Table 1_AI-driven spatial analysis of tumor-infiltrating lymphocytes predicts chemo-immuno

An AI pipeline analyzing 32 hematoxylin-eosin-stained breast cancer samples achieved substantial agreement with pathologists (κ=0.73) for TLS recognition and 0.92 accuracy for TIL classification. The dataset, created by Hai Liang and last updated in April 2026, contains spatial metrics derived from these images to predict chemo-immunotherapy response. A random forest model built from these spatial features achieved an AUC of 0.81 for predicting Miller-Payne response grades.

Use Cases

• Train predictive models for chemo-immunotherapy response based on spatial lymphocyte-tumor cell distances.
• Validate AI-based histopathology analysis tools against pathologist annotations for TLS detection.
• Investigate correlations between stromal lymphocyte density and treatment response grades in TNBC.
• Develop feature extraction pipelines for quantifying lymphocyte aggregation patterns in tumor microenvironments.

Strengths

• AI-derived metrics are validated against pathologist assessments with a κ score of 0.73 and 0.92 accuracy.
• The predictive model built from the spatial features achieved an AUC of 0.81 (95% CI: 0.76–0.86).
• Data is shared under a permissive CC-BY-4.0 license.

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 13.8 KB, indicating limited scope.

Provenance

Source

Hai Liang via figshare

Collection Method

AI analysis of 32 HE-stained breast cancer samples from TNBC cases treated with neoadjuvant chemo-immunotherapy.

Time Range

null

Freshness

Last updated 2026-04-16 05:26:03; freshness should be verified.

Geography

null