Lung Adenocarcinoma Immune Subtyping Model with TIL Spatial Patterns

A model developed in 2026 by Xia Li for predicting immune subtypes in lung adenocarcinoma. It is based on spatial distribution patterns of tumor-infiltrating lymphocytes quantified from H&E-stained whole-slide images of 503 patients from the TCGA cohort. The model achieved an AUC of 0.839 in internal validation and 0.927 in an external cohort.

Use Cases

• Predicting high- vs. low-immunity subgroups in lung adenocarcinoma based on TIL spatial features.
• Validating computational pathology models against immunohistochemical markers like CD3+, CD8+, CD20+, and CD68+ cell densities.
• Linking morphological phenotypes from H&E slides to molecular immune subtypes for translational research.

Strengths

• Model validation includes an external cohort with a high AUC of 0.927.
• Automated annotation model achieved a 95.09% Dice score for tissue contour segmentation.
• Dataset is licensed under CC-BY-4.0, permitting open reuse and modification.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 17.0 KB, indicating it likely contains model parameters or summary statistics, not raw image data.

Provenance

Source

figshare, authored by Xia Li.

Collection Method

Derived from analysis of the TCGA lung adenocarcinoma cohort and external validation cohorts using transcriptomic analysis and deep learning-based image annotation.

Time Range

Publication date suggests analysis circa 2026.

Freshness

Last updated 2026-05-08 05:40:46; freshness should be verified.

Geography

null