Eur Radiol: Perhaps understanding the natural process of lung sub-solid nods can more accurately diagnose lung nods!

With the application of thin-layer HRCT, lung nods are becoming more and more early and clearer! According to the presence of solid ingredients in the nodding stove, the lung sub-solid nodding stove (SSN) can be divided into pure grinding glass density cooker (pGGN), sub-solid nodding stove (PSN).
Most persistent SSN stem pathology has been confirmed to be mostly pre-immersion or inflamed adenocarcinomas, including atypical adenoma-like hyperplus (AAH), in-place cancer (AIS) and inflamed adenocarcinoma (AIC);
, however, after many years of follow-up SSN often manifests it as an inert, stable, or slow-growing process, how long that follow-up is indeed a problem.
to solve this profile, it is critical to identify the natural processes of SSN and IAC and to identify SSNs that can progress into IAC! Most previous studies have been exploring the histological type and immersion of SSN, but no studies have been conducted to evaluate the growth rate of SSN and the natural processes of IAC.
contrary, many studies have looked at the natural processes of SSN or risk factors for SSN growth, but have not focused on the pathological types of growth of SSN.
only a small number of studies focused on the multiplies of SSN, but did not explain the growth pattern of SSN.
recently, great progress has been made in the detection, segmentation and classification of lung nods using deep learning techniques.
recently, a paper published on Eur Radiol called Nature history of pathologically confirmed pulmonary subsolid nodules with deep learning-assisted nodule segmentation explores the natural processes of lung subsume nodules (SSNs) of different pathological types through deep learning-assisted nodule segmentation.
retrospective study included 95 SSN patients who underwent long-term follow-up and surgical removal between June 2012 and June 2019.
to use Dr. The deep learning of wise system detects and divides SSN images of preoperative follow-up CT.
SSNs were divided into immersive adenocarcinoma (IAC, n s 47) and non-IAC (n s 48) groups.
SSN was divided into growth groups (n s 68), non-growth (n s 22) and emerging (n s 5) groups according to the changes in nodding stoves during preoperative follow-up.
we analyzed the cumulative percentage and growth patterns of SSN growth and identified important factors for IAC diagnosis and SSN growth.
picture. After 58.8 months of preoperative follow-up, pGGN progressed to PSN, with new positive ingredients visible.
VDT and MDT distributions are 2630.0, 2482.0 days.
picture. In the new pGGN in preoperative follow-up, the average follow-up time before the visible air bubble sign was 42.1±17.0 months.
SSNs in the IAC showed more growth or new appearances than non-IAC groups (89.4% to 64.6%, p s 0.009).
IAC's volume multiplies time is less significant than non-IAC (1436.0±1188.2 and 2087.5±1799.7 days, p s 0.077).
the median size multiplile time of the IAC is significantly shorter than that of non-IAC (821.7 to 1944.1 days, p s 0.001).
the division of the leaf (p s 0.002) and SSN size (p s 0.004) are important factors in distinguishing the IAC.
the first 70 months of follow-up, the cumulative growth rate of IAC was significantly higher than that of non-IAC.
SSN growth model may be consistent with the exponential model.
the initial volume (p s 0.042) is a predictor of SSN growth.
this study shows that IAC, which manifests ite as SSN, often manifests it as an inert growth process.
IAC has a higher average growth rate than non-IAC.
and initial volume are important signs of IAC diagnosis.
is more likely to increase the initial volume of the SSN.