Pulmonary nodule database and prediction model construction method and system based on column diagram

Abstract

The invention belongs to the technical field of data analysis, and discloses a method and a system for constructing a pulmonary nodule database and a prediction model based on a column diagram. The method comprises the steps of collecting LDCT features and clinical related information of benign and malignant pulmonary nodule patients with clear pathological diagnosis to construct a benign and malignant pulmonary nodule database; incorporating a plurality of LDCT indexes and serum CEA levels into a Logistic regression model by adopting an entry method, and forming a pulmonary nodule malignant prediction model; and performing external verification on the constructed pulmonary nodule malignant prediction model. The invention provides an effective and practical malignant pulmonary nodule prediction model jointly applying the LDCT index and the serum CEA level, the malignant probability of the pulmonary nodule can be accurately predicted, and a basis is provided for rapid and reasonable clinical management and effective clinical treatment of a pulmonary nodule patient. The method overcomes the problems of high LDCT false positive rate in lung cancer screening and insufficient sensitivity and specificity in individual application of tumor markers.

Description

Technical field [0001] The present invention belongs to the field of data analysis, and in particular, the present invention relates to a line of lung nodules database and a predictive model, a system, terminal, and storage medium. Background technique [0002] At present, lung cancer is a malignant tumor having a global morbidity and mortality. The incidence and mortality rate of lung cancer are also first. Due to the lack of typical clinical symptoms, only non-specific symptoms such as cough, sputum, chest pain, fever, physical strength. Most lung cancer patients have diagnosed more in the late stage and lose their best time. Early lung cancer is hidden and diversified, clinical characteristics and signs are not obvious, and nearly 60% of patients have been in advanced period of lung cancer. Studies have shown that the five-year survival rate of early lung cancer reached 70%, while the five-year survival rate of late lung cancer was only 16%. With the improvement of people's li...

AI Technical Summary

Problems solved by technology

At the same time, LDCT has some unavoidable shortcomings. LDCT can find nodules in 25% of high-risk groups. After further inspection and tracking, only 4% of them are finally diagnosed as lung cancer, and the false positive rate is as high as 96%. The screening indicators for high-risk groups are strict. Only 30% of all lung cancer patients meet the high-risk group screening criteria of LDCT, and most lung cancer patients cannot be found by LDCT screening

[0003] Through the above analysis, the problems and defects of the existing technology are: the false positive rate of LDCT in lung cancer screening is high, and the sensitivity and specificity of tumor markers alone are not enough; , do not meet the indications for puncture and bronchoscopic biopsy, surgical exploration may be required to confirm the diagnosis, which brings a huge economic burden to patients; for patients with pulmonary nodules that cannot be judged as benign or

Images