# Whole Slide Image Classification on TCGA-LUAD

Source: [Whole Slide Image Classification on TCGA-LUAD](<https://www.sota2.com/research/sota/whole-slide-image-classification-on-tcga-luad>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:28:59.489Z

Evaluation of Whole Slide Image classification performance on the TCGA-LUAD dataset, reporting metrics such as Accuracy and AUROC.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Whole Slide Image Classification |
| Dataset (including any stated split) | TCGA-LUAD |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Accuracy](<https://www.sota2.com/research/sota/whole-slide-image-classification-on-tcga-luad.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [AUROC](<https://www.sota2.com/research/sota/whole-slide-image-classification-on-tcga-luad.md?metric=AUROC>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| CLAM MB | 69.01 | Not recorded | 68fa6fe81e209131cea48127 |
| Ours (Structured State Space Models) | 68.79 | Not recorded | 68fa6fe81e209131cea48127 |
| CLAM SB | 68.59 | Not recorded | 68fa6fe81e209131cea48127 |
| TransMIL | 63.48 | Not recorded | 68fa6fe81e209131cea48127 |
| Max-pooling | 63.28 | Not recorded | 68fa6fe81e209131cea48127 |
| Mean-pooling | 62.61 | Not recorded | 68fa6fe81e209131cea48127 |
| Transformer | 57.74 | Not recorded | 68fa6fe81e209131cea48127 |
| LSTM | 53.89 | Not recorded | 68fa6fe81e209131cea48127 |

### CLAM MB

Accuracy: 69.01. Result ID: 6990f1a83a6c314849047cd4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fe81e209131cea48127.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5f041e209131ce9e6f7a.md>)

Recorded evaluation configuration: {"Number of parameters":"791 048","Inference time (ms)":"5.85"}. This record does not assert independent reproduction.

### Ours (Structured State Space Models)

Accuracy: 68.79. Result ID: 6990f1a83a6c314849047cd6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fe81e209131cea48127.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fe81e209131cea48127.md>)

[Implementation](<https://github.com/MICS-Lab/s4_digital_pathology>)

Recorded evaluation configuration: {"state dimension":"128"}. This record does not assert independent reproduction.

### CLAM SB

Accuracy: 68.59. Result ID: 6990f1a83a6c314849047cd3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fe81e209131cea48127.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5f041e209131ce9e6f7a.md>)

Recorded evaluation configuration: {"Number of parameters":"790 791","Inference time (ms)":"0.84"}. This record does not assert independent reproduction.

### TransMIL

Accuracy: 63.48. Result ID: 6990f1a83a6c314849047cd5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fe81e209131cea48127.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b2d1e209131cea2c9e3.md>)

[Implementation](<https://github.com/szc19990412/TransMIL>)

Recorded evaluation configuration: {"Number of parameters":"2 672 146","Inference time (ms)":"8.58"}. This record does not assert independent reproduction.

### Max-pooling

Accuracy: 63.28. Result ID: 6990f1a83a6c314849047cd0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fe81e209131cea48127.md>)

Recorded evaluation configuration: {"Number of parameters":"1 025","Inference time (ms)":"77.49"}. This record does not assert independent reproduction.

### Mean-pooling

Accuracy: 62.61. Result ID: 6990f1a83a6c314849047ccf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fe81e209131cea48127.md>)

Recorded evaluation configuration: {"Number of parameters":"1 025","Inference time (ms)":"5.60"}. This record does not assert independent reproduction.

### Transformer

Accuracy: 57.74. Result ID: 6990f1a83a6c314849047cd1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fe81e209131cea48127.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703d1e209131cea49db1.md>)

[Implementation](<https://github.com/tensorflow/tensor2tensor>)

Recorded evaluation configuration: {"Number of parameters":"1 054 978","Inference time (ms)":"2.60"}. This record does not assert independent reproduction.

### LSTM

Accuracy: 53.89. Result ID: 6990f1a83a6c314849047cd2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fe81e209131cea48127.md>)

Recorded evaluation configuration: {"Number of parameters":"789 250","Inference time (ms)":"320.52"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
