Method and system for assisting early diagnosis of head and neck squamous cell carcinoma

Abstract

The invention relates to the technical field of early diagnosis of squamous-cell carcinoma, and provides a method for assisting early diagnosis of head and neck squamous-cell carcinoma, which comprises the following steps: S1, recording existing video data obtained by utilizing an MET targeted near-infrared fluorescence imaging technology, and training a supervised learning model by utilizing a video data set; s2, building a classifier system, inputting to-be-classified video data by adopting the classifier system, and segmenting and judging pathological properties of a focus by applying the trained supervised learning model; s3, comparing the efficiency of the established classifier system with the efficiency of diagnosis of early-stage head and neck squamous cell carcinoma by a human physician; and S4, screening key features related to occurrence and development of the head and neck squamous cell carcinoma through a classifier system, connecting the key features with multiple omics data in series to form a comprehensive data set, building an intermediate model based on the comprehensive data set, and performing bioinformatics analysis by using the intermediate model. The early detection rate of HNSCC is increased, the accuracy of biopsy is improved, and early accurate tumor diagnosis is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of early diagnosis of squamous cell carcinoma, and in particular to a method and system for assisting early diagnosis of head and neck squamous cell carcinoma. Background Art

[0002] Head and neck squamous cell carcinoma (HNSCC) mainly refers to squamous cell carcinoma in the nasal cavity, paranasal sinuses, oral cavity, oropharynx, and larynx. Its incidence accounts for about 5% of all tumors in the body, with more than 500,000 new cases worldwide each year (referenced by Siegel RL, Miller KD, Wagle NS and Jemal A, published in "CA: A Cancer Journal For Clinicians", Volume 73, Issue 1, 2023, "Cancer statistics, 2023", and by Sung H, Ferlay J, Siegel RL, Laversanne M, SoerjomataramI, Jemal A et al., published in "CA: A Cancer Journal For Clinicians", Volume 71, Issue 3, 2021, "Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries", and it is one of the major cancers with the fastest rising incidence today (referenced by Mody MD, Rocco JW, Yom SS, Haddad RI and Saba NF "Head and neck cancer" written by Bhat GR, Hyole RG and Li J and published in "Adv Cancer Res 2021, Vol. 152, "Head and neck squamous cell carcinoma" by Johnson DE, Burtness B, Leemans CR, Lui VWY, Bauman JE and Grandis JR, published in "Nat Rev Dis Primers 2020, Vol. 6, No. 1". At the same time, although HNSCC is not the most common malignant tumor, it is one of the top five malignant tumors with the highest mortality rate in the world, with more than 380,000 deaths each year (referenced by "Head and neck cancer: Current challenges and future perspectives" written by Bhat GR, Hyole RG and Li J and published in "Adv Cancer Res 2021, Vol. 152".According to data reported in Nature, despite the advancement of medical development and the rapid development of treatment methods (surgery, radiotherapy, chemotherapy, targeted therapy, immunotherapy), the five-year survival rate of HNSCC patients has not increased significantly. The high local recurrence rate (50%~60%) has led to a mortality rate of nearly 50% for HNSCC, and an average five-year survival rate of only about 42% (referenced by Leemans CR, Snijders PJ and Brakenhoff RH, published in "Nature Reviews Cancer" Vol. 18, No. 5, 2018 "The molecular landscape of head and neck cancer").

[0003] The low early diagnosis rate of HNSCC is one of the important reasons for this dilemma. A review published in the New England Journal in 2020 pointed out that improving the early diagnosis rate of HNSCC and the accuracy of preoperative staging are the key to prolonging patient survival and improving the quality of life (refer to "Head and Neck Cancer" written by Chow LQ and published in the New England Journal of Medicine, Vol. 382, ​​No. 1, 2020). Worldwide, the early diagnosis rate of HNSCC is only 44% to 55% ("Challenges in diagnosing head and neck cancer in primary health care" by Nieminen M, Aro K, Mäkitie A, Harlin V, Kainulainen S, Jouhi L, et al., published in Annals of Medicine, Vol. 53, No. 1, 2021), and the average treatment delay due to misdiagnosis and missed diagnosis is more than 13 months, and more than 60% of the patients initially diagnosed are already in the advanced stage (referenced to "Diagnosis and referral delays in primary care for oral squamous cell cancer: a systematic review" by Grafton-Clarke C, Chen KW and Wilcock J, published in British Journal of General Practice, Vol. 69, No. 679, 2019, and "ClinicalTrials.gov identifier: NCT0262976574" by Ligier K, Dejardin O, Launay L, Benoit E, Babin E, Bara S et al., published in BMC Cancer, Vol. 16, No. 1, 2016, “Health professionals and the early detection of head and neck cancers: a population-based study in a high incidence area”. Although HNSCC is superficial, the main reason for the difficulty in early diagnosis is that early cancer is hidden and has various manifestations, and it is difficult to distinguish it from common benign mucosal lesions such as ulcers and leukoplakia with the naked eye.Epidemiological studies have shown that the incidence of oral ulcers, leukoplakia, etc. in the population is about 2-4%, of which about 5.9-14.0% progress to squamous cell carcinoma each year (authored by Li C, Zhang Q, Sun K, Jia H, Shen X, Tang G, et al., "Autofluorescence imaging as a noninvasive tool of risk stratification formalignant transformation of oral leukoplakia: A follow-up cohort study" published in "Oral Oncology" 2022, Issue 130). In the initial stage of this evolution process, it is often manifested only as microscopic cytological changes. Clinicians often cannot accurately biopsy and obtain early cancerous tissue based on their empirical judgment of the naked eye, which directly leads to clinical misdiagnosis and missed diagnosis. If HNSCC patients can be diagnosed early and treated in time, their 5-year survival rate is expected to increase significantly from the current average of less than 50% to about 85% ("Predictors of locoregional control in stage I / II oral squamous cell carcinoma classified by AJCC 8th edition" written by Subramaniam N, Balasubramanian D, Murthy S, Kumar N, Vidhyadharan S, Vijayan SN et al., published in "European Journal of Surgical Oncology" Vol. 45, No. 11, 2019, pp. 2126-2130). Therefore, timely and accurate detection of small cancer nests in the early stages of HNSCC tumor evolution is the key to effectively assist early diagnosis, and is also the bottleneck and breakthrough for ultimately improving patient prognosis and survival time.

[0004] The rapid development of near-infrared fluorescence (NIRF) imaging hardware equipment and related probes has provided a key non-invasive visualization technology for the early diagnosis of HNSCC. Near-infrared fluorescence imaging has the characteristics of high sensitivity, no ionizing radiation, simple operation, and low cost, and is particularly suitable for long-term and multiple real-time imaging. At present, handheld near-infrared real-time fluorescence imaging devices for HNSCC have gradually matured, and some have entered the clinical trial stage (see "Rapid, non-invasive fluorescence margin assessment: Optical specimen mapping in oral squamous cell carcinoma" written by van Keulen S, van den Berg NS, Nishio N, Birkeland A, Zhou Q, Lu G, et al., published in "Oral Oncology" 2019, Issue 88, and "Intraoperative tumor assessment using real-time molecular imaging in head and neck cancer patients" written by van Keulen S, Nishio N, Fakurnejad S, van den Berg NS, Lu G, Birkeland A, et al., published in "Journal of the American College of Surgeons" 2019, Volume 229, Issue 6, and "Fluorescent Molecular Imaging Can Improve Intraoperative Sentinel Imaging" written by Krishnan G, van den Berg NS, Nishio N, Kapoor S, Pei J, Freeman L, et al. Margin Detection in Oral Squamous Cell Carcinoma" was published in Journal of Nuclear Medicine: Official Publication, Society of Nuclear Medicine, Vol. 63, No. 8, 2022). Near-infrared fluorescence imaging is sensitive to tissue signals within a depth of 5-6 mm. HNSCC originates from squamous cells on the surface of the oral and pharyngeal mucosa, and early lesions are superficial, which can give full play to the advantages of near-infrared fluorescence imaging. On this basis, the addition of targeting elements that respond to tumor-specific molecular markers can further improve its sensitivity, specificity and early diagnostic performance.

[0005] Mesenchymal-epithelial transition factor (MET) has been confirmed to be a driver of many tumors (see "MET-dependent solid tumours—molecular diagnosis and targeted therapy" by Guo R, Luo J, Chang J, Rekhtman N, Arcila M, Drilon A, published in Nature Reviews Clinical Oncology 2020). A study published in Clinical Cancer Research in 2016 first reported that HNSCC had high expression of MET ("HGF / Met Signaling in Head and Neck Cancer: Impact on the Tumor Microenvironment" by Hartmann S, Bhola NE and Grandis JR, published in Clinical Cancer Research, Vol. 22, No. 16, 2016). A Science Signaling interview also pointed out that MET protein is highly expressed at the very early stage of squamous cell carcinoma (see "Science Signaling Podcast for 21 June 2016: MET and skin cancer" by Yuspa SH and VanHook AM, published in the journal Science Signaling in 2016, 9 (433): c14). Preliminary studies by the applicant's research group also showed that the positive expression rate of MET protein in HNSCC is nearly 80%, which is significantly higher than that in the surrounding normal tissues (about 10%) (reference to "Peptide functionalized upconversion / NIR II luminescent nanoparticles for targeted imaging and therapy of oral squamous cell carcinoma" written by Lin B, Wu J, Wang Y, Sun S, Yuan Y, Tao X, etc., published in "Biomaterials Science" Volume 9, Issue 3, 2021); through the detection of tongue specimens at consecutive intervals of the induction time of the spontaneously induced oral squamous cell carcinoma mouse model, we found that MET showed recognizable high expression in the very early stage of HNSCC tumorigenesis.The above research results all suggest that MET is a key molecular marker for the early diagnosis of HNSCC (refer to "Chapter Seven- MET receptor in oncology: From biomarker to therapeutic target" written by Malik R, Mambetsariev I, Fricke J, Chawla N, Nam A, Pharaon R, etc., included in "Advances in Cancer Research", Volume 147, published by Academic Press in 2020, pages 259-301) and a potential biological target.

[0006] In order to verify the feasibility of near-infrared fluorescence imaging technology combined with MET-targeted fluorescent probes for early diagnosis of HNSCC, the applicant's research group previously constructed a small molecule probe cMBP-ICG targeting MET, and confirmed its effectiveness and safety through basic and animal experimental studies. A Chinese invention patent has been applied for (near-infrared fluorescent imaging agent targeting c-Met and its use, patent number: 202210118240.6). In the same year, the "Clinical Trial to Evaluate the Effectiveness of Preoperative Fluorescence Imaging of Oral Lesions with MET Targeted Probes" (registration number: ChiCTR2200058058) was completed, and it was found that the sensitivity and specificity of the cMBP-ICG probe for tumor lesions were 100% and 75%, respectively, and the accuracy of tumor boundary recognition was increased from 60% based on the experience of senior surgeons to 83% based on fluorescence-assisted judgment (referenced by Wang J, Li S, Wang K, Zhu L, Yang L, Zhu Y, etc. "A c-MET-Targeted Topical Fluorescent Probe cMBP-ICG Improves Oral Squamous Cell Carcinoma Detection in Humans", published in "Annals of Surgical Oncology", Volume 30, Issue 1, 2023). Surface-applied fluorescent contrast agents are safe, fast, and convenient, and will help optimize and develop new targeted agents; they are expected to become a non-innovative technology for diagnosis and follow-up in the future (see "ASO Author Reflections: Topically Applied Targeted Near-Infrared Fluorescence Imaging Probe Helps Detect Head and Neck Squamous Cell Carcinoma" written by Wang J, Yuan Y, and Tao X, published in "Annals of Surgical Oncology" in 2022).

[0007] As a new type of molecular imaging technology, targeted near-infrared fluorescence imaging technology is still in the promotion and learning stage in clinical applications (the accuracy results of the above clinical trials are based on the judgment of near-infrared imaging diagnostic physicians with two years of experience). However, similar to all medical imaging data, the raw video and image data of targeted near-infrared fluorescence imaging contain a large number of explicit and invisible features. Although the application of artificial intelligence (AI) in near-infrared fluorescence medical imaging is still blank (refer to "Randomized Clinical Trials of Machine Learning Interventions in Health Care: A Systematic Review" written by Plana D, Shung DL, Grimshaw AA, Saraf A, Sung JJY and Kann BH, published in "JAMA Network Open" Volume 5, Issue 9, 2022, article number e2233946-e2233946), it can be inferred that AI technology can provide strong support for this process. In other medical imaging fields, AI-assisted technology has been shown to enhance image quality, accurately read scan results, accelerate the diagnostic process and reduce diagnostic bias. A large number of studies represented by gastroenteroscopy and cystoscopy have shown that in the face of video image information, real-time computer-aided diagnosis systems can provide experienced human physicians with an additional sensitivity of approximately 20% to 25%, and shorten the diagnosis time by nearly 30 minutes per case (see "Impact of Artificial Intelligence on Miss Rate of Colorectal Neoplasia" written by Wallace MB, Sharma P, Bhandari P, East J, Antonelli G, Lorenzetti R, etc., published in "Gastroenterology", Volume 163, Issue 1, 2022, and "An Artificial Intelligence System for the Detection of Bladder Cancer via Cystoscopy: A Multicenter Diagnostic Study" written by Wu S, Chen X, Pan J, Dong W, Diao X, Zhang R, etc., published in "Journal of the National Cancer Institute", Volume 114, Issue 2, 2022).In the early diagnosis of squamous cell carcinoma, Zhou et al. (authored by dos Santos DFD, deFaria PR, Travençolo BAN, do Nascimento MZ, "Automated detection of tumor regions from oral histological whole slide images using fully convolutional neural networks", published in "Biomedical Signal Processing and Control" 2021, Issue 69, Article Number 102921) developed a machine learning-based automatic detection of laryngeal squamous cell carcinoma in a dual-modal optical imaging microscope. Through the analysis of the pathological image test set, the AUC of the detection result reached 0.981; Tang et al. (authored by Tang H, Li G, Liu C, Huang D, Zhang X, Qiu Y, etc., "Diagnosis of lymph nodemetastasis in head and neck squamous cell carcinoma using deep learning", published in "Laryngoscope Investigative Otolaryngology" 2022, Volume 7, Issue 1 The first issue of the journal Head & Neck used a deep convolutional neural network to extract high-dimensional features from pathological images, making the detection accuracy of cervical lymph node HNCSC exceed 97.3%, improving the accuracy, sensitivity and specificity of the test results. Lopez-Janeiro et al. ("Fully convolutional networks in multimodal nonlinear microscopy images for automated detection of head and neck carcinoma: Pilot study" written by Rodner E, Bocklitz T, von Eggeling F, Ernst G, Chernavskaia O, Popp J et al., published in Head & Neck, Vol. 41, No. 1, 2019) established a salivary gland sample diagnosis model based on machine learning, making the accuracy of malignant tumor detection 84.6%. These methods all use artificial intelligence methods to analyze pathological or imaging data and effectively improve the accuracy of detection.Therefore, we have reason to believe that in the face of more information-rich targeted near-infrared fluorescence imaging data, artificial intelligence has the ability to fully empower human physicians, deeply mine molecular imaging information, further improve the accuracy of early diagnosis of HNSCC, and even reveal the more hidden early biological evolution process of squamous cell carcinoma.

[0008] In summary, MET targeted near-infrared fluorescence imaging can achieve early visualization of head and neck squamous cell carcinoma and improve its early diagnosis rate; artificial intelligence-assisted diagnosis system is expected to deeply mine the molecular biology information and molecular imaging information provided by this new molecular imaging technology, integrate and analyze them, and ultimately affect clinical decision-making. Based on the applicant's previous research, the present invention intends to carry out the following four parts of research in accordance with the research idea of ​​"development of artificial intelligence-assisted early diagnosis system for head and neck squamous cell carcinoma based on MET targeted near-infrared fluorescence imaging technology (hereinafter referred to as targeted diagnosis system) → verification of targeted diagnosis system → exploration of the biological significance of key extracted features of targeted diagnosis system": (1) Data preparation: Based on the MET targeted near-infrared fluorescence imaging video obtained in the previous study, data cleaning and data post-processing such as data segmentation, standardization, and manual delineation of the region of interest are performed, and the pathological results of the paraffin section of the lesion are used as the gold standard to add labels to prepare data for supervised learning model training. (2) Establishment and verification of the targeted diagnosis system model; initially build a classifier system for automatic segmentation and determination of the pathological properties of lesions, evaluate the model and tune hyperparameters through cross-validation between data subsets, analyze learning curves, integrate models, deploy the targeted diagnosis system to the clinical diagnosis and treatment environment, and monitor and debug model performance in real time. (3) Comparison of the diagnostic efficacy of the targeted diagnosis system model and human physicians for early head and neck squamous cell carcinoma; under the same test protocol, test the diagnostic efficacy of the human physician naked eye group, fluorescence imaging + AI-assisted group, human physician fluorescence imaging-assisted group, and human physician human physician fluorescence imaging + AI-assisted group, and perform statistical analysis and result interpretation. (4) Combine multi-omics data to explore the biological significance of key features; conduct multiple data integration analyses such as concatenation, construction of intermediate models, and conversion of key features screened by the targeted diagnosis system and multi-omics data of cases collected in previous studies to mine the biological information of the occurrence and development of HNSCC, so that the model's decisions can be understood and explained, so as to feed back into clinical decision-making. Summary of the invention

[0009] In view of the above problems, the purpose of the present invention is to provide a method and system for assisting the early diagnosis of head and neck squamous cell carcinoma, so as to improve the early detection rate of HNSCC, improve the accuracy of biopsy, achieve early and accurate tumor diagnosis, and thus provide a new intelligent visualization solution for improving surgical resection rate and reducing postoperative recurrence rate. The successful implementation of the present invention is also expected to expand the research method of artificial intelligence-assisted molecular imaging diagnosis to other medical visualization design strategies, which has broad clinical application prospects and basic research value.

[0010] The above-mentioned object of the present invention is achieved through the following technical solutions: A method for assisting early diagnosis of head and neck squamous cell carcinoma comprises the following steps: S1: Pre-collect a large amount of existing video data related to lesions obtained by using MET targeted near-infrared fluorescence imaging technology as a video data set, perform data cleaning and post-processing on the video data set, and use the video data set to train a supervised learning model capable of identifying and classifying lesions; S2: Building a classifier system for targeted diagnosis as an AI aid, using the classifier system to input video data to be classified, performing preprocessing and feature extraction of the video data to be classified, and applying the trained supervised learning model to segment and determine the pathological nature of the lesion; S3: Comparison of the effectiveness of the classifier system constructed for targeted diagnosis with that of human physicians in diagnosing early-stage head and neck squamous cell carcinoma; S4: The classifier system is used to screen key features related to the occurrence and development of head and neck squamous cell carcinoma, and the key features are connected in series with multi-omics data to form a comprehensive data set. Based on the comprehensive data set, an intermediate model is constructed for processing and analyzing multi-omics data, as well as identifying biomarkers and model pathways related to the occurrence and development of head and neck squamous cell carcinoma. The intermediate model is used to perform bioinformatics analysis to explore the biological basis and therapeutic targets of head and neck squamous cell carcinoma.

[0011] Furthermore, in step S1, the video data set is cleaned and post-processed, and the supervised learning model with the ability to identify and classify lesions is trained using the video data set, specifically: Performing data cleaning and post-processing on each of the video data in the video data set, and dividing each of the video data into separate frames, so as to analyze and process the video data of each frame; Performing standardization processing on the video data of each frame, including brightness and contrast adjustment, color standardization, and size standardization, to ensure consistency and comparability of the video data; Manually outlining a region of interest related to the lesion in each frame of the video data, and adding a feature label related to the pathological property to each frame of the video data; The supervised learning model is trained by a supervised learning method using the video data to which the feature labels have been added.

[0012] Furthermore, after step S1, the method further includes: Paraffin section technology was used to obtain the pathological results of the lesions, which served as the gold standard for evaluating the performance of the classifier system.

[0013] Further, in step S2, the classifier system for targeted diagnosis is constructed as the AI ​​aid, the video data to be classified is input by the classifier system, preprocessing and feature extraction of the video data to be classified are performed, and the trained supervised learning model is applied to segment and determine the pathological properties of the lesion, specifically: Building a preliminary framework of the classifier system, the framework can process the input video data to be classified, preprocess and extract features of the video data to be classified, and apply the trained supervised learning model to segment and determine the pathological properties of the lesion, wherein the preprocessing includes operations such as removing noise, adjusting the brightness, contrast and color balance of the image, and the feature extraction includes extraction of information representing the characteristics of the lesion, such as the color, texture, shape and size of the image; Evaluating the performance of the classifier system by comparing the pathological results with the gold standard, and optimizing and adjusting the classifier system according to the results of the performance evaluation to improve the accuracy and reliability of the classifier system; Analyze and report the results of the development and evaluation of the classifier system, including the performance indicators of the classifier system, case analysis of success and failure after segmentation and judgment by the classifier system, and the direction in which the classifier system can be improved.

[0014] Furthermore, in step S3, the effectiveness of the constructed classifier system for targeted diagnosis is compared with that of human physicians in diagnosing early-stage head and neck squamous cell carcinoma, specifically: Design a unified test protocol, including specific test tasks, evaluation criteria, data sets, and evaluation metrics; Four groups of diagnostic methods were set up, including a human physician naked eye group, a fluorescence imaging + AI-assisted group, a human physician fluorescence imaging-assisted group, and a human physician fluorescence imaging + AI-assisted group, and different diagnostic methods were tested respectively. The human physician naked eye group was a human doctor who made a diagnosis only with naked eyes, the fluorescence imaging + AI-assisted group was a human doctor who made a diagnosis without relying on human physicians and only used fluorescence imaging technology and AI-assisted diagnosis, the human physician fluorescence imaging-assisted group was a human physician who made a diagnosis using fluorescence imaging technology, and the human physician fluorescence imaging + AI-assisted group was a human physician who made a diagnosis with the assistance of fluorescence imaging technology and AI. According to the designed test protocol, each group is tested, and the diagnostic performance of each group is recorded, and the diagnostic performance of each group is evaluated using the predetermined evaluation criteria including accuracy, recall rate, and FI score; Statistical analysis was performed to compare the diagnostic efficacy between different groups, including the use of analysis of variance (ANOVA) and t-test to determine whether the efficacy differences were significant, and to interpret the results of the statistical analysis, including the advantages and disadvantages of each group, and whether fluorescence imaging technology and AI assistance could significantly improve the diagnostic efficacy.

[0015] Further, in step S4, the key features related to the occurrence and development of head and neck squamous cell carcinoma are screened by the classifier system, the key features are connected in series with the multi-omics data to form a comprehensive data set, and an intermediate model for processing and analyzing the multi-omics data and identifying the biomarkers and model pathways related to the occurrence and development of the head and neck squamous cell carcinoma is built based on the comprehensive data set, specifically: Screening out the key features related to the occurrence and development of head and neck squamous cell carcinoma through the classifier system; Collect multi-omics data including genomics, transcriptomics, proteomics, and metabolomics; The screened key features are connected in series with the collected multi-omics data to form the comprehensive data set, and the intermediate model is built based on the comprehensive data set, wherein the intermediate model can process and analyze the multi-omics data, and identify biomarkers and model pathways related to the occurrence and development of head and neck squamous cell carcinoma; The output of the intermediate model is transformed, including converting the expression level or variation data of the biomarker into a form that is more easily understood and used, to facilitate analysis and interpretation.

[0016] Further, in step S4, the intermediate model is used to perform bioinformatics analysis to explore the biological basis and therapeutic targets of head and neck squamous cell carcinoma, specifically: Using the intermediate model to perform bioinformatics analysis to explore the biological basis and therapeutic targets of head and neck squamous cell carcinoma; Analyze and evaluate the interpretability of the intermediate model to ensure that the decision-making process of the intermediate model can be understood and explained, including using model interpretation techniques such as LIME or SHAP to understand how the intermediate model uses input data to make decisions; Feed the interpretive results of the intermediate model and the findings of bioinformatics analysis back into clinical decision-making to provide better patient management and treatment options, and continuously optimize and update the model and analysis methods based on clinical feedback and new research findings to maintain relevance and accuracy in head and neck squamous cell carcinoma research and clinical applications.

[0017] An auxiliary head and neck squamous cell carcinoma early diagnosis system for executing the auxiliary head and neck squamous cell carcinoma early diagnosis method as described above, comprising: A data preparation module is used to pre-collect a large amount of existing video data related to lesions obtained using MET targeted near-infrared fluorescence imaging technology as a video data set, perform data cleaning and post-processing on the video data set, and use the video data set to train a supervised learning model capable of identifying and classifying lesions; A targeted diagnosis system establishment module is used to build a classifier system for targeted diagnosis as an AI aid, use the classifier system to input video data to be classified, perform preprocessing and feature extraction of the video data to be classified, and apply the trained supervised learning model to segment and determine the pathological nature of the lesion; A diagnostic performance comparison module, used to compare the diagnostic performance of the classifier system constructed for targeted diagnosis with that of human physicians in diagnosing early-stage head and neck squamous cell carcinoma; The biological significance exploration module is used to screen key features related to the occurrence and development of head and neck squamous cell carcinoma through the classifier system, concatenate the key features with multi-omics data to form a comprehensive data set, build an intermediate model for processing and analyzing multi-omics data based on the comprehensive data set, and identify biomarkers and model pathways related to the occurrence and development of head and neck squamous cell carcinoma, and use the intermediate model to perform bioinformatics analysis to explore the biological basis and therapeutic targets of head and neck squamous cell carcinoma.

[0018] A computer device comprises a memory and one or more processors, wherein the memory stores computer codes, and when the computer codes are executed by the one or more processors, the one or more processors execute the above method.

[0019] A computer-readable storage medium stores computer codes. When the computer codes are executed, the above method is executed.

[0020] Compared with the prior art, the present invention has at least one of the following beneficial effects: (1) The present invention uses MET targeted probes combined with near-infrared fluorescence imaging technology as the data source for machine learning. This method utilizes the high sensitivity and deep tissue penetration of near-infrared fluorescence imaging, combined with the high specificity of MET probes, to more accurately mark and identify the lesion area of ​​head and neck squamous cell carcinoma (HNSCC). This technology provides a high-quality training data set with richer information levels for machine learning algorithms, which greatly enhances the accuracy of disease identification and classification. With these data, the classifier system as an AI model can be trained to obtain the precise boundaries and biological characteristics of the lesions, thereby providing a new perspective for the early diagnosis of HNSCC. In addition, matching these image data with high-quality pathological results and immunohistochemistry results provides a "gold standard" for verifying the accuracy of the machine learning model, further enhancing the value and reliability of the AI-assisted system in actual clinical applications. This multimodal data fusion strategy not only optimizes the existing diagnostic process, but also enhances the interpretability of the model, providing doctors with an interpretable diagnostic basis, reflecting the application potential of artificial intelligence in the field of precision medicine.

[0021] (2) By analyzing multi-omics data related to MET, including gene expression, protein expression, and other molecular changes, we are able to identify biological events that affect the course of HNSCC and use these events to build and iteratively design more accurate artificial intelligence models. This method can extract key information in the disease mechanism and provide dynamic and continuous biomarker evaluation for early diagnosis and formulation of treatment strategies. Through continuous model iteration, it is possible to further predict the responsiveness to targeted drug therapy, thereby achieving personalized treatment. In clinical decision-making, this research feature allows doctors to obtain data-driven insights that are automatically learned from large-scale data sets through deep learning algorithms. This iterative design approach ensures the timeliness and adaptability of the model, enabling it to continue to advance with the latest research discoveries and clinical practice. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The overall flow chart of the method for assisting early diagnosis of head and neck squamous cell carcinoma of the present invention; Figure 2 The overall design roadmap for the method of assisting early diagnosis of head and neck squamous cell carcinoma of the present invention; Figure 3 It is a data processing flow chart of the present invention; Figure 4A flow chart for the development and verification of a supervised learning model for the present invention as an AI model; Figure 5 This is the overall structural diagram of the system for assisting early diagnosis of head and neck squamous cell carcinoma according to the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0024] Those skilled in the art will appreciate that, unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0025] Head and neck squamous cell carcinoma (HNSCC) accounts for about 90% of head and neck malignant tumors. Early cancer is hidden and the early diagnosis rate is low, which is one of the key factors for the high recurrence rate and poor prognosis of patients. Targeted near-infrared fluorescence imaging technology provides new possibilities for improving the early imaging ability of tumors. In the early stage, a targeted MET small molecule probe was constructed, and preliminary clinical studies verified that its accuracy in showing early oral mucosal cancer was increased from 56% to about 84%.

[0026] By utilizing MET-targeted near-infrared fluorescence imaging technology, early visual diagnosis of head and neck squamous cell carcinoma can be achieved, thereby significantly improving the early diagnosis rate of the disease. With the introduction of artificial intelligence-assisted diagnosis systems, it is expected that the rich molecular biology and molecular imaging information provided by this new molecular imaging technology will be deeply mined and analyzed. By integrating and analyzing this information, we can better understand the biological mechanisms of lesions and thus influence clinical decision making. This project plans to carry out the following three major studies based on the applicant's previous research results: (1) Establishment and verification of targeted diagnosis system model: Introduce machine learning to conduct in-depth mining of the above molecular imaging data and establish a classifier system for targeted diagnosis.

[0027] Initially build a classifier system for automatic segmentation and determination of the pathological properties of lesions. Evaluate the performance of the model on different data subsets through cross-validation and perform hyperparameter tuning. Analyze the learning curve to identify possible overfitting or underfitting problems and take appropriate measures to solve these problems. Perform model integration to improve the stability and accuracy of the classifier system. Deploy the targeted diagnosis system in the clinical diagnosis and treatment environment, and monitor and debug the performance of the model in real time to ensure its accuracy and reliability.

[0028] (2) Diagnostic efficacy ratio: The effectiveness of the targeted diagnosis model is verified by comparing it with the diagnosis of human physicians.

[0029] Design a unified test protocol to evaluate the efficacy of different diagnostic methods (human physician naked eye diagnosis, fluorescence imaging + AI-assisted diagnosis, fluorescence imaging-assisted diagnosis, and fluorescence imaging + AI-assisted diagnosis) in the early diagnosis of head and neck squamous cell carcinoma. Compare the diagnostic efficacy of different diagnostic methods through statistical analysis, and interpret and discuss the results.

[0030] (3) Combine multi-omics data to explore the biological significance of key features: Combine existing clinical data, imaging data, pathological data and other multi-omics data to explore the key features suggested by the targeted diagnostic model and mine the biological information of the occurrence and development of HNSCC; ultimately, the model's decisions can be understood and explained so that they can be fed back into clinical decision-making.

[0031] The key features screened by the targeted diagnostic system are integrated with multi-omics data (including genomics, transcriptomics, proteomics, and metabolomics data) collected in previous studies. An intermediate model is constructed to analyze and interpret the integrated data to explore the bioinformatics basis of HNSCC occurrence and development. The biological significance of the key features identified by the model is explored, and these findings are fed back into clinical decision-making to provide more informed clinical diagnosis and treatment plans.

[0032] The present invention introduces artificial intelligence into the field of molecular imaging, which is expected to improve the early detection rate of HNSCC and the accuracy of biopsy, achieve early and accurate tumor diagnosis, and provide new intelligent diagnosis and treatment solutions for improving surgical resection rates and reducing postoperative recurrence rates. At the same time, the successful implementation of the present invention is also expected to expand this innovative research method to other medical visualization design strategies, which has broad clinical application prospects and basic research value. At the social and medical level, the smooth implementation of this project will not only promote the advancement of early diagnosis technology for head and neck squamous cell carcinoma, but also provide strong decision-making support for clinicians, ultimately improving the diagnosis and treatment effects and quality of life of patients.

[0033] The following is described by specific examples: First embodiment like Figure 1 and 2 As shown, this embodiment provides a method for assisting early diagnosis of head and neck squamous cell carcinoma, comprising the following steps: S1: Pre-collect a large amount of existing video data related to lesions obtained using MET targeted near-infrared fluorescence imaging technology as a video data set, perform data cleaning and post-processing on the video data set, and use the video data set to train a supervised learning model capable of identifying and classifying lesions.

[0034] like Figure 3 As described above, this embodiment pre-collects a large amount of existing video data related to lesions obtained using MET targeted near-infrared fluorescence imaging technology as a video data set, screens and excludes all NIRF format video data in the video data set, performs video quality control, and incorporates the video data that passes the quality control into the video data set.

[0035] After the video data set is determined, the video data set is cleaned and post-processed, and the video data set is used to train a supervised learning model capable of identifying and classifying lesions, specifically: Each of the video data in the video data set is cleaned and post-processed, and each of the video data is divided into separate frames so as to analyze and process the video data of each frame. Video processing software or programming libraries (such as OpenCV) can be used to read video files and save them frame by frame as image files.

[0036] The video data of each frame is subjected to standardization processing including brightness and contrast adjustment, color standardization, and size standardization to ensure the consistency and comparability of the video data.

[0037] (1) Brightness and contrast adjustment: The average brightness value of an image can be calculated, and then the brightness can be adjusted to a specific range as needed. For example, the contrast of an image can be enhanced through linear transformation or histogram equalization to make the details in the image clearer.

[0038] Using image enhancement algorithms such as adaptive histogram equalization (AHE) or contrast-limited adaptive histogram equalization (CLAHE), local contrast can be enhanced without changing the overall brightness.

[0039] (2) Color standardization: If the image is in color, the color space can be converted to a standard color space, such as RGB or HSV. Then, the color channels can be normalized, for example, by adjusting the color values ​​to a specific range or a normal distribution with a mean of zero and a variance of one.

[0040] For multiple frames of images, the color consistency can be ensured by calculating the average color value of all images and adjusting the color of each frame to be close to this average value.

[0041] (3) Dimension standardization: If the images are not of the same size, you can resize them to the same size. This can be done with an image scaling algorithm such as bilinear interpolation or nearest neighbor interpolation. Make sure to maintain the proportions and clarity of the image while resizing.

[0042] The region of interest related to the lesion in each frame of the video data is manually outlined, and a feature label related to the pathological property is added to each frame of the video data.

[0043] There are many image annotation tools available, such as LabelImg, VGG Image Annotator (VIA), etc. These tools allow users to manually draw rectangles, polygons, or other shapes on the image to mark the region of interest. The annotator needs to have a certain level of expertise in the image in order to accurately identify and outline the area associated with the lesion. To ensure the accuracy of the annotation, a quality control check can be performed. The same set of images can be annotated by multiple annotators, and their results can then be compared to determine the consistency of the annotation. If there are differences, discussions and corrections can be made. Regular training and evaluation of annotators should be conducted to improve the quality and accuracy of the annotation. Professional pathologists or doctors are required to determine the appropriate feature labels based on the characteristics and pathological properties of the lesions. For example, based on the morphology, color, size, edge and other characteristics of the lesion, it can be labeled as benign, malignant, inflammatory and other different pathological states. To ensure the consistency of the label, clear annotation guidelines and standards can be formulated. Annotators should follow these guidelines to add labels to avoid subjective differences. The label can be reviewed and verified, such as another professional checking the annotated image to ensure the accuracy of the label.

[0044] like Figure 4 As shown, the supervised learning model is trained by the supervised learning method using the video data to which the feature labels have been added. The supervised learning model is also the AI ​​model used subsequently. During training, the video data set is randomly divided into a training set and a validation set. The model is trained by the training set and validated by the validation set.

[0045] After step S1, the method further includes: using paraffin section technology to obtain pathological results of lesions as a gold standard for evaluating the performance of the classifier system.

[0046] Paraffin sectioning is a commonly used method for histopathological examination. First, a tissue sample containing the lesion is obtained from the patient. Then, the tissue sample undergoes a series of treatments, including fixation (usually using a fixative such as formalin to keep the cells and structures in the tissue stable), dehydration (removing water from the tissue for subsequent immersion in paraffin), transparency (using an organic solvent to make the tissue transparent to facilitate paraffin penetration), wax immersion (immersing the tissue in liquid paraffin to allow the paraffin to penetrate into the tissue), and embedding (placing the wax-immersed tissue in a mold, pouring in liquid paraffin, and waiting for the paraffin to cool and solidify to form a wax block containing the tissue).

[0047] Next, a microtome is used to cut the wax block into extremely thin slices (usually a few microns thick). These slices can be mounted on glass slides and stained (such as hematoxylin-eosin staining) to observe the cell morphology, structure, and pathological changes of the tissue under a microscope.

[0048] Pathologists observe the tissue morphology on paraffin sections through a microscope and diagnose the lesions based on indicators such as cell morphology, size, arrangement, characteristics of the cell nucleus, and the presence of abnormal cells. For example, if the presence of cancer cells, obvious cell atypia, and disordered tissue structure are observed, it may be diagnosed as a malignant tumor; if the cell morphology is relatively normal, without obvious atypia and abnormal proliferation, it may be diagnosed as a benign lesion or normal tissue.

[0049] Since paraffin section pathology results are obtained through direct observation and diagnosis of tissues by professional pathologists, they have high accuracy and reliability and are therefore used as the gold standard for classifier systems. This means that when evaluating the performance of a classifier system, its predictions are compared with the paraffin section pathology results.

[0050] S2: Build a classifier system for targeted diagnosis as an AI aid, use the classifier system to input the video data to be classified, perform preprocessing and feature extraction of the video data to be classified, and apply the trained supervised learning model to segment and determine the pathological nature of the lesion.

[0051] Building a preliminary framework of the classifier system, the framework can process the input video data to be classified, preprocess and extract features of the video data to be classified, and apply the trained supervised learning model to segment and determine the pathological properties of the lesion, wherein the preprocessing includes operations such as removing noise, adjusting the brightness, contrast and color balance of the image, and the feature extraction includes extraction of information representing the characteristics of the lesion, such as the color, texture, shape and size of the image; Develop and integrate automatic segmentation algorithms to identify and segment lesions in images without human intervention. Use the trained model to automatically determine the pathological nature of the lesions, such as benign or malignant.

[0052] Evaluating the performance of the classifier system by comparing the pathological results with the gold standard, and optimizing and adjusting the classifier system according to the results of the performance evaluation to improve the accuracy and reliability of the classifier system; Analyze and report the results of the development and evaluation of the classifier system, including the performance indicators of the classifier system, case analysis of success and failure after segmentation and judgment by the classifier system, and the direction in which the classifier system can be improved.

[0053] For example: (1) Performance indicators of classifiers: Accuracy: Measures the ability of a classifier to correctly determine the pathological nature of a lesion. It is usually calculated by dividing the number of correctly classified samples by the total number of samples. For example, if a classifier makes judgments on 100 lesion samples and 85 of them are correct, then the accuracy is 85%.

[0054] Sensitivity: Also known as recall, it measures the ability of a classifier to correctly identify positive samples (such as malignant lesions). High sensitivity means that the classifier can better detect true positive cases and reduce the possibility of missed diagnoses.

[0055] Specificity: Measures the ability of a classifier to correctly identify negative samples (such as benign lesions). High specificity means that the classifier is better able to exclude true negative cases and reduce the possibility of misdiagnosis.

[0056] Precision: reflects how finely the classifier determines the lesion. For example, if the classifier determines that a lesion is malignant but specifies its possible subtype or grade, then the precision is high.

[0057] Other indicators: may also include F1 value (an indicator that comprehensively considers accuracy and sensitivity), area under the ROC curve, etc.

[0058] (2) Case analysis of success and failure: Successful cases: Select some cases where the classifier system accurately judged the pathological nature of the lesion for analysis. For example, for a complex lesion image, the classifier correctly judged it as malignant, which was consistent with the gold standard pathological result. Analyzing successful cases can summarize the advantages and effective methods of the classifier system, such as specific image features, algorithm advantages, etc.

[0059] Failure cases: Analyze cases where the classifier system made an incorrect judgment. For example, the classifier mistakenly judged a benign lesion as malignant, or failed to detect a small malignant lesion. By analyzing failure cases, we can find out the shortcomings of the system, such as incomplete feature extraction and algorithm limitations.

[0060] (3) Possible directions for improvement: Algorithm improvement: Based on performance indicators and case analysis, propose directions for improving the classifier algorithm. For example, adopt more advanced machine learning algorithms, optimize feature extraction methods, adjust parameters, etc.

[0061] Data augmentation: If you find that the classifier system performs poorly on certain types of lesions, you can consider adding relevant types of training data to improve the system's generalization ability.

[0062] Combining with other technologies: Explore the possibility of combining with other technologies to improve the performance of the classifier, such as combining with other imaging technologies, introducing clinical information, etc.

[0063] User interface optimization: If the classifier system is designed for use by clinicians, one can also consider optimizing the user interface to make it easier to operate and understand.

[0064] S3: Compare the performance of the classifier system constructed for targeted diagnosis with that of human physicians in diagnosing early-stage head and neck squamous cell carcinoma.

[0065] In this embodiment, step S3 is specifically as follows: Design a unified test protocol, including specific test tasks, evaluation criteria, data sets, and evaluation indicators. The following are specific examples: (1) Specific tasks of the test: Clarify the goal of the test, such as comparing the diagnostic efficacy of different diagnostic methods (naked eye diagnosis, fluorescence imaging-assisted diagnosis, AI-assisted diagnosis, etc.) for early head and neck squamous cell carcinoma.

[0066] Determine the specific operational procedures for the test, such as having a doctor or model diagnose a given set of images and record the diagnosis results.

[0067] (2) Evaluation criteria: Define the criteria for judging whether the diagnosis is correct or not. For example, the pathological results of the paraffin section of the lesion are used as the gold standard to compare with the results of different diagnostic methods.

[0068] It is possible to consider setting different levels of accuracy requirements. For example, for the diagnosis of malignant lesions, a certain level of accuracy is required.

[0069] (3) Dataset: Select a dataset for testing. This dataset should be representative and cover different types and severity of early-stage head and neck squamous cell carcinoma cases.

[0070] The dataset can come from multiple sources, such as clinical cases in hospitals, medical research databases, etc. Ensure the image quality and annotation accuracy in the dataset.

[0071] (4) Evaluation indicators: Determine the metrics used to measure the effectiveness of different diagnostic methods. Common evaluation metrics include precision, recall, F1 score, etc.

[0072] Precision measures the overall accuracy of the diagnosis; recall measures the proportion of positive cases (such as malignant lesions) correctly identified; and the F1 score takes into account both precision and recall.

[0073] Other indicators can also be selected according to specific needs, such as specificity (the proportion of negative cases correctly identified), accuracy, etc.

[0074] Four groups of diagnostic methods were set up, including a human physician naked eye group, a fluorescence imaging + AI-assisted group, a human physician fluorescence imaging-assisted group, and a human physician fluorescence imaging + AI-assisted group, and different diagnostic methods were tested separately. Among them, the human physician naked eye group was a human doctor who made a diagnosis only by naked eyes, the fluorescence imaging + AI-assisted group was a human doctor who made a diagnosis without relying on human physicians and only used fluorescence imaging technology and AI assistance, the human physician fluorescence imaging-assisted group was a human physician who made a diagnosis using fluorescence imaging technology, and the human physician fluorescence imaging + AI-assisted group was a human physician who made a diagnosis with the assistance of fluorescence imaging technology and AI.

[0075] According to the designed test protocol, each group is tested, and the diagnostic performance of each group is recorded, and the diagnostic performance of each group is evaluated using the predetermined evaluation criteria including accuracy, recall rate, and FI score; Statistical analysis was performed to compare the diagnostic efficacy between different groups, including the use of analysis of variance (ANOVA) and t-test to determine whether the efficacy differences were significant, and to interpret the results of the statistical analysis, including the advantages and disadvantages of each group, and whether fluorescence imaging technology and AI assistance could significantly improve the diagnostic efficacy.

[0076] ANOVA (Analysis of Variance): When comparing the diagnostic performance of multiple groups, analysis of variance can be used. ANOVA can test whether there are significant differences in the means of multiple groups. If the ANOVA results show significant differences between groups, further post hoc tests can be performed to determine which specific groups have differences. For example, ANOVA can be used to determine whether indicators such as accuracy and recall under different diagnostic methods are significantly different.

[0077] t-test: If you are comparing the diagnostic efficacy of two groups, such as the naked eye group and the fluorescence imaging-assisted group, you can use the t-test. The t-test can test whether there is a significant difference in the means of two samples. Similarly, the results of the t-test can be used to determine whether the two diagnostic methods are significantly different in specific indicators.

[0078] S4: The classifier system is used to screen key features related to the occurrence and development of head and neck squamous cell carcinoma, and the key features are connected in series with multi-omics data to form a comprehensive data set. Based on the comprehensive data set, an intermediate model is constructed for processing and analyzing multi-omics data, as well as identifying biomarkers and model pathways related to the occurrence and development of head and neck squamous cell carcinoma. The intermediate model is used to perform bioinformatics analysis to explore the biological basis and therapeutic targets of head and neck squamous cell carcinoma.

[0079] In this embodiment, step S4 is specifically: Screening out the key features related to the occurrence and development of head and neck squamous cell carcinoma through the classifier system; Collect multi-omics data including genomics, transcriptomics, proteomics, and metabolomics; The screened key features are connected in series with the collected multi-omics data to form the comprehensive data set, and the intermediate model is built based on the comprehensive data set, wherein the intermediate model can process and analyze the multi-omics data, and identify biomarkers and model pathways related to the occurrence and development of head and neck squamous cell carcinoma; The output of the intermediate model is transformed, including converting the expression level or variation data of the biomarker into a form that is more easily understood and used, to facilitate analysis and interpretation.

[0080] Using the intermediate model to perform bioinformatics analysis to explore the biological basis and therapeutic targets of head and neck squamous cell carcinoma; Analyze and evaluate the interpretability of the intermediate model to ensure that the decision-making process of the intermediate model can be understood and explained, including using model interpretation techniques such as LIME or SHAP to understand how the intermediate model uses input data to make decisions; Feed the interpretive results of the intermediate model and the findings of bioinformatics analysis back into clinical decision-making to provide better patient management and treatment options, and continuously optimize and update the model and analysis methods based on clinical feedback and new research findings to maintain relevance and accuracy in head and neck squamous cell carcinoma research and clinical applications.

[0081] Second embodiment like Figure 5As shown, this embodiment provides an auxiliary head and neck squamous cell carcinoma early diagnosis system for executing the auxiliary head and neck squamous cell carcinoma early diagnosis method as in the first embodiment, comprising: Data preparation module 1 is used to pre-collect a large amount of existing video data related to lesions obtained by using MET targeted near-infrared fluorescence imaging technology as a video data set, perform data cleaning and post-processing on the video data set, and use the video data set to train a supervised learning model capable of identifying and classifying lesions; A targeted diagnosis system establishment module 2 is used to build a classifier system for targeted diagnosis as an AI aid, use the classifier system to input video data to be classified, perform preprocessing and feature extraction of the video data to be classified, and apply the trained supervised learning model to segment and determine the pathological properties of the lesion; A diagnostic efficacy comparison module 3 is used to compare the efficacy of the classifier system constructed for targeted diagnosis with that of human physicians in diagnosing early head and neck squamous cell carcinoma; Biological significance exploration module 4 is used to screen key features related to the occurrence and development of head and neck squamous cell carcinoma through the classifier system, connect the key features with multi-omics data in series to form a comprehensive data set, build an intermediate model for processing and analyzing multi-omics data based on the comprehensive data set, and identify biomarkers and model pathways related to the occurrence and development of head and neck squamous cell carcinoma, and use the intermediate model to perform bioinformatics analysis to explore the biological basis and therapeutic targets of head and neck squamous cell carcinoma.

[0082] A computer-readable storage medium stores computer code. When the computer code is executed, the above method is executed. A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program. The program can be stored in a computer-readable storage medium. The storage medium can include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0083] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

[0084] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0085] It should be noted that the above embodiments can be freely combined as needed. The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered as the protection scope of the present invention.

Claims

1. A method for assisting the early diagnosis of head and neck squamous cell carcinoma, characterized in that: The following steps are involved: S1: Pre-collect a large amount of existing video data related to lesions obtained by using MET targeted near-infrared fluorescence imaging technology as a video data set, perform data cleaning and post-processing on the video data set, and use the video data set to train a supervised learning model capable of identifying and classifying lesions; S2: Building a classifier system for targeted diagnosis as an AI aid, using the classifier system to input video data to be classified, performing preprocessing and feature extraction of the video data to be classified, and applying the trained supervised learning model to segment and determine the pathological nature of the lesion; S3: Comparison of the effectiveness of the classifier system constructed for targeted diagnosis with that of human physicians in diagnosing early-stage head and neck squamous cell carcinoma; S4: The classifier system is used to screen key features related to the occurrence and development of head and neck squamous cell carcinoma, and the key features are connected in series with multi-omics data to form a comprehensive data set. Based on the comprehensive data set, an intermediate model is constructed for processing and analyzing multi-omics data, as well as identifying biomarkers and model pathways related to the occurrence and development of head and neck squamous cell carcinoma. The intermediate model is used to perform bioinformatics analysis to explore the biological basis and therapeutic targets of head and neck squamous cell carcinoma.

2. The method for assisting early diagnosis of head and neck squamous cell carcinoma according to claim 1, characterized in that: In step S1, the video data set is cleaned and post-processed, and the supervised learning model with the ability to identify and classify lesions is trained using the video data set, specifically: Performing data cleaning and post-processing on each of the video data in the video data set, and dividing each of the video data into separate frames, so as to analyze and process the video data of each frame; Performing standardization processing on the video data of each frame, including brightness and contrast adjustment, color standardization, and size standardization, to ensure consistency and comparability of the video data; Manually outlining a region of interest related to the lesion in each frame of the video data, and adding a feature label related to the pathological property to each frame of the video data; The supervised learning model is trained by a supervised learning method using the video data to which the feature labels have been added.

3. The method for assisting early diagnosis of head and neck squamous cell carcinoma according to claim 1, characterized in that: After step S1, the method further includes: Paraffin section technology was used to obtain the pathological results of the lesions, which served as the gold standard for evaluating the performance of the classifier system.

4. The method for assisting early diagnosis of head and neck squamous cell carcinoma according to claim 3, characterized in that: In step S2, the classifier system for targeted diagnosis is built as the AI ​​aid, the video data to be classified is input by the classifier system, preprocessing and feature extraction of the video data to be classified are performed, and the trained supervised learning model is applied to segment and determine the pathological properties of the lesion, specifically: Building a preliminary framework of the classifier system, the framework can process the input video data to be classified, preprocess and extract features of the video data to be classified, and apply the trained supervised learning model to segment and determine the pathological properties of the lesion, wherein the preprocessing includes operations such as removing noise, adjusting the brightness, contrast and color balance of the image, and the feature extraction includes extraction of information representing the characteristics of the lesion, such as the color, texture, shape and size of the image; Evaluating the performance of the classifier system by comparing the pathological results with the gold standard, and optimizing and adjusting the classifier system according to the results of the performance evaluation to improve the accuracy and reliability of the classifier system; Analyze and report the results of the development and evaluation of the classifier system, including the performance indicators of the classifier system, case analysis of success and failure after segmentation and judgment by the classifier system, and the direction in which the classifier system can be improved.

5. The method for assisting early diagnosis of head and neck squamous cell carcinoma according to claim 1, characterized in that: In step S3, the effectiveness of the constructed classifier system for targeted diagnosis is compared with that of human physicians in diagnosing early head and neck squamous cell carcinoma, specifically: Design a unified test protocol, including specific test tasks, evaluation criteria, data sets, and evaluation metrics; Four groups of diagnostic methods were set up, including a human physician naked eye group, a fluorescence imaging + AI-assisted group, a human physician fluorescence imaging-assisted group, and a human physician fluorescence imaging + AI-assisted group, and different diagnostic methods were tested respectively. The human physician naked eye group was a human doctor who made a diagnosis only with naked eyes, the fluorescence imaging + AI-assisted group was a human doctor who made a diagnosis without relying on human physicians and only used fluorescence imaging technology and AI-assisted diagnosis, the human physician fluorescence imaging-assisted group was a human physician who made a diagnosis using fluorescence imaging technology, and the human physician fluorescence imaging + AI-assisted group was a human physician who made a diagnosis with the assistance of fluorescence imaging technology and AI. According to the designed test protocol, each group is tested, and the diagnostic performance of each group is recorded, and the diagnostic performance of each group is evaluated using the predetermined evaluation criteria including accuracy, recall rate, and FI score; Statistical analysis was performed to compare the diagnostic efficacy between different groups, including the use of analysis of variance (ANOVA) and t-test to determine whether the efficacy differences were significant, and to interpret the results of the statistical analysis, including the advantages and disadvantages of each group, and whether fluorescence imaging technology and AI assistance could significantly improve the diagnostic efficacy.

6. The method for assisting early diagnosis of head and neck squamous cell carcinoma according to claim 1, characterized in that: In step S4, the classifier system is used to screen key features related to the occurrence and development of head and neck squamous cell carcinoma, the key features are connected in series with multi-omics data to form a comprehensive data set, and an intermediate model for processing and analyzing multi-omics data and identifying biomarkers and model pathways related to the occurrence and development of head and neck squamous cell carcinoma is built based on the comprehensive data set, specifically: Screening out the key features related to the occurrence and development of head and neck squamous cell carcinoma through the classifier system; Collect multi-omics data including genomics, transcriptomics, proteomics, and metabolomics; The screened key features are connected in series with the collected multi-omics data to form the comprehensive data set, and the intermediate model is built based on the comprehensive data set, wherein the intermediate model can process and analyze the multi-omics data, and identify biomarkers and model pathways related to the occurrence and development of head and neck squamous cell carcinoma; The output of the intermediate model is transformed, including converting the expression level or variation data of the biomarker into a form that is more easily understood and used, to facilitate analysis and interpretation.

7. The method for assisting early diagnosis of head and neck squamous cell carcinoma according to claim 1, characterized in that: In step S4, bioinformatics analysis is performed using the intermediate model to explore the biological basis and therapeutic targets of head and neck squamous cell carcinoma, specifically: Using the intermediate model to perform bioinformatics analysis to explore the biological basis and therapeutic targets of head and neck squamous cell carcinoma; Analyze and evaluate the interpretability of the intermediate model to ensure that the decision-making process of the intermediate model can be understood and explained, including using model interpretation techniques such as LIME or SHAP to understand how the intermediate model uses input data to make decisions; Feed the interpretive results of the intermediate model and the findings of bioinformatics analysis back into clinical decision-making to provide better patient management and treatment options, and continuously optimize and update the model and analysis methods based on clinical feedback and new research findings to maintain relevance and accuracy in head and neck squamous cell carcinoma research and clinical applications.

8. An assisted early diagnosis system for head and neck squamous cell carcinoma for executing the assisted early diagnosis method for head and neck squamous cell carcinoma according to any one of claims 1 to 7, characterized in that: include: A data preparation module is used to pre-collect a large amount of existing video data related to lesions obtained using MET targeted near-infrared fluorescence imaging technology as a video data set, perform data cleaning and post-processing on the video data set, and use the video data set to train a supervised learning model capable of identifying and classifying lesions; A targeted diagnosis system establishment module is used to build a classifier system for targeted diagnosis as an AI aid, use the classifier system to input video data to be classified, perform preprocessing and feature extraction of the video data to be classified, and apply the trained supervised learning model to segment and determine the pathological nature of the lesion; A diagnostic performance comparison module, used to compare the diagnostic performance of the classifier system constructed for targeted diagnosis with that of human physicians in diagnosing early-stage head and neck squamous cell carcinoma; The biological significance exploration module is used to screen key features related to the occurrence and development of head and neck squamous cell carcinoma through the classifier system, concatenate the key features with multi-omics data to form a comprehensive data set, build an intermediate model for processing and analyzing multi-omics data based on the comprehensive data set, and identify biomarkers and model pathways related to the occurrence and development of head and neck squamous cell carcinoma, and use the intermediate model to perform bioinformatics analysis to explore the biological basis and therapeutic targets of head and neck squamous cell carcinoma.

9. A computer device comprising a memory and one or more processors, wherein the memory stores computer codes, and when the computer codes are executed by the one or more processors, the one or more processors execute the method according to any one of claims 1 to 7. 10 . A computer-readable storage medium storing a computer code. When the computer code is executed, the method according to claim 1 is executed.

Images