Cancer TNM intelligent staging system and device and storage medium

Abstract

The invention discloses a cancer TNM intelligent staging system and device and a storage medium, the cancer TNM intelligent staging system comprises an information acquisition module, an identification module and a typing output module, the information acquisition module is used for acquiring cancer diagnosis information; the identification module is used for identifying typing selection and staging selection corresponding to the diagnosis information; wherein the typing selection and the staging selection are both determined based on the diagnosis information and a preset rule base; and the typing output module is used for outputting a cancer TNM staging result corresponding to the diagnosis information according to the typing selection and the staging selection. The cancer TNM staging process is managed in a flow mode, staging errors caused by human factors such as mistaken and missed information are reduced, and therefore the staging accuracy is improved to a certain degree.

Description

Technical Field

[0001] This invention relates to the field of intelligent healthcare, and more particularly to a cancer TNM intelligent staging system, device, and storage medium. Background Technology

[0002] The TNM (Tumor, Node, Metastasis) staging system for cancer is a widely accepted method in the medical community for assessing the severity and extent of cancer spread. It describes the cancer's condition using three key indicators: T represents the size and depth of the tumor, N indicates whether the tumor has spread to nearby lymph nodes, and M indicates whether the cancer has metastasized to other parts of the body. This staging system provides physicians with a standardized tool to help them assess cancer patients' conditions and develop personalized treatment plans, while also providing patients with information about disease prognosis.

[0003] However, existing cancer TNM staging studies mostly focus on the management of clinical data, while lacking necessary oversight of the staging process, which is detrimental to the accuracy of cancer TNM staging results. Summary of the Invention

[0004] The purpose of this invention is to provide a cancer TNM intelligent staging system, device, and storage medium to solve the problems mentioned in the background art.

[0005] To achieve the above-mentioned objectives, one aspect of the present invention provides a cancer TNM intelligent staging system, comprising an information acquisition module, an identification module, and a typing output module, wherein:

[0006] The information acquisition module is used to obtain cancer diagnosis information;

[0007] The identification module is used to identify the subtyping selection and staging selection corresponding to the diagnostic information; wherein, the subtyping selection and the staging selection are both determined based on the diagnostic information and a preset rule base;

[0008] The typing output module is used to output the cancer TNM staging result corresponding to the diagnostic information based on the typing selection and the staging selection.

[0009] Furthermore, the diagnostic information is a description of the patient's disease, including medical records, disease name, pathological classification, examination results, patient identity, attending physician, and examining physician, which is entered by the doctor.

[0010] Furthermore, the preset rule base includes a classification rule base and a staging rule base; the classification rule base includes a classification of diagnosis and treatment stages and corresponding classification standards; the staging rule base includes several disease codes and staging standards, each disease code corresponding to several staging standards, and the staging standards include the progression of the primary tumor, regional lymph nodes, and distant metastases.

[0011] Furthermore, the classification rule base includes diagnostic and treatment stage classifications such as pTNM, cTNM, ypTNM, and rTNM.

[0012] Furthermore, the diagnostic information is matched with a preset rule base to obtain subtyping and staging predictions based on the matching results; the subtyping selection is compared with the subtyping prediction, and the staging selection is compared with the staging prediction; the cancer TNM staging results are output based on the comparison results.

[0013] Furthermore, the system matches medical records with the classification rule base to obtain classification predictions; it matches disease names with disease codes; if a match is successful, it matches the medical records corresponding to the disease names with the staging standards corresponding to the successfully matched disease codes; if a match is successful, it obtains staging predictions.

[0014] Furthermore, if the typing selection is consistent with the typing prediction, and the staging selection is consistent with the staging prediction, then the cancer TNM staging result is output based on the typing selection and staging selection; if the typing selection is inconsistent with the typing prediction, then the first critical conflict point in the medical record is located based on the classification criteria corresponding to the typing selection and the classification criteria corresponding to the typing prediction; if the staging selection is inconsistent with the staging prediction, then the second critical conflict point in the medical record is located based on the staging criteria corresponding to the staging selection and the staging criteria corresponding to the staging prediction; based on the first and second critical conflict points, corresponding risk warnings are issued.

[0015] Furthermore, several types of medical images are acquired, and combined with medical records, cross-modal target recognition is performed to obtain tumor location information, tumor size information, and tumor invasion depth information; the tumor location information, tumor size information, and tumor invasion depth information are combined to perform staging standard matching and obtain staging prediction.

[0016] A second aspect of the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the operating functions of the above-described system.

[0017] A third aspect of the present invention provides a computer-storable medium having a computer program stored thereon, which, when executed by a processor, implements the operational functions of the above-described system.

[0018] Compared with existing technologies, this system and method have the following advantages:

[0019] 1. This invention first obtains diagnostic information to ensure the necessary patient information is verified. Then, based on the diagnostic information and a preset rule base, it performs subtyping and staging selection to ensure consistency in the standards used by doctors when staging and subtyping. Finally, only after the subtyping and staging selection steps are completed can the cancer TNM staging result be output, ensuring the integrity of the doctor's operation process.

[0020] 2. This invention improves the personalization and precision of cancer treatment by standardizing process management and incorporating predictive models, which helps to improve patients' survival rate and quality of life.

[0021] 3. The method of automatically matching cancer TNM staging results based on staging selection and typing selection in this invention can reduce the workload of medical professionals and improve the overall efficiency of medical services.

[0022] 4. This invention can reduce misdiagnosis and optimize treatment processes by improving the accuracy of cancer staging, thereby reducing medical costs and bringing economic savings to medical institutions. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the structure of a cancer TNM intelligent staging system.

[0024] Figure 2 This is a diagram illustrating the application environment of a cancer TNM intelligent staging system.

[0025] Figure 3 A flowchart illustrating how to use the TNM intelligent staging system for cancer. Detailed Implementation

[0026] 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.

[0027] like Figure 1 The diagram shown illustrates the system structure of this invention. This invention provides a TNM intelligent staging system for cancer. It includes: an information acquisition module 402, an identification module 404, and a typing output module 406.

[0028] The cancer TNM intelligent staging system provided in this invention can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on other network servers. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, etc. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0029] The information acquisition module is used to obtain diagnostic information.

[0030] The diagnostic information mainly describes the patient's disease, such as the disease name, pathological classification, and examination results, and also includes basic information such as the patient's identity, attending physician, and examining physician.

[0031] Diagnostic information is entered by doctors and can include all examination information and diagnostic records related to the patient's current visit, as well as historical medical records related to the disease being treated, such as the previous medical records of patients returning for follow-up visits. Comprehensive and continuous diagnostic information is beneficial for a more complete understanding of disease changes.

[0032] Basic information in the diagnosis can help verify the patient's identity, avoid human error, and ensure the orderly conduct of the medical process.

[0033] In this embodiment, to facilitate management and subsequent work, it is necessary to standardize the description of disease names. Therefore, a selection mode is used to select disease names instead of manual input. Furthermore, the selection of disease names adopts a progressively more detailed approach. For example, by selecting "cancer," "lung cancer," "malignant tumor of the main bronchus," and "malignant tumor of the left main bronchus" in sequence, the disease name entry is completed.

[0034] The identification module is used to identify the subtyping and staging selections corresponding to the diagnostic information; both subtyping and staging selections are determined based on the diagnostic information and a preset rule base.

[0035] The default rule base is the TNM staging rule base, which defines various criteria for determining cancer staging. This step requires doctors to manually determine the subtype and staging selection based on the criteria in the default rule base.

[0036] In this embodiment, to facilitate more accurate diagnosis by doctors, the system automatically retrieves a preset rule base and displays the definitions of each subtype and stage on the screen when doctors select subtypes and stages. For example, the definition of T1 lung cancer is a tumor with a maximum diameter of 3 cm or less, surrounded by the lung or visceral pleura, and not invading the upper lobe bronchi (not within the main bronchus) on bronchoscopy. Doctors can independently select the subtype and stage based on the definitions of subtype and stage.

[0037] The typing output module outputs the cancer TNM staging results corresponding to the diagnostic information based on the typing and staging selections.

[0038] After the doctor completes the diagnostic information, classification selection, and staging selection, the system matches the preset rule base to finally determine the patient's cancer TNM staging result. For example, after the doctor selects T1 (tumor maximum diameter less than or equal to 3cm, surrounded by the lung or visceral pleura, and bronchoscopy shows no invasion of the upper lobe bronchus (not in the main bronchus)), N1 (with ipsilateral bronchial and / or ipsilateral hilar lymph node and intrapulmonary lymph node metastasis, including through direct invasion), and M0 (no distant metastasis), the system will output the patient's cancer TNM staging result as III A1.

[0039] In one embodiment, the preset rule base includes a typing rule base and a staging rule base; the typing rule base includes a classification of treatment stages and corresponding classification criteria; the staging rule base includes several disease codes and staging criteria, each disease code corresponding to several staging criteria, and the staging criteria include the progression of the primary tumor, regional lymph nodes, and distant metastases.

[0040] The classification rule base includes four categories for diagnosis and treatment stages: pTNM, cTNM, ypTNM, and rTNM. The corresponding classification criteria are shown in Table 1.

[0041] Table 1 Cancer TNM Classification Rule Base

[0042] TNM classification Type Description pTNM Pathological staging is applicable to patients who have undergone surgical treatment. It is a staging process determined by a pathologist after a detailed examination of the removed tumor tissue. cTNM Clinical staging is applicable to untreated patients and assesses cancer stage through clinical examination, imaging studies, and other non-invasive diagnostic methods. ypTNM Clinical staging applies to patients who have received neoadjuvant therapy (such as chemotherapy or radiotherapy). Neoadjuvant therapy refers to treatment administered before primary treatment (usually surgery). The prefix "y" indicates the post-treatment status, while "p" indicates that the staging is based on pathological assessment. rTNM Recurrence staging is used for patients with recurrent cancer to describe the stage of cancer recurrence. "r" represents recurrence. This staging system assesses the severity of cancer when it recurs after a period of remission.

[0043] The staging rule base covers cancers including thyroid cancer, intrahepatic bile duct cancer, cervical cancer, lung cancer, liver cancer, breast cancer, pancreatic cancer, colorectal cancer, stomach cancer, kidney cancer, esophageal cancer, and prostate cancer. The rules for each cancer include disease codes and TNM staging standards. The rules for thyroid cancer, intrahepatic bile duct cancer, cervical cancer, and lung cancer are listed below; other rules will not be elaborated upon.

[0044] Table 2 Thyroid Cancer Disease Codes

[0045] Disease coding Disease Name C73.x00 Intrahepatic cholangiocarcinoma C73.x00x003 Bile duct cancer

[0046] Table 3. Thyroid Cancer Classification

[0047] Disease coding Disease Name Differentiated thyroid cancer Papillary carcinoma, follicular carcinoma, poorly differentiated carcinoma, Hütter cell carcinoma Undifferentiated thyroid carcinoma Undifferentiated thyroid carcinoma medullary carcinoma medullary carcinoma

[0048] Table 4 TNM staging criteria for thyroid cancer

[0049] Primary tumor (T) standard TX Primary tumor cannot be assessed T0 No evidence of primary tumor T1 The tumor is confined to the thyroid gland and has a maximum diameter of ≤2cm. T1a The tumor is confined to the thyroid gland and has a maximum diameter of ≤1cm. T1b The tumor is confined to the thyroid gland, and its maximum diameter is less than 1 cm and less than or equal to 2 cm. T2 The tumor is confined to the thyroid gland, with a maximum diameter >2cm but ≤4cm. T3 Tumors confined to the thyroid gland with a maximum diameter >4 cm, or any tumor with significant extrathyroidal invasion involving the band-like muscle. T3a The tumor is confined to the thyroid gland and has a maximum diameter >4cm. T3b Any tumor with significant extraglandular invasion involving the band muscles (including the sternohyoid, sternothyroid, thyrohyoid, and omohyoid muscles) T4 Significant extraglandular invasion beyond the bandgap muscle T4a Any tumor that invades subcutaneous soft tissue, larynx, trachea, esophagus, or recurrent laryngeal nerve T4b Any tumor that invades the prevertebral fascia, surrounds the carotid artery, or mediastinal vessels Regional lymph nodes (N) standard NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N0a With ≥1 lymph node that is benign according to cytological or pathological examination N0b There was no radiological or clinical evidence to suggest regional lymph node metastasis. N1 With regional lymph node metastasis N1a Metastasis to zone VI (pretracheal, paratracheal, prelarynx / Delphian lymph nodes) or zone VII (superior mediastinal lymph nodes), including unilateral and bilateral metastases. N1b Metastasis to lateral cervical (regions I, II, III, IV, or V) lymph nodes or retropharyngeal lymph nodes, including unilateral, bilateral, and contralateral metastases. Distant transfer (M) standard M0 No distant transfer M1 Accompanied by distant migration

[0050] Table 5 TNM staging grouping of differentiated thyroid cancer

[0051] Age at diagnosis T N M Phased grouping <55 years old Any T Any N M0 I <55 years old Any T Any N M1 II ≥55 years old T1 N0 / NX M0 I ≥55 years old T1 N1 M0 II ≥55 years old T2 N0 / NX M0 I ≥55 years old T2 N1 M0 II ≥55 years old T3a / T3b Any N M0 II ≥55 years old T4a Any N M0 III ≥55 years old T4b Any N M0 IVA ≥55 years old Any T Any N M1 IVB

[0052] Table 6 TNM staging grouping of differentiated thyroid cancer

[0053] T N M Phased grouping T1-T3a N0 / NX M0 IVA T1-T3a N1 M0 IVB T3b Any N M0 IVB T4 Any N M0 IVB Any T Any N M1 IVC

[0054] Table 7 TNM staging grouping of medullary thyroid carcinoma

[0055] T N M Phased grouping T1 N0 M0 I T2 N0 M0 II T3 N0 M0 II T1-3 N1a M0 III T4a Any N M0 IVA T1-3 N1b M0 IVA T4b Any N M0 IVB Any T Any N M1 IVC

[0056] Table 8. Disease Codes for Hepatobiliary Cholangiocarcinoma

[0057] Disease coding Disease Name C22.100 Intrahepatic cholangiocarcinoma C22.101 Bile duct cancer

[0058] Table 9 TNM staging criteria for hepatobiliary carcinoma

[0059] Primary tumor (T) standard TX Primary tumor cannot be assessed T0 No evidence of primary tumor Tis Carcinoma in situ (intra-biliary tumor) T1 Solitary tumor without vascular invasion, ≤5cm or >5cm T1a Solitary tumor ≤5cm, without vascular invasion T1b A single tumor >5cm in size, without vascular invasion. T2 A single tumor with intrahepatic vascular invasion, or multiple tumors with or without vascular invasion. T3 Tumor penetrates visceral peritoneum T4 Tumor directly invades local extrahepatic structures Regional lymph nodes (N) standard NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 With regional lymph node metastasis Distant transfer (M) standard M0 No distant transfer M1 Accompanied by distant migration

[0060] Table 10 TNM staging grouping of hepatobiliary carcinoma

[0061] T N M Phased grouping Tis N0 M0 0 T1a N0 M0 IA T1b N0 M0 IB T2 N0 M0 II T3 N0 M0 IIIA T4 N0 M0 IIIB Any T N1 M0 IIIB Any T Any N M1 IV

[0062] Table 11 Cervical Cancer Disease Code

[0063] coding name C53.000 Cervical endometrial malignancy C53.000x001 Malignant tumors of the cervical endometrium C53.100 external cervical malignant tumors C53.800 Cervical entanglement transcends the damage caused by malignant tumors. C53.801 Malignant tumor of cervical stump C53.900 Cervical cancer C53.900x001 Cervical cancer

[0064] Table 12 TNM staging criteria for cervical cancer

[0065] Primary tumor (T) standard T0 No evidence of primary tumor Tis Carcinoma in situ T1 Cervical cancer confined to the cervix (invasion of the uterine body will not be considered). T1a Microscopically, invasive carcinoma with a maximum invasive depth of ≤5.0 mm for all visible lesions. T1a1 Interstitial infiltration depth ≤3mm T1a2 Interstitial infiltration depth >3mm and ≤5mm T1b Grossly visible cancerous lesions confined to the cervix; depth of invasion >5mm (microscopic lesions larger than stage IA); involvement of lymphatic / vascular spaces does not alter tumor stage, and the extent of tumor invasion is not considered. T1b1 The largest visible cancerous lesion is ≤2cm in diameter and has an invasion depth >5mm. T1b2 The largest visible cancerous lesion is >2cm in diameter and ≤4cm in size. T1b3 The largest visible cancerous lesion is >4cm in diameter. T2 The cervical tumor has extended beyond the uterus, but has not reached the pelvic wall or the lower third of the vagina. T2a The tumor involved the upper two-thirds of the vagina, without parametrial invasion. T2a1 The largest diameter of the lesion visible to the naked eye is ≤4cm T2a2 The largest visible lesion diameter is >4cm T2b There was some seepage around the uterus, but it did not reach the basin wall. T3 The tumor involves the lower third of the vagina and / or extends to the pelvic wall, and / or causes hydronephrosis or renal insufficiency, and / or involves the pelvic and / or para-aortic lymph nodes. (The pelvic wall refers to the muscles, fascia, neurovascular structures, and bony components of the pelvis; obliteration of the space between the pelvic wall and the tumor during rectal examination indicates FIGO stage III.) T3a The tumor involved the lower third of the vagina but did not invade the pelvic wall. T3b Tumors involving the pelvic wall and / or causing hydronephrosis or no hydronephrosis T4 The tumor invades the bladder or rectal mucosa (confirmed by biopsy) or invades adjacent organs (vesicular edema is not classified as IVA). Regional lymph nodes (N) standard Nx Regional lymph node metastasis cannot be assessed N0 No regional lymph node metastasis N0(i+) Regional lymph node tumor cells (ITCs): tumor lesions within the lymph node with a diameter <0.2 cm, or a single tumor cell within a single lymph node, or clusters of <200 cells. N1 Regional lymph node metastasis, limited to pelvic lymph nodes N1mi Pelvic lymph node metastasis, 0.2mm < maximum diameter ≤ 2mm N1a Pelvic lymph node metastasis, with a maximum diameter >2mm N2 Regional lymph node metastasis: Para-aortic lymph nodes N2mi Para-aortic lymph node metastasis, 0.2mm < maximum diameter ≤ 2mm N2a Para-aortic lymph node metastasis, with a maximum diameter >2mm Distant transfer (M) standard M0 No distant transfer M1 Distant metastasis (including metastasis to inguinal lymph nodes, peritoneum, lungs, liver, bones, etc., but excluding metastasis to pelvic and para-aortic lymph nodes and vagina)

[0066] Table 13 TNM staging grouping for cervical cancer

[0067] T N M Phased grouping T1a1 0 0 IA1 T1a2 0 0 IA2 T1b1 0 0 IB1 T1b2 0 0 IB2 T1b3 0 0 IB3 T2a1 0 0 IIA1 T2a2 0 0 IIA2 T2b 0 0 IIB T3a 0 0 IIIA T3b 0 0 IIIB any 1r 0 IIIC1r any 1p 0 IIICp any 2r 0 IIIC2r any 2p 0 IIIC2p any any 1r IVAr / IVBr any any 1p IVAp / IVBp

[0068] Table 14 Lung Cancer Disease Code

[0069] coding name C34.000 Malignant tumor of the main bronchus C34.000x002 Malignant tumor of the left main bronchus C34.000x003 Malignant tumor of the right main bronchus C34.001 malignant tumors of the pulmonary hilum C34.100x003 Malignant tumor in the upper lobe of the left lung C34.100x004 malignant tumor in the upper lobe of the right lung C34.101 Malignant tumor of the upper lobe of the lung C34.102 superior sulcus malignant tumor C34.200x001 malignant tumor in the middle lobe of the right lung C34.201 Malignant tumors of the middle lobe of the lung C34.300x003 Malignant tumor in the lower lobe of the left lung C34.300x004 Malignant tumor in the lower lobe of the right lung C34.301 lower lobe malignant tumor C34.800 The bronchus and lungs intersect to overcome damage from malignant tumors. C34.800x001 Malignant tumor in the middle and upper lobe of the right lung C34.800x002 Malignant tumor in the middle and lower lobe of the right lung C34.800x003 Malignant tumors in the upper and lower lobes of the left lung C34.801 Malignant tumors of the middle and upper lobes of the lung C34.802 Malignant tumors of the middle and lower lobes of the lung C34.803 Malignant tumors of the upper and lower lobes of the lung C34.900x001 Malignant lung tumors C34.900x004 left lung malignant tumor C34.900x005 malignant tumor of the right lung C34.900x006 Malignant tumors of both lungs C34.900x008 Multiple malignant tumors in the lungs C34.901 Malignant tumors of the bronchus C34.902 Malignant tumors of bronchioles

[0070] Table 15 TNM staging criteria for lung cancer

[0071] Primary tumor (T) standard T0 No evidence of primary tumor T1 The tumor's maximum diameter is less than or equal to 3 cm, it is surrounded by the lung or visceral pleura, and bronchoscopy shows that it has not invaded the upper lobe bronchi (not within the main bronchus). T1a Tumors with a maximum diameter of 1 cm or less, uncommon superficial spreading tumors, regardless of size, are classified as T1a if their invasion is limited to the bronchial wall, even if they may invade the main bronchus. T1b The maximum diameter of the tumor is greater than or equal to 1 cm, but less than or equal to 2 cm. T1c The maximum diameter of the tumor is greater than or equal to 2 cm, but less than or equal to 3 cm. T1mi Microinvasive adenocarcinoma; adenocarcinoma (maximum diameter ≥ 3 cm) with predominantly scaly growth, and invasive components with a maximum diameter ≤ 5 mm. T2 A tumor with a maximum diameter greater than 3 cm but less than or equal to 5 cm, or any of the following characteristics: 1. Involvement of the main bronchus, regardless of its distance from the carina, but not involving the carina; 2. Invasion of the visceral pleura (PL1 or PL2); 3. Spread to the hilum with atelectasis or obstructive pneumonia, including partial or total atelectasis / obstructive pneumonia. T2 tumors with these characteristics that are less than or equal to 4 cm or whose size is indeterminate are classified as T2a; those greater than 4 cm but less than or equal to 5 cm are classified as T2b. T2a The tumor's maximum diameter is greater than 3 cm but less than or equal to 4 cm. T2b The tumor's maximum diameter is greater than 4 cm but less than or equal to 5 cm. T3 Tumors with a maximum diameter greater than 5 cm but less than or equal to 7 cm, or those that directly invade any of the following sites: parietal pleura (PL3), chest wall (including superior sulcus tumors), phrenic nerve, pericardium, or metastatic nodules in the same lobe. T4 Tumors with a maximum diameter greater than 7 cm, or tumors of any size invading the following sites: diaphragm, mediastinum, heart, major blood vessels, trachea, recurrent laryngeal nerve, esophagus, vertebral body, carina, or ipsilateral heterolobar metastatic nodules. Tis Carcinoma in situ; squamous cell carcinoma in situ (SCIS); adenocarcinoma in situ (AIS): adenocarcinoma with a pure squamous growth pattern, with a maximum diameter less than or equal to 3 cm. Tx The size of the primary tumor cannot be measured; or cancer cells are found in sputum exfoliated cells or bronchial lavage fluid, but imaging examinations and bronchoscopy fail to detect the primary tumor. Regional lymph nodes (N) standard N0 No regional lymph node metastasis N1 With ipsilateral bronchial and / or ipsilateral hilar lymph node and intrapulmonary lymph node metastasis, including through direct invasion of... N2 With ipsilateral mediastinal lymph node and / or subcarinal lymph node metastasis N3 With metastasis to contralateral mediastinal lymph nodes, contralateral hilum, ipsilateral or contralateral scalene or supraclavicular lymph nodes Nx Regional lymph nodes cannot be evaluated Distant transfer (M) standard M0 No distant transfer M1 There is a long-distance transfer M1 Accompanied by a single metastatic lesion located in a single extrathoracic organ (including a single distant lymph node). M1a Metastatic tumor nodules in the contralateral lung; tumors with metastatic pleural or pericardial nodules; malignant pleural or pericardial effusions. Most pleural (pericardial) effusions are neoplastic, but in a minority of patients, multiple pathological microscopic examinations of the pleural (pericardial) effusions are negative, the fluid is non-bloody and non-exudative, and based on comprehensive clinical judgment, it is not related to the tumor. Such effusions are not used as staging indicators. Mx Distant transfer cannot be evaluated

[0072] Table 16 TNM Staging Groups for Lung Cancer

[0073] TMN combination Installments T#N#M1 IV T#N#M1a IVA T#N#M1b IVB T1aN0M0 IA T1bN0M0 IB T1cN0M0 IC T1N0M0 I T1N1M0 IIIA1 T2aN0M0 IIA T2bN0M0 IIB T2N0M0 II T2N1M0 IIIA1 T3aN0M0 IIIA2 T3aN1M0 IIIA2 T3bN0M0 IIIB T3bN1M0 IIIB T3cN0M0 IIIC T3cN1M0 IIIC

[0074] In the table, T# represents "any T"; N# represents "any N".

[0075] In practice, doctors first complete the subtyping selection, and then perform detailed selection of cancer T, N, and M, that is, complete the staging selection. Finally, the system will output the TNM subtyping results in Table 1 and the staging grouping results in Table 4, that is, complete the output of the cancer TNM staging results.

[0076] In one embodiment, such as Figure 3 The diagram shows a flowchart of TNM staging for cancer. To standardize the doctor's workflow, after the doctor enters the patient's diagnostic information, they must click the "OK" button to complete the diagnostic information collection. Simultaneously, the system extracts the disease name and matches it against the disease code in the preset rule base. If the match fails, the subtyping and staging selection processes are not initiated, and the process ends; if the match is successful, the subtyping and staging selection processes are initiated. After a successful disease name match, subtyping selection is initiated first. The system retrieves the preset rule base and outputs the pTNM, cTNM, ypTNM, and rTNM subtyping selection content, providing detailed definitions for each subtyping and prompting the doctor to complete the subtyping selection. The doctor completes the cancer TNM subtyping selection according to the prompts. After the system recognizes the doctor's completion of the TNM subtyping selection, it prompts the doctor to perform further selections for T, N, and M sub-options. The system provides the names and definitions of each sub-option, and the doctor independently completes the selections for the T, N, and M sub-options, clicks "OK," and the TNM staging information entry is complete. The system recognizes the doctor's selection results for the T, N, and M sub-options. If all three categories (T, N, and M) are entered, the system will prompt that the cancer TNM staging result is about to be output. If not all entries are completed, the system will remind the doctor of the missing information and invite them to re-enter the TNM staging information. Finally, the system collects the doctor's subtype and staging selection information, matches it with a preset rule base, determines the patient's cancer TNM staging result, and provides the doctor with a reminder of the cancer TNM staging result.

[0077] Compared to existing installment standards, this system enables full-process management of installments, resolving issues of information inconsistency and management discrepancies that easily arise in practical applications.

[0078] In one embodiment, diagnostic information is matched with a preset rule base, and subtyping and staging predictions are obtained based on the matching results; subtyping selection is compared with subtyping predictions, staging selection is compared with staging predictions, and cancer TNM staging results are output based on the comparison results.

[0079] This embodiment employs a predictive model to predict diagnostic information, obtaining classification and staging predictions based on classification and staging standards in a preset rule base. Taking lung cancer as an example, staging is primarily based on tumor size. Therefore, the predictive model uses image processing technology to identify tumors in medical images included in the diagnostic information, obtaining information such as tumor size and location of spread, and matching it with the definitions of lung cancer-related staging in the preset rule base to determine the staging prediction. Existing models can be used for prediction; specific details of the predictive model will not be elaborated here. In other embodiments, semantic analysis technology can also be combined to identify textual content such as medical records and medical image results, obtaining information such as tumor size and location of spread to assist in obtaining staging predictions.

[0080] This embodiment utilizes algorithms and big data analysis to assist in staging cancer patients' conditions, improving the accuracy and reliability of staging. Simultaneously, intelligent processing reduces the possibility of human error and enhances staging efficiency.

[0081] In one embodiment, diagnostic information includes disease name and medical records; matching the diagnostic information with a preset rule base, and obtaining classification prediction and staging prediction based on the matching results, includes: matching the medical records with the classification rule base to obtain classification prediction; matching the disease name with the disease code; if the match is successful, matching the medical records corresponding to the disease name with the staging criteria corresponding to the successfully matched disease code; if the match is successful, obtaining staging prediction.

[0082] This embodiment employs semantic analysis technology to process written medical records and identify the patient's treatment stage as pathological staging, untreated clinical staging, treated clinical staging, or relapse staging. Specifically, the semantic analysis tool used can be Natural Language Processing (NLP), which identifies specific words such as "relapse," "first visit," and "post-operative" to obtain a classification prediction. This step can improve the accuracy of semantic analysis by constraining the vocabulary used in medical record writing. In the process of obtaining the staging prediction, a pre-trained prediction model can be used to achieve staging, as explained above.

[0083] In one embodiment, outputting cancer TNM staging results based on comparison results includes: if the typing selection is consistent with the typing prediction and the staging selection is consistent with the staging prediction, then outputting cancer TNM staging results based on the typing selection and staging selection.

[0084] When both the subtyping and staging selections are consistent with the predictions, it can be determined that the current subtyping and staging selections are highly accurate, and the cancer TNM staging results output by the system are more accurate and reliable.

[0085] Based on the comparison results, the cancer TNM staging results are output, including: if the subtyping selection is inconsistent with the subtyping prediction, the first critical conflict point in the medical record is located according to the classification criteria corresponding to the subtyping selection and the classification criteria corresponding to the subtyping prediction; if the staging selection is inconsistent with the staging prediction, the second critical conflict point in the medical record is located according to the staging criteria corresponding to the staging selection and the staging criteria corresponding to the staging prediction; based on the first and second critical conflict points, corresponding risk warnings are issued.

[0086] For example, if the system identifies a discharge record related to this disease in the medical record, it will likely predict the subtyping as ypTNM or rTNM according to the subtyping rule base. However, if the subtyping selection is cTNM, the discharge record becomes the first critical point of conflict. The system will issue a risk warning indicating the existence of a discharge record corresponding to this disease, suggesting a possible error in the subtyping selection, and the doctor needs to closely monitor and confirm whether to modify it. Similarly, if the system determines the staging prediction based on the size and location of the lesion identified from the medical record or medical images, and this differs from the staging selection entered by the doctor, the portion recording the lesion size and location becomes the second critical point of conflict. The system will issue a risk warning indicating that the actual lesion size exceeds the lesion size corresponding to the staging selection, suggesting a possible error in the staging selection, and the doctor needs to closely monitor and confirm whether to modify it.

[0087] In one embodiment, several types of medical images are acquired, and combined with medical records, cross-modal target recognition is performed to obtain tumor location information, tumor size information, and tumor invasion depth information; the tumor location information, tumor size information, and tumor invasion depth information are combined to perform staging standard matching to obtain staging prediction.

[0088] Specifically, various types of medical images are collected, such as X-rays, CT scans, MRIs, and ultrasounds. These images should come from different patients, covering various disease types and stages to ensure data diversity and representativeness. First, the medical image data is processed. For example, grayscale normalization is performed on the medical images to adjust the pixel values ​​to a uniform range, eliminating the influence of different equipment and acquisition conditions. Images are cropped to remove irrelevant background information, and then scaled to a uniform size for easier subsequent processing. Data augmentation techniques, such as rotation, flipping, and translation, are used to increase the diversity of the image data and improve the model's generalization ability.

[0089] Next, the text information in the medical records is processed. The medical records are cleaned and segmented into words. The word embedding technology GloVe is used to convert the words into low-dimensional vector representations.

[0090] Then, DenseNet is used to extract features from medical images, and GRU is used to process sequence data to capture contextual information in the text. Features from different modalities are fused at multiple scales in the intermediate layers of the model.

[0091] Finally, a multimodal neural network model is constructed, the fused features are input into the classifier, and GAM is introduced to assign weights to different modal features, thereby enhancing the model's focus on key information.

[0092] This invention improves the personalization and accuracy of cancer treatment by standardizing process management and incorporating predictive models, thus contributing to improved patient survival rates and quality of life. Simultaneously, the method of automatically matching cancer TNM staging results based on staging and subtyping selection reduces the workload of medical professionals and improves the overall efficiency of healthcare services. Finally, improving the accuracy of cancer staging reduces misdiagnosis and optimizes treatment processes, thereby lowering healthcare costs and bringing economic savings to medical institutions.

[0093] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0094] The modules in the aforementioned cancer TNM intelligent staging system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0095] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement all the steps in the above method embodiments.

[0096] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements all the steps in the above method embodiments.

[0097] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for installment payments, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0098] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0099] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A cancer TNM intelligent staging system, characterized in that, It includes an information acquisition module, a recognition module, and a classification output module, among which: The information acquisition module is used to obtain cancer diagnosis information; The identification module is used to identify the subtyping selection and staging selection corresponding to the diagnostic information; wherein, the subtyping selection and the staging selection are both determined based on the diagnostic information and a preset rule base; The typing output module is used to output the cancer TNM staging result corresponding to the diagnostic information based on the typing selection and the staging selection.

2. The intelligent TNM staging system for cancer according to claim 1, characterized in that, The diagnostic information is a description of the patient's disease, including medical records, disease name, pathological classification, examination results, patient identity, attending physician, and examining physician, which is entered by the doctor.

3. The cancer TNM intelligent staging system according to claim 1, characterized in that, The preset rule base includes a classification rule base and a staging rule base; the classification rule base includes a classification of diagnosis and treatment stages and corresponding classification standards; the staging rule base includes several disease codes and staging standards, each disease code corresponds to several staging standards, and the staging standards include the progression of the primary tumor, regional lymph nodes and distant metastases.

4. The cancer TNM intelligent staging system according to claim 1, characterized in that, The classification rule base includes diagnostic and treatment stage classifications such as pTNM, cTNM, ypTNM, and rTNM.

5. The intelligent TNM staging system for cancer according to claim 1, characterized in that, The diagnostic information is matched with a preset rule base, and the subtype prediction and staging prediction are obtained based on the matching results. The subtype selection is compared with the subtype prediction, and the staging selection is compared with the staging prediction. The cancer TNM staging result is output based on the comparison results.

6. A cancer TNM intelligent staging system according to claim 5, characterized in that, The system matches medical records with the classification rule base to obtain classification predictions; it matches disease names with disease codes; if a match is successful, it matches the medical records corresponding to the disease names with the staging standards corresponding to the successfully matched disease codes; if a match is successful, it obtains staging predictions.

7. A cancer TNM intelligent staging system according to claim 1, characterized in that, If the typing selection is consistent with the typing prediction, and the staging selection is consistent with the staging prediction, then the cancer TNM staging result is output based on the typing selection and staging selection; if the typing selection is inconsistent with the typing prediction, then the first critical conflict point in the medical record is located based on the classification criteria corresponding to the typing selection and the classification criteria corresponding to the typing prediction. If the staging selection is inconsistent with the staging prediction, then the second critical conflict point in the medical record is located based on the staging criteria corresponding to the staging selection and the staging criteria corresponding to the staging prediction. Based on the first and second critical points of conflict, corresponding risk warnings are issued.

8. A cancer TNM intelligent staging system according to claim 1, characterized in that, Several types of medical images are acquired and combined with medical records to perform cross-modal target recognition, obtaining tumor location information, tumor size information, and tumor invasion depth information; the tumor location information, tumor size information, and tumor invasion depth information are combined to perform staging standard matching and obtain staging prediction.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the operational functions of the system according to any one of claims 1 to 8.

10. A computer-storable medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the operational functions of the system according to any one of claims 1 to 8.

Images