Eur Radiol: How does metastasis occur in front of non-small cell lung cancer patients without invasive accuracy?

For patients with a history of tumor disease, the imaging assessment of lymph node metastasis depends primarily on the diameter size of the lymph nodes.
this, it is not possible to < metastatic lymph nodes with a diameter of >1cm or inflammatory reactive lymph nodes with a diameter of >1cm.
of the clinical decision-making system and prognostic evaluation of lung cancer patients is very important for the accurate stageing of the vertical lymph nodes (MedLNs).
's current diagnosis of MedLNs includes imaging tests such as 18F-FDG PET/CT, contrast-enhanced HRCT, and pathological examinations, such as a bronchid ultrasound-guided physioscopic puncture biopsy (EBUS-TBNA), a perigation laparoscopic biopsy, and a MedLNs biopsy.
Despite the non-invasive advantages of imaging examination methods, recent studies have shown that the sensitivity, specificity and positive predictive values of imaging examination are 50-79%, 72-94% and 58-63%, respectively.
, EBUS-TBNA was slightly more accurate, with sensitivity and negative predictions of 78-89% and 91-92%, respectively.
, however, EBUS-TBNA is an invasive test and is not applicable to all patients with non-small cell lung cancer (NSCLC), with some lymph nodes not sufficiently sourced.
that, it is necessary to find or establish a high diagnostic efficiency and non-invasive imaging method! A paper published recently on Eur Radiol called Machine learning-based diagnostic method of pre-18 F-FDG PET/CT for evaluating mediastinal lymph nodes in non-cell lung cancer by using 18F-fluoride deoxygenated glucose (FDG) electron emission fault scanning/computer fault scanning (PET/CT) establishes the best machine learning (ML) model to evaluate metastatic lymph nodes (MedLNs) of non-small cell lung cancer and compare the diagnosis with that of a nuclear medicine physician.
the study included 1,329 MedLNs.
the diagnostic effectiveness of improving decision trees, logic regression, support vector machines, neural networks, and decision forest models.
compared the diagnostic efficacy of the best ML model with that of physicians.
ML methods are divided into ML (MLq) with only quantitative variables and ML (MLc) that adds clinical information.
we analyzed the intake of 18F-FDG based on MedLNs.
Tutual.2 ML Model Sample Graph . Subjects incorporated flowcharts. The TOP40 study, the most important feature in primary tumors and integrated lymph nodes, found that the elevated decision tree model obtained higher sensitivity and negative predictors than doctors, but lower specificity and positive predictions.
differences between the accuracy of physicians and MLq (79.8% vs. 76.8%, p s 0.067).
accuracy of MLc is significantly higher than that of doctors (81.0% to 76.8%, p s 0.009).
in MedLNs with low intake of 18F-FDGs, ML accuracy was significantly higher than physicians (70.0% vs. 63.3%, p s 0.018).
.ML model and ROC curve comparison of physician evaluation results. The ROC curve efficacy of different evaluation methods was evaluated in groups based on vertical lymph node intake This study showed that although there was no significant difference in accuracy between MLq and physicians, MLc's diagnostic efficacy was better than that of MLq and physicians.
ML method is helpful for evaluating low metabolic MedLN.
, adding clinical information to quantitative variables of 18F-FDG PET/CT can improve ML diagnostic results.
is written later: whether the metastasis of the lymph nodes is an important indicator of clinical treatment choice and patient prognosis.
the evaluation of vertical lymph nodes by HRCT is mainly dependent on morphological signs, although non-invasive, but its evaluation effect is limited by lymph node morphology and enhanced heterogeneity.
sensitivity and specificity to early diagnosis of lymph node metastasis.
18F FDG PET/CT is highly sensitive to metastasis lesions, and then joint machine learning to carry out more feature mining and normative evaluation, the comprehensive evaluation of NSCLC is more perfect and accurate!