Fig. 1
From: A Cross Spatio-Temporal Pathology-based Lung Nodule Dataset
The workflow for generating the dataset. Firstly, the Electronic Medical Record System (EMRS) is used to identify cases diagnosed with pulmonary occupying lesions within the past six years. Subsequently, these cases are filtered using the Pathology Information System (P.I.S), retaining only those with available pathology information. Finally, one or multi-time CT sequences of the patient are exported from the Picture Archiving and Communication System (PACS). After data extraction, the dataset is categorized into two distinct categories: classification and detection. Under the guidance of expert physicians, annotators extract the coordinate information of the nodules from the CT sequences, based on the corresponding pathological information, and records it in the CSV file. In addition, post-segmentation lung data is also provided.
