loss function is: In the formula, The total loss of the multi-period outcome prediction model. For short-term sub-fusion loss, For intermediate sub-fusion loss, For long-term sub-fusion loss; In the formula, i represents the corresponding number for short-term, medium-term, and long-term. , where j is the corresponding number of the classification head used for short-term, medium-term, and long-term outcome probability prediction. Focal Loss is the prediction result of the gold standard for the i-th period on the j-th class head. In the formula, For the gold standard of the i-th period, Let be the positive probability output by the j-th classification header. Let be the negative probability output by the j-th classification header. An adjustment factor to balance the contribution of difficult and easy samples to the loss; Among them, when This indicates that the positive result has not turned into a positive outcome. This indicates that the negative result has been confirmed. when hour, middle Used to constrain the loss of positive samples predicted by the current classification head. Used to constrain the loss of negative samples predicted by the current classification head; when hour, The gold standard used to constrain the positive probability output by the previous classification head using the current period's positive samples; when hour, The negative probability of the classification head output is used to constrain the negative sample output in the later period using the gold standard of the current period's negative samples.

[0011] In a preferred embodiment of the present invention, the Adam optimizer undergoes end-to-end iterative training, dynamically calculating the short-term, medium-term, and long-term total losses during the training process. Then, backpropagation is performed until the multi-period outcome prediction model achieves the best accuracy on the validation set, thus obtaining the optimal multi-period outcome prediction model.

[0012] Compared with the prior art, the present invention has the following advantages: This invention designs three classification heads corresponding to short-term (1 month), medium-term (3 months), and long-term (6 months) prediction tasks. The intermediate classification features of the previous period are fused in a cascade manner as auxiliary inputs for the next period to guide the temporal information. Furthermore, an improved cross-period fusion loss function based on Focal Loss is introduced to achieve continuity constraints between prediction results of different periods. This enables the constructed multi-period outcome prediction model to output high-precision outcome probabilities for three time points using a single acute-phase CT image. Attached Figure Description

[0013] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0014] Figure 1 A flowchart of a multi-period outcome prediction method based on multi-class head concatenation and inter-period fusion loss provided in an embodiment of the present invention; Figure 2 This is a structural diagram of the multi-period outcome prediction model provided in an embodiment of the present invention; Figure 3 Performance evaluation of multi-class head concatenation and inter-period fusion loss provided in embodiments of the present invention; Figure 4 Granularity performance evaluation of inter-period fusion loss provided in embodiments of the present invention; Figure 5 Evaluation of the effectiveness of the multi-period outcome prediction model provided in the embodiments of the present invention; Figure 6 This is a graph showing the predicted outcome of the regression probability provided in an embodiment of the present invention. Figure 7 This is a structural diagram of Wo_Both provided in an embodiment of the present invention; Figure 8 The W_CH structure diagram provided in the embodiments of the present invention; Figure 9 This is a structural diagram of Wo_CFL provided in an embodiment of the present invention; Figure 10 This is a structural diagram of the encoder provided in an embodiment of the present invention; Figure 11 Performance evaluation of different encoder structures provided in embodiments of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] like Figure 1 As shown, this invention provides a multi-period outcome prediction method based on multi-class head concatenation and inter-period fusion loss, comprising the following steps: Acquire chest CT images of acute lung diseases and perform standardized resampling, cropping and intensity normalization on the chest CT images to unify the chest CT images into a three-dimensional input tensor with fixed voxel size; A multi-period outcome prediction model consisting of an encoder and a multi-classification head cascade module was established, and a cross-period fusion loss function based on Focal loss was designed to constrain the consistency of multi-period predictions. The Adam optimizer is used for end-to-end training of the multi-period outcome prediction model. With the goal of minimizing the fusion loss, the multi-period outcome prediction model is iteratively optimized to obtain the optimal multi-period outcome prediction model. Using the optimal multi-stage outcome prediction model, based on the patient's chest CT images, the model outputs the short-term, medium-term, and long-term probabilities of the patient's outcome, such as... Figure 6 As shown.

[0017] Encoders are used to extract high-dimensional image features from chest CT images. The encoder uses a ResNet encoder, such as Figure 10 As shown, other types of encoders can also be used, among which the ResNet encoder has the best performance. It consists of an input convolutional layer, multiple residual blocks, an average pooling layer, and a linear transform layer. The structure expression of the encoder is as follows: In the formula, Features of high-dimensional images Chest CT image, It is a ResNet encoder.

[0018] Ablation analysis was performed on the ResNet encoder structure. The prediction performance of the improved ResNet encoder of this invention was compared with that of commonly used DenseNet and ViTNet encoders. In the comparison, the network architecture remained unchanged except for the encoder components. The prediction results of the three compared encoders are shown below. Figure 11 As shown, the ResNet encoder (ours) provided by this invention achieved the best prediction performance across all follow-up periods. The ViTNet encoder had the lowest performance metrics. It can be observed that ResNet can achieve more reasonable localization. Therefore, this demonstrates both the effectiveness of the encoder for the current problem (multi-period outcome prediction) and the superior performance of the ResNet encoder in solving the current problem.

[0019] In constructing a multi-stage outcome prediction model for predicting the short-term, medium-term, and long-term probabilities of patient outcomes, this invention selects three classification heads corresponding to short-term (1 month), medium-term (3 months), and long-term (6 months) prediction tasks during the model structure design stage. These classification heads are then cascaded and integrated with intermediate features from the previous period as auxiliary inputs for the next period. This approach guides the time-series information, enables the synchronous execution of multi-stage outcome prediction tasks, and improves outcome prediction efficiency.

[0020] The multi-classification head cascade module includes three cascaded classification heads, each consisting of two linear layers and a softmax activation layer. The first classification head is used to classify high-dimensional image features. Given the input, output the short-term outcome probability; The structural expression for the first classification head is: In the formula, The short-term regression probability is the output of the first classifier. , These are the first and second linear layers in the classification head, respectively. High-dimensional image features Intermediate classification features generated through the first linear layer of the first classification head. This is the softmax activation layer.

[0021] The second classification head is used to classify high-dimensional image features. and intermediate classification features Compositional fusion characteristics Given the input, output the intermediate outcome probability; The structural expression for the second classification head is: In the formula, This represents the intermediate regression probability output by the second classification head. , These are the first and second linear layers in the classification head, respectively. For fusion features Intermediate classification features generated through the first linear layer of the second classification head. This is a softmax activation layer, and Concat is the feature concatenation operator.

[0022] The third classification head is used to classify high-dimensional image features. Intermediate classification features and intermediate classification features Compositional fusion characteristics Given the input, output the long-term outcome probability; The structural expression for the third classification head is: In the formula, This represents the long-term regress probability output by the third classification head. , These are the first and second linear layers in the classification head, respectively. This is a softmax activation layer, and Concat is the feature concatenation operator.

[0023] In this invention, the first classification head is based solely on high-dimensional image features. The output transition probability is used. The second and third classifiers introduce the intermediate features of the first and the first two classifiers respectively and then combine them as auxiliary information to participate in the prediction. Through the additional design of the cascaded information transmission path between the three classifiers, the classifiers in later periods (medium and long term) can incorporate the information of the earlier periods when making predictions.

[0024] Compared to the first classification head, the second and third classification heads, corresponding to the March and June periods respectively, not only include the image features extracted by the encoder but also incorporate the intermediate classification features from their respective preceding classification heads. Taking the third classification head corresponding to the June period as an example, its input is a concatenation of image features and the intermediate classification features from the first and second classification heads. This cascading classification feature fusion strategy indirectly introduces short-, medium-, and long-term temporal prior information into the classification head's learning target mapping, thus allowing the longer-term classification head to refer to the shorter-term classification response when making decisions. This also aligns with the gradual progression of short-, medium-, and long-term outcomes in clinical practice.

[0025] In constructing a multi-period outcome prediction model for predicting the short-term, medium-term, and long-term probabilities of patient outcomes, this invention introduces an improved inter-period fusion loss function based on Focal Loss during the model structure training phase. This achieves continuity constraints between prediction results at different periods, thereby improving the accuracy of outcome prediction.

[0026] This invention employs Focal Loss as a constraint for each classification head. This loss enhances the model's focus on difficult samples by increasing the weight of those samples, thereby mitigating the problem of imbalanced samples. Furthermore, each classification head includes an additional inter-period loss term. This term incorporates the current period's gold standard label into the supervision of classification heads in other periods, constraining the temporal consistency of prediction results. This additionally constructed inter-period consistency loss term achieves continuity constraints between prediction results from different periods.

[0027] The continuity constraint ensures that the predictions of a multi-phase prediction model are as consistent as possible in time across the three phases. For example, if the intermediate-term prediction leans towards a positive outcome (higher probability of negative), then the long-term prediction for that patient should also lean towards a positive outcome (higher probability of negative). This avoids the shortcomings of other existing methods that do not consider the temporal consistency of patient outcomes when making predictions across different phases. Specifically, it may result in the same patient being predicted to have a positive outcome in the intermediate phase, but then being predicted not to have a positive outcome in the long term, which does not reflect the actual prognostic trend.

[0028] This invention designs a cross-period fusion loss based on Focal Loss to constrain the consistency and continuity of predictions made by the network across different time periods. This cross-period fusion total loss... Composed of three sub-loss items The components correspond to the three monthly periods, 1, 3, and 6, as shown in the following formula. Each sub-fusion loss term includes not only the Focal Loss constraint for the current period, but also constraints on the positive predictions for the preceding months and the negative predictions for the following months.

[0029] Focal Loss is a loss function specifically designed to address class imbalance. By dynamically adjusting the weights of easily classified and difficult-to-classify samples, it makes the model pay more attention to difficult-to-classify samples, thereby improving the model's performance in object detection and rare class identification.

[0030] The fusion loss function is: In the formula, The total loss of the multi-period outcome prediction model. For short-term sub-fusion loss, For intermediate sub-fusion loss, For long-term sub-fusion loss; In the formula, i represents the corresponding number for short-term, medium-term, and long-term. , where j is the corresponding number of the classification head used for short-term, medium-term, and long-term outcome probability prediction. Focal Loss is the prediction result of the gold standard for the i-th period on the j-th class head. In the formula, For the gold standard of the i-th period, Let be the positive probability output by the j-th classification header. Let be the negative probability output by the j-th classification header. To balance the adjustment factor for the contribution of difficult and easy samples to the loss, it is empirically set to 2, which gives the model the ability to adaptively focus on difficult samples during iterative updates. Among them, when This indicates that the positive result has not turned into a positive outcome. This indicates that the negative result has been confirmed. when hour, middle Used to constrain the loss of positive samples predicted by the current classification head. Used to constrain the loss of negative samples predicted by the current classification head; when hour, The gold standard used to constrain the positive probability output by the previous classification head using the current period's positive samples; when hour, The negative probability of the classification head output is used to constrain the negative sample output in the later period using the gold standard of the current period's negative samples.

[0031] Taking the medium term (3-month period) as an example, the corresponding sub-fusion loss term is... As in the formula: like Figure 2 As shown, the constraints corresponding to the three loss terms are illustrated (dashed lines of different colors). Taking the constraint of the loss term corresponding to the mid-term classification head as an example (the constraint consists of two parts: the constraint on the current period and the constraints on the short-term and long-term): First, the mid-term gold standard of the current sample will supervise the output of the current mid-term prediction (that is, if the gold standard is positive, it will constrain the current mid-term classification head to make a positive prediction as much as possible, and vice versa). In addition, through the additional design of the loss term ( Item 1 and item 3 Constraints are applied to both the short-term and long-term ends, specifically: a) When the interim gold standard for the sample is positive, proceed through item 1 (corresponding to...) Figure 2 The intermediate loss term points to the red dotted line of the short-term classification head, which can additionally supervise the short-term classification head to make it as likely as possible to make a short-term positive judgment (in clinical practice, if a patient's intermediate loss is positive, then their short-term loss can also be inferred to be positive). b) When the interim gold standard for the sample is negative, proceed through item 3 (corresponding to...). Figure 2 The intermediate-term loss term points to the red dashed line of the long-term classification head, which can additionally supervise the long-term classification head to make a long-term negative judgment as much as possible (similarly, if a patient is negative in the intermediate stage, their long-term should also remain negative).

[0032] This invention introduces temporal modeling and consistency constraints from both the network structure and loss function mechanism aspects, which significantly improves the performance and usability of multi-period prediction models.

[0033] The Adam optimizer undergoes end-to-end iterative training, dynamically calculating the short-term, medium-term, and long-term total loss during training. Then, backpropagation is performed until the multi-period outcome prediction model achieves the best accuracy on the validation set, thus obtaining the optimal multi-period outcome prediction model.

[0034] This invention performs ablation analysis at the granular level of architecture and fusion path. At the architecture level, it explores whether multi-class head concatenation and inter-period fusion loss components truly contribute to prediction performance through implementation. Specifically, three additional unique settings are implemented: 1) without multi-class head concatenation and inter-period fusion loss (Wo_Both), such as... Figure 7 As shown, 2) only multi-class header cascade (W_CH), such as Figure 8 As shown in Figure 3), only the inter-period fusion loss (W_CFL) exists, as... Figure 9 As shown, and along with 4) a multi-period prediction model (Ours) that simultaneously possesses multi-class head concatenation and inter-period fusion loss, such as Figure 2 As shown, the comparison is performed, and the results are as follows. Figure 3 As shown, the combination of multi-class head concatenation and inter-period fusion loss generally outperforms the models with the other three settings. Furthermore, adding concatenated heads or inter-period fusion significantly improves performance. The results indicate that the synergistic effect of multi-class head concatenation and inter-period fusion loss leads to significantly more accurate predictions than the model with the Wo_Both setting. It can be observed that the combination of multi-class head concatenation and inter-period fusion loss can highlight more reasonable regions for more reliable predictions.

[0035] like Figure 4As shown, this invention performs ablation analysis on the granularity of inter-period fusion loss. Specifically, it compares the full inter-period fusion path (Ours) at the short, medium, and long time points with two partial inter-period fusion settings at the short-to-long and medium-to-long time points, respectively denoted as Short&Long and Medium&Long. The corresponding performance of Ours, Short&Long, and Medium&Long is shown in... Figure 4 The results show that the fusion of learning features from the three periods is crucial, and further demonstrates that temporal relationships can be better modeled by considering the cross-period fusion of all time points.

[0036] like Figure 5 As shown, this invention compares four state-of-the-art methods with a multi-phase prediction model (Ours) using multi-class head cascade and interphase fusion loss to evaluate the effectiveness of the proposed multi-phase prediction model in predicting the probability of multi-phase outcomes from acute-phase CT images. The four methods include: 1) Mikhail et al. developed a U-Net-based multitask model for pneumonia classification, in which pneumonia segmentation is used as an auxiliary task to enhance feature capture (from CT-Based COVID-19 triage: Deep multitask learning improves joint identification and severity quantification.).

[0037] 2) Simon et al. input randomly selected 2D CT slices from different slices into the Inception-ResNet-v2 model to predict the prognosis of progressive fibrotic lung disease (from Deep Learning-based Outcome Prediction in Progressive Fibrotic Lung Disease Using High-Resolution Computed Tomography).

[0038] 3) Yun et al. used CNN to extract depth features from multiple CT slices of chronic obstructive pulmonary disease to predict 3-year and 5-year survival rates (derived from Deep radiomics-based survival prediction inpatients with chronic obstructive pulmonary disease).

[0039] 4) Wang et al. developed an attention alignment model to capture high-risk lesion changes from mammograms to predict the occurrence of breast cancer within 1 to 5 years (in International Conference on Medical Image Computing and Computer-Assisted Intervention.).

[0040] To ensure fair comparison, all methods were trained on the same settings on the dataset of this invention: input was restricted to acute-phase lung CT scans, and output was reformulated to predict short-, medium-, and long-term outcomes using a classification head. The performance of the four comparison methods and the method proposed in this invention was evaluated on the test set of this invention, as detailed below. Figure 5 The proposed method achieved best predictive performance across all time periods, with an average AUC of 0.832 and an accuracy of 0.779. The methods of Simon et al. and Yun et al. significantly underperformed other methods, possibly because 2D slice input cannot provide a global representation of pneumonia lesions. Mikhail et al.'s multi-task framework achieved comparable results in short-term predictions, possibly attributed to improved lesion feature extraction capabilities through a segmentation auxiliary task. Similarly, the attention alignment mechanism designed by Wang et al., designed to focus on changes in high-risk lesions, demonstrated strong performance in short-term predictions. However, the method of this invention outperformed the comparative methods in most short, medium, and long-term periods. Therefore, the effectiveness of the cascaded multi-classifier head architecture with inter-period fusion loss was further confirmed, explicitly modeling the temporal relationships between different follow-up stages.

[0041] This invention designs three classification heads corresponding to short-term (1 month), medium-term (3 months), and long-term (6 months) prediction tasks. The intermediate classification features of the previous period are fused in a cascade manner as auxiliary inputs for the next period to guide the temporal information. Furthermore, an improved cross-period fusion loss function based on Focal Loss is introduced to achieve continuity constraints between prediction results of different periods. This enables the constructed multi-period outcome prediction model to output high-precision outcome probabilities for three time points using a single acute-phase CT image.

[0042] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. A multi-period outcome prediction method based on multi-class head concatenation and inter-period fusion loss, characterized in that, Includes the following steps: Acquire chest CT images of acute lung diseases and perform standardized resampling, cropping and intensity normalization on the chest CT images; A multi-period outcome prediction model consisting of an encoder and a multi-classification head cascade module was established, and a cross-period fusion loss function based on Focal loss was designed to constrain the consistency of multi-period predictions. The Adam optimizer is used for end-to-end training of the multi-period outcome prediction model. With the goal of minimizing the fusion loss, the multi-period outcome prediction model is iteratively optimized to obtain the optimal multi-period outcome prediction model. Using the optimal multi-stage outcome prediction model, the short-term, medium-term, and long-term outcomes of patients are output based on their chest CT images.

2. The multi-period outcome prediction method based on multi-class head concatenation and inter-period fusion loss according to claim 1, characterized in that: The encoder is used to extract high-dimensional image features from chest CT images. .

3. The multi-period outcome prediction method based on multi-class head concatenation and inter-period fusion loss according to claim 2, characterized in that: The multi-classification head cascade module includes three cascaded classification heads, each consisting of two linear layers and a softmax activation layer. The first classification head is used to classify high-dimensional image features. Given the input, output the short-term outcome probability; The structural expression of the first classification head is: In the formula, The short-term regression probability is the output of the first classifier. , These are the first and second linear layers in the classification head, respectively. High-dimensional image features Intermediate classification features generated through the first linear layer of the first classification head. This is the softmax activation layer.

4. The multi-period outcome prediction method based on multi-class head concatenation and inter-period fusion loss according to claim 3, characterized in that: The second classification head is used to classify high-dimensional image features. and the intermediate classification features Compositional fusion characteristics Given the input, output the intermediate outcome probability; The structural expression for the second classification head is: In the formula, This represents the intermediate regression probability output by the second classification head. , These are the first and second linear layers in the classification head, respectively. For fusion features Intermediate classification features generated through the first linear layer of the second classification head. This is a softmax activation layer, and Concat is the feature concatenation operator.

5. The multi-period outcome prediction method based on multi-class head concatenation and inter-period fusion loss according to claim 4, characterized in that: The third classification head is used to classify high-dimensional image features. The intermediate classification features and the intermediate classification features Compositional fusion characteristics Given the input, output the long-term outcome probability; The structural expression of the third classification head is: In the formula, This represents the long-term regress probability output by the third classification head. , These are the first and second linear layers in the classification head, respectively. This is a softmax activation layer, and Concat is the feature concatenation operator.

6. The multi-period outcome prediction method based on multi-class head concatenation and inter-period fusion loss according to claim 5, characterized in that: The fusion loss function is: In the formula, The total loss of the multi-period outcome prediction model. For short-term sub-fusion loss, For intermediate sub-fusion loss, For long-term sub-fusion loss; In the formula, i represents the corresponding number for short-term, medium-term, and long-term. , where j is the corresponding number of the classification head used for short-term, medium-term, and long-term outcome probability prediction. FocalLoss is the prediction result of the gold standard for the i-th period on the j-th class head. In the formula, For the gold standard of the i-th period, Let be the positive probability output by the j-th classification header. Let be the negative probability output by the j-th classification header. An adjustment factor to balance the contribution of difficult and easy samples to the loss; Among them, when This indicates that the positive result has not turned into a positive outcome. This indicates that the negative result has been confirmed. when hour, middle Used to constrain the loss of positive samples predicted by the current classification head. Used to constrain the loss of negative samples predicted by the current classification head; when hour, The gold standard used to constrain the positive probability output by the previous classification head using the current period's positive samples; when hour, The negative probability of the classification head output is used to constrain the negative sample output in the later period using the gold standard of the current period's negative samples.

7. The multi-period outcome prediction method based on multi-class head concatenation and inter-period fusion loss according to claim 6, characterized in that: The Adam optimizer undergoes end-to-end iterative training, dynamically calculating the short-term, medium-term, and long-term total loss during the training process. Then, backpropagation is performed until the multi-period outcome prediction model achieves the best accuracy on the validation set, thus obtaining the optimal multi-period outcome prediction model.

Images