Wang Y et al. PATQUANT

This is a dataset of 283 images used for interobserver agreement study in the work:

"AI algorithm for lung adenocarcinoma pattern quantification (PATQUANT): International validation and advanced risk stratification superior to conventional grading" by Wang Y. et al.

These images were analysed by 13 pathologists and AI tool (PATQUANT).

The dataset includes:

1) Images of lung adenocarcinomas (conventional invasive non-mucinous adenocarcinoma, resection cases).

2) CSV files containing all evaluation results:

*Raw results with percentages of single pattern for each grader.

*Processed results with dominant and secondary pattern / image for each grader.

Pattern coding:

1- ACINAR

2- LEPIDIC  

3- SOLID    

4- MICROPAPILLARY    

5- PAPILLARY    

6- COMPLEX_GLANDULAR

3) Code used for training, test experiments, and WSI processing.

The detailed description of the code components is provided in code_description.txt

We do not provide trained checkpoints.

More details on composition of dataset, principles of analysis, and other aspects can be found in the publication:

[Publication DOI link - appears after publication]

This dataset for academic, non-commercial research only.

The original publication should be cited in case of using the dataset in further publications.