一种基于热层析的良恶性乳腺肿瘤分析方法

An analytical method of benign and malignant breast tumors based on thermal tomography

目的乳腺肿瘤的早发现一直都是治疗的关键。当人体发生病变时,功能性改变(如温度、新陈代谢等)往往会早于形态学改变。但大多数乳腺成像手段(如X线、超声等)只能在乳腺组织产生变化后确定病变,主要用于临床诊断和术后分析而缺乏对乳腺肿瘤的早期分析。因此,提出了一种基于热层析的简单快捷且无侵入性的乳腺肿瘤分析方法。方法通过红外热像仪提取生物体体表温度后,利用Pennes方程对生物体三维温度场进行反演从而得到生物体内热源的信息;将肿瘤看作是一个复杂热源,用MATLAB建立它的模型,并在该模型基础上提出一种区分良恶性肿瘤的方法。该模型可测出肿瘤的深度与大小,分别对应肿瘤的解剖位置和新陈代谢状态,并分别分析280例和80例数据来验证该模型及方法的有效性和准确性。结果MATLAB中模型的相关系数R^2大多在0.9以上,用SPSS 26.0对该区分方法做了Kappa分析和卡方分析,Kappa=0.9,P<0.01。结论模型对肿瘤的预测良好,且在此基础上提出的区分良恶性肿瘤方法具有统计学意义。

Objective Early detection of breast tumors has always been the key to treatment.When a disease occurs in the human body,functional changes(such as temperature,metabolism,etc.)tend to be much earlier than morphological changes.Because most of mammography methods that can only identify the lesions after the changes of breast tissue are mainly used for clinical diagnosis and postoperative analysis,lack early analysis of breast tumors,a new method based on thermal tomography is proposed,which is simple,fast and non-invasive.Methods After extracting the living body’s surface temperature with an infrared thermal imager,the method uses the Pennes equation to invert the three-dimensional temperature field of the organism and then the heat intensity of the internal heat source is detected.The tumor is regarded as a complex heat source established by MATLAB,a method to distinguish benign and malignant breast tumors.The model can measure the depth and size of the tumor.The depth corresponds to the anatomical position of the tumor,and the intensity reflects its metabolic state.The effectiveness and accuracy of the proposed model and method is verified by analyzing the data of 280 cases and 80 cases respectively.Results The model’s R^2 is more than 0.9 in most cases.The Kappa analysis result of this method is that Kappa is equal to 0.9,the result of the chi-square is that P is less than 0.01.Conclusions The model predicts tumors well,and the method of distinguishing benign and malignant tumors proposed on this basis has statistical significance.