Deep Learning Signature for Predicting the Novel Grading System of Clinical Stage I Lung Adenocarcinoma

Deep Learning Signature Based on PET-CT Images for Predicting the Novel Grading System of Clinical Stage I Lung Adenocarcinoma

The purpose of this study is to evaluate the performance of a PET/ CT-based deep learning signature for predicting the grade 3 tumors based on the novel grading system in clinical stage stage I lung adenocarcinoma based on a multicenter prospective cohort.

Inclusion Criteria:

(1) Participants scheduled for surgery for radiological finding of pulmonary lesions from the preoperative thin-section CT scans; (2) The maximum diameter of lesion less than 4 cm on CT scans; (3) The maximum short axis diameter of lymph nodes less than 1 cm on CT scan; (4) The SUVmax of hilar and mediastinal lymph nodes less than 2.5; (5) Pathological confirmation of primary lung adenocarcinoma; (5) Age ranging from 20-75 years; (6) Obtained written informed consent.

Exclusion Criteria:

(1) Multiple lung lesions; (2) Poor quality of PET-CT images; (3) Participants with incomplete clinical information; (4) Mucinous adenocarcinomas; (5) Participants who have received neoadjuvant therapy.