# AD Detection on in-house OCTA dataset (test) Source: [AD Detection on in-house OCTA dataset (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:29:16.496Z Evaluation of Alzheimer's Disease (AD) detection performance on the test split of the in-house OCTA dataset, measured using ACC, AUROC, and Kappa. 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 | AD Detection | | Dataset (including any stated split) | in-house OCTA dataset (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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](): Higher is better; unit: PERCENT. - [AUROC](): Higher is better; unit: SCALAR. - [Kappa](): Higher is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Polar-Net | 85.32 | Not recorded | 68fa71211e209131cea4f5e0 | | Polar-Net | 85.18 | Not recorded | 68fa71211e209131cea4f5e0 | | Polar-Net | 83.88 | Not recorded | 68fa71211e209131cea4f5e0 | | Polar-Net | 81.91 | Not recorded | 68fa71211e209131cea4f5e0 | | ResNet-34 | 81.25 | Not recorded | 68fa71211e209131cea4f5e0 | | MUCO-Net | 79.68 | Not recorded | 68fa71211e209131cea4f5e0 | | EfficientNet-B3 | 79.42 | Not recorded | 68fa71211e209131cea4f5e0 | | ViT-Base | 79.04 | Not recorded | 68fa71211e209131cea4f5e0 | | VAN-B6 | 77.07 | Not recorded | 68fa71211e209131cea4f5e0 | | HorNet-SGF | 76.02 | Not recorded | 68fa71211e209131cea4f5e0 | | SwinV2-T | 76.01 | Not recorded | 68fa71211e209131cea4f5e0 | | ConvNeXt-S | 75.62 | Not recorded | 68fa71211e209131cea4f5e0 | ### Polar-Net Accuracy: 85.32. Result ID: 69919c6e3a6c314849095a35. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prior Knowledge Matrix":true}. This record does not assert independent reproduction. ### Polar-Net Accuracy: 85.18. Result ID: 69919c6e3a6c314849095a34. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Full Proposed Model":true}. This record does not assert independent reproduction. ### Polar-Net Accuracy: 83.88. Result ID: 69919c6e3a6c314849095a33. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Transformation":false}. This record does not assert independent reproduction. ### Polar-Net Accuracy: 81.91. Result ID: 69919c6e3a6c314849095a32. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"PFEM":false}. This record does not assert independent reproduction. ### ResNet-34 Accuracy: 81.25. Result ID: 69919c6e3a6c314849095a2a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Version":"34 layers"}. This record does not assert independent reproduction. ### MUCO-Net Accuracy: 79.68. Result ID: 69919c6e3a6c314849095a31. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### EfficientNet-B3 Accuracy: 79.42. Result ID: 69919c6e3a6c314849095a2b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Version":"B3"}. This record does not assert independent reproduction. ### ViT-Base Accuracy: 79.04. Result ID: 69919c6e3a6c314849095a2f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Version":"Base"}. This record does not assert independent reproduction. ### VAN-B6 Accuracy: 77.07. Result ID: 69919c6e3a6c314849095a2e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Version":"B6"}. This record does not assert independent reproduction. ### HorNet-SGF Accuracy: 76.02. Result ID: 69919c6e3a6c314849095a2d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Architecture":"Recursive Gated Convolution"}. This record does not assert independent reproduction. ### SwinV2-T Accuracy: 76.01. Result ID: 69919c6e3a6c314849095a30. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Version":"Tiny"}. This record does not assert independent reproduction. ### ConvNeXt-S Accuracy: 75.62. Result ID: 69919c6e3a6c314849095a2c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Version":"Small"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.