Cancer progression assessment method and system therefor

Abstract

The invention relates to a cancer progression evaluation method and system, and the evaluation method comprises the following steps: inputting a plurality of immune cell population data through an input device, and storing the immune cell population data in a storage device; accessing the storage device by the processor, calculating immune entropy values of the plurality of immune cell population data, normalizing the immune entropy values, and establishing a cancer progression evaluation model by using the normalized immune entropy values through a decision tree algorithm; obtaining immune cell population data to be evaluated through an input device, and performing an interpretation program by a processor to obtain a cancer progress interpretation result; and outputting the cancer progress interpretation result by means of an output device. The system comprises an input device, a storage device, a processor and output equipment.

Description

Technical Field

[0001] The present invention relates to a cancer progression assessment method and system thereof, and more particularly to an assessment method and system thereof that utilizes a decision tree algorithm to analyze immune cell population data obtained by flow cytometry analysis to accurately predict cancer progression. Background Art

[0002] Due to changes in modern diet and lifestyle habits, as well as factors such as environmental pollution and radiation caused by the rapid development of civilization, modern people are prone to various diseases. In severe cases, these diseases may cause mutations in human cells, leading to abnormal proliferation of these mutated cells and the formation of tumors. If the disease continues to worsen and forms a malignant tumor, it is called cancer.

[0003] The World Health Organization (WHO) points out that cancer is one of the leading causes of death worldwide, claiming nearly 10 million lives in 2021. In Taiwan, cancer has been ranked first among the top ten causes of death for decades in a row, and due to the rapid aging population and unhealthy lifestyles, the number of cancer cases is expected to continue to rise.

[0004] In recent years, with the rapid development of machine learning and the continuous integration of medicine and artificial intelligence, the use of computer methods to assist in the study of related issues in the medical and biological fields has become a powerful tool.

[0005] Current cancer treatments are primarily categorized as surgical resection, radiotherapy, chemotherapy, targeted therapy, and cell therapy. Cytokine-Induced Killer (CIK) cell therapy is currently the most successfully approved treatment under the Special Management Measures, outperforming these traditional cancer therapies. However, the global success rate remains at only 10-30%, and there are no predictive systems on the market to evaluate the efficacy of cell therapy.

[0006] In summary, there is an urgent need for effective assessment methods to intervene in the clinical practice of cancer treatment, providing physicians with additional patient immune-related data as a reference to accelerate medical decision-making, increase patient treatment willingness, and reduce unnecessary waste of resources. Summary of the Invention

[0007] In view of the above-mentioned problems in conventional cancer treatment assessment, the present invention aims to provide a cancer progression assessment method and system thereof to reduce the problems of misjudgment caused by manual assessment and difficulty in making quick medical decisions.

[0008] According to one purpose of the present invention, a method for assessing cancer progression is proposed, which includes the following steps: inputting multiple immune cell population data through an input device and storing them in a storage device; accessing the storage device through a processor to calculate the immune entropy values ​​of the multiple immune cell population data, and normalizing the immune entropy values ​​of the multiple immune cell population data, and then calculating the immune entropy values ​​of the normalized multiple immune cell population data using a decision tree algorithm to establish a cancer progression assessment model; obtaining the immune cell population data to be assessed through the input device, performing a judgment procedure with the processor to obtain a cancer progression judgment result; accessing the storage device through an output device to output the cancer progression judgment result.

[0009] The immune cell population data are obtained by collecting peripheral blood samples from cancer patients undergoing CIK treatment before entering the treatment course, and analyzing the peripheral blood samples using flow cytometry to obtain the immune cell population data of the cancer patients.

[0010] Regularization can improve model training speed and accuracy. Based on the characteristics of immune cell population data and experimental comparisons, Min-Max Normalization is used to scale the immune entropy values ​​of multiple immune cell population data sets between [0, 1] without changing the original data distribution. Among them, immune cell population data refers to the proportion of different cell types in the entire cell population.

[0011] The cancer progression results are classified according to the Response Evaluation Criteria in Solid Tumors (RECIST) criteria, which are divided into four stages based on changes in tumor size during cancer treatment: complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD).

[0012] The plurality of immune cell population data specifically includes 82 immune cell population data, and thus has 82 feature points for machine learning. The types of the 82 immune cell population data are shown in Table 1:

[0013] Table 1: List of 82 immune cell population data according to the present disclosure

[0014]

[0015]

[0016]

[0017]

[0018] Before establishing a cancer progression assessment model, peripheral blood samples obtained from different cancer patients were first divided into four categories (i.e., CR, PR, SD, and PD) according to RECIST criteria. Then, the immune entropy values ​​of multiple immune cell population data (i.e., the 82 feature points) in the blood samples of individual cancer patients were calculated and normalized. The immune entropy values ​​of each normalized immune cell population data were then used as indicators for training using a decision tree algorithm. Through supervised learning, the normalized immune entropy values ​​of the multiple immune cell population data were divided into training, validation, and testing groups, and the model was continuously trained in a loop. When the test group achieved the best accuracy, the key feature points (i.e., key immune cell population data) of the optimal decision tree were identified to obtain the cancer progression assessment model.

[0019] Key feature points were selected using the decision tree's classification nodes as weights, and their contribution and importance were verified using SHAP. SHAP (SHapley Additive exPlanation) is a Python program used to interpret model predictions. By analyzing each feature point and calculating its Shapley value, the contribution and importance of each feature point to the prediction are assessed.

[0020] Finally, the data of the immune cell population to be evaluated is evaluated using the cancer progression assessment model to obtain a cancer progression interpretation result. From the cancer progression interpretation result, it can be known whether the current cancer progression interpretation result of the cancer patient to be evaluated is in one of the four stages: CR, PR, SD or PD.

[0021] According to another object of the present invention, a cancer progression assessment system is proposed, which includes an input device, a storage device, a processor, and an output device. The input device is used to input a plurality of immune cell group data and immune cell group data to be assessed; the storage device is connected to the input device and is used to store a plurality of immune cell group data and immune cell group data to be assessed; the output device is connected to the storage device and is used to output a cancer progression judgment result; the processor is connected to the storage device and executes a plurality of instructions to perform the following steps: calculating immune entropy values ​​of a plurality of immune cell group data and normalizing the immune entropy values ​​of the plurality of immune cell group data; calculating the immune entropy values ​​of the normalized plurality of immune cell group data using a decision tree algorithm to establish a cancer progression assessment model; performing a judgment procedure on the immune cell group data to be assessed based on the cancer progression assessment model to obtain a cancer progression judgment result; and accessing the storage device via the output device to output the cancer progression judgment result.

[0022] As described above, the cancer progression assessment method and system of the present invention can quickly and accurately determine the current stage of cancer progression in cancer patients, thereby reducing the burden on physicians and alleviating the problem of inconsistent judgment standards among different physicians. This facilitates subsequent physicians to accurately assess the cancer patient's condition and make appropriate medical decisions as quickly as possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] To make the technical features, contents, advantages and effects of the present invention more apparent, the present invention is described in detail below in the form of embodiments with reference to the accompanying drawings.

[0024] Figure 1 is a flow chart of a method for assessing cancer progression according to an embodiment of the present invention;

[0025] Figure 2 is a schematic diagram of an optimal decision tree architecture according to an embodiment of the present invention;

[0026] Figure 3 Schematic diagram of the contribution and importance of key feature points of the optimal decision tree architecture using SHAP to verify the embodiment of the present invention, where X0 represents Tc, X1 represents NK, and X2 represents Inter.monocyte;

[0027] Figure 4 Schematic diagram of the accuracy of different algorithms in evaluating cancer progression;

[0028] Figure 5 FIG. 4 is a schematic diagram of a cancer progression assessment system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0029] To help the Review Committee understand the technical features, content, advantages, and effects achievable by the present invention, the present invention is described in detail below using embodiments in conjunction with the accompanying drawings. The figures used herein are for illustrative purposes only and to assist in the description. They may not reflect the actual proportions and precise configurations of the present invention after implementation. Therefore, the proportions and configurations in the accompanying drawings should not be interpreted to limit the scope of the present invention in actual implementation. This is to be noted.

[0030] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the meanings commonly understood by one of ordinary skill in the art to which the invention belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present invention, and will not be interpreted as idealized or overly formal meanings unless explicitly defined as such herein.

[0031] See also Figure 1, which is a flow chart of a method for evaluating cancer progression according to an embodiment of the present invention. Figure 1 As shown, the cancer progression assessment method includes the following steps (S1-S4):

[0032] Step S1: Input a plurality of immune cell population data through an input device and store them in a storage device.

[0033] The collected immune cell population data is input into the system's storage device via an input device. The input device described here is not limited to a flow cytometer for obtaining immune cell population data. Immune cell population data stored in a medical institution's database can also be input into the system database via physical lines, file transmission from a storage device, or via wired or wireless network transmission to serve as training data for model construction.

[0034] Among them, multiple immune cell population data were obtained by collecting peripheral blood samples from a single cancer patient undergoing CIK treatment before entering the treatment course, and then analyzing the peripheral blood samples using a flow cytometer; therefore, each cancer patient's peripheral blood sample has multiple immune cell population data, specifically 82 immune cell population data.

[0035] In addition, before establishing the cancer progression assessment model, peripheral blood samples obtained from different cancer patients were first divided into four categories (i.e., CR, PR, SD, and PD) according to RECIST criteria.

[0036] Step S2: The processor accesses the storage device to calculate the immune entropy values ​​of the plurality of immune cell population data, and normalizes the immune entropy values ​​of the plurality of immune cell population data. The normalized immune entropy values ​​of the plurality of immune cell population data are then calculated using a decision tree algorithm to establish a cancer progression assessment model.

[0037] The processor reads the data of multiple immune cell populations stored in the storage device, first calculates the immune entropy value (Immunity Entropy) of the multiple immune cell population data (i.e., the 82 immune cell population data, as shown in Table 1 above) in the blood sample of an individual cancer patient, and then uses the minimum-maximum method to normalize the immune entropy value of the immune cell population data. The immune entropy values ​​of the multiple immune cell population data are proportionally scaled between [0, 1]. After normalization, the model training speed and accuracy can be improved. The immune entropy value is calculated as shown in the following formula:

[0038]

[0039] In the above formula, P represents probability, and B1, B2, ... through Bn represent the first, second, ... through nth sample values ​​of the B cell population, respectively. The same applies to the remaining cell populations and is not further described here. The above formula calculates a total of 82 immune entropy values ​​for each immune cell population data.

[0040] Next, the immune entropy values ​​of each normalized immune cell population data (82 in total, equivalent to 82 feature points) were used as indicators for training using a decision tree algorithm, and the model was continuously trained in a loop. When the test group achieved the best accuracy, the key feature points of the optimal decision tree (i.e., key immune cell population data) were identified to obtain a cancer progression assessment model.

[0041] Among them, the immune entropy values ​​of the normalized immune cell population data were divided into training group, validation group and test group, with the distribution ratio being 64% for the training group, 16% for the validation group and 20% for the test group, that is, (training group + validation group): test group = 8:2, and training group: test group = 8:2.

[0042] See also Figure 2 , which is a schematic diagram of the optimal decision tree architecture of an embodiment of the present invention. Figure 2 As shown in the figure, the key feature points of the best decision tree are selected by taking the classification nodes of the decision tree as weights, and three important key feature points are selected, namely CD3+CD8+Tc (CD3+CD8+ cytotoxic T cells), CD3 - CD56+CD16+NK (CD3-CD56+CD16+ natural killer cells) and CD14 + +CD16+Inter.Monocyte (CD14++CD16+intermediate monocytes), and use SHAP to verify the contribution and importance of these key feature points.

[0043] Please refer to Figure 3 , which is a schematic diagram of the contribution and importance of key feature points of the optimal decision tree architecture of the embodiment of the present invention using SHAP, where X0 represents Tc, X1 represents NK, and X2 represents Inter.monocyte. Figure 3 As shown, the contributions of Tc, NK, and Intermonocyte at each node vary significantly. Intermonocyte is most important for classifying cancer patients in the PR stage, NK is most important for classifying cancer patients in the SD stage, and Tc is most important for classifying cancer patients in the CR and PD stages. This shows that these key features are important factors in determining the four stages of CR, PR, SD, or PD.

[0044] Step S3: Obtain the data of the immune cell population to be evaluated through the input device, and use the processor to perform a judgment program to obtain a cancer progression judgment result.

[0045] The cancer progression assessment model established in step S2 is used to evaluate the immune cell population data to obtain a cancer progression interpretation result. The cancer progression interpretation result indicates whether the patient's cancer is currently at one of four stages: CR, PR, SD, or PD. The input device described here is the same as previously described and will not be further described here.

[0046] See also Figure 4 , which is a schematic diagram of the accuracy of different algorithms in assessing cancer progression. Figure 4 As shown, multiple immune cell population data (i.e., the 82 immune cell population data) were used to establish a cancer progression assessment model using other machine learning algorithms such as support vector machine (SVM) and K-nearest neighbor algorithm (KNN). Their accuracy (correctness) was much lower than that of the decision tree algorithm. Among them, the accuracy of SVM was only 42.85%, the accuracy of KNN was slightly higher at 64.28%, and the accuracy of the decision tree was as high as 85.71%. Therefore, it can be seen that the cancer progression assessment model established using the decision tree algorithm is superior to other types of algorithms.

[0047] Step S4: Accessing the storage device via the output device to output the cancer progression assessment result.

[0048] The cancer progression judgment result obtained in step S3 can be further outputted via an output device. The output device disclosed in this embodiment can include various display interfaces, such as a computer screen, a display, or a handheld device display.

[0049] See also Figure 5 , which is a schematic diagram of a cancer progression assessment system according to an embodiment of the present invention. Figure 5 As shown, the cancer progression assessment system 20 may include an input device 21 , a storage device 22 , a processor 23 , and an output device 24 .

[0050] In this embodiment, input device 21 is a flow cytometer, which collects and analyzes peripheral blood samples from cancer patients undergoing CIK therapy before treatment to obtain immune cell population data. In another embodiment, input device 21 is not limited to a flow cytometer. Input device 21 may include an input interface of an electronic device such as a personal computer, smartphone, or server, including a touch screen, keyboard, or mouse, to transmit immune cell population data via a file format. Alternatively, historical data may be uploaded to the memory of storage device 22 via wireless network transmission, wireless communication, or conventional wired internet access. The memory may include read-only memory, flash memory, disk, or a cloud database.

[0051] Next, the cancer progression assessment system 20 accesses the storage device 22 through the processor 23. The processor 23 may include a central processing unit, an image processor, a microprocessor, etc. in a computer or server. It may include a multi-core processing unit or a combination of multiple processing units. The processor 23 executes instructions to access the multiple immune cell group data in the storage device 22 for a training program, and accesses the immune cell group data to be evaluated for a reading program. In detail, the training program is to normalize the multiple immune cell group data originally in the storage device 22 by calculating the immune entropy values ​​of the multiple immune cell group data, and then calculate the immune entropy values ​​of the normalized multiple immune cell group data using a decision tree algorithm to establish a cancer progression assessment model.

[0052] Next, the immune cell population data to be evaluated is calculated by the established cancer progression evaluation model through the interpretation program, and is classified as one of CR, PR, SD, and PD according to the RECIST criteria to obtain a cancer progression interpretation result. The output device 24 accesses the storage device 22 to output the cancer progression interpretation result. The output device 24 can include various display interfaces, such as a computer screen, a monitor, or a handheld device display.

[0053] The above-mentioned cancer progression assessment method and system can significantly reduce the workload of physicians and reduce errors in manual interpretation that may lead to deviations in cancer progression diagnosis. Furthermore, using the cancer progression assessment method and system disclosed herein, a cancer patient's current cancer progression stage can be quickly and accurately assessed with an assessment accuracy of over 80%, thereby facilitating subsequent physicians to correctly assess the cancer patient's condition and make appropriate medical decisions as quickly as possible.

[0054] The above description is for illustrative purposes only and is not intended to be limiting. Any equivalent modifications or variations that do not depart from the spirit and scope of the present invention should be included in the scope of the appended patent applications.

Claims

1. A method for assessing cancer progression, comprising the following steps: Step S1: inputting a plurality of immune cell population data through an input device and storing the data in a storage device; Step S2: Calculating, by a processor, immune entropy values ​​of the plurality of immune cell population data by accessing the storage device, and normalizing the immune entropy values ​​of the plurality of immune cell population data. Then, computing the normalized immune entropy values ​​of the plurality of immune cell population data using a decision tree algorithm to establish a cancer progression assessment model. Step S3: obtaining data of an immune cell population to be evaluated through the input device, and performing a judgment process on the processor to obtain a cancer progression judgment result; and Step S4: Accessing the storage device through an output device to output the cancer progression judgment result, wherein In step S1, the immune cell population data are first classified into four categories according to the solid tumor response evaluation criteria (RECIST criteria): complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD).

2. The cancer progression assessment method according to claim 1, wherein the input device is a flow cytometer. 3 . The cancer progression assessment method according to claim 1 , wherein the plurality of immune cell population data are specifically 82 immune cell population data. 4 . The cancer progression assessment method according to claim 1 , wherein the immune entropy values ​​of the plurality of immune cell population data are normalized using a minimum-maximum method.

5. The method for assessing cancer progression according to any one of claims 1 to 3, wherein the normalized immune entropy values ​​of the plurality of immune cell population data are divided into a training group, a validation group, and a test group, and the distribution ratios are 64% for the training group, 16% for the validation group, and 20% for the test group. 6 . The cancer progression assessment method according to claim 1 , wherein the cancer progression assessment model comprises an optimal decision tree having three key immune cell populations. 7 . The method for assessing cancer progression according to claim 6 , wherein the three key immune cell populations are cytotoxic T cells (Tc), natural killer (NK) cells, and intermediate monocytes.

8. The method for evaluating cancer progression according to claim 7, wherein Cytotoxic T cells (Tc) are most important for classifying complete response (CR) and progressive disease (PD); Natural killer (NK) cells are most important for classifying stable disease (SD); and Intermediate monocytes are most important for classifying partial response (PR).

9. A cancer progression assessment system comprising: An input device for inputting a plurality of immune cell population data and data of an immune cell population to be evaluated; a storage device connected to the input device, storing the data of the plurality of immune cell populations and the data of the immune cell population to be evaluated; an output device, connected to the storage device, for outputting a cancer progression assessment result; as well as A processor, connected to the storage device, executes a plurality of commands to perform the following steps: Calculating immune entropy values ​​of the plurality of immune cell population data and normalizing the immune entropy values ​​of the plurality of immune cell population data; Calculating the normalized immune entropy values ​​of the plurality of immune cell population data using a decision tree algorithm to establish a cancer progression assessment model; Performing an interpretation process on the immune cell population data to be evaluated according to the cancer progression assessment model to obtain a cancer progression interpretation result; and The storage device is accessed through the output device to output the cancer progression judgment result.

10. The cancer progression assessment system as claimed in claim 9, wherein the input device is a flow cytometer.

Images