# Patient-level Fracture Detection on RSNA Source: [Patient-level Fracture Detection on RSNA]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:31:32.464Z Evaluation of patient-level fracture detection performance on the RSNA dataset, reporting Accuracy, Precision, Sensitivity, Specificity, F1-Score, and ROC-AUC. 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 | Patient-level fracture detection | | Dataset (including any stated split) | RSNA (patient-level) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 15 of 15 | | 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. - [Precision](): Higher is better; unit: PERCENT. - [Sensitivity](): Higher is better; unit: PERCENT. - [Specificity](): Higher is better; unit: PERCENT. - [F1-Score](): Higher is better; unit: PERCENT. - [ROC-AUC](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | RSNA First Place Winner | 84.39 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | Proposed Network (Score Fusion) | 83.06 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | Proposed Network (Score Fusion) | 83.06 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | Proposed Method (Score Fusion) | 83.06 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | Proposed Network | 82.67 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | Proposed Method | 82.67 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | Proposed Network | 81.18 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | Mahajan et al. | 76 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | DenseNet121 | 66.67 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | EfficientNetB0 | 62 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | DenseNet121 | 61.42 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | DenseNet121 | 60.99 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | DenseNet121 | 59.26 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | DenseNet121 | 48.34 | Not recorded | 6971af3ae25ad2fa9d1edc13 | | DenseNet121 | 47.6 | Not recorded | 6971af3ae25ad2fa9d1edc13 | ### RSNA First Place Winner Accuracy: 84.39. Result ID: 69979d2f5698665ec4f31683. [Reporting paper]() Recorded evaluation configuration: {"Dataset Split":"Full","Test Set Size":"2019 Patients","Cross Validation":"true"}. This record does not assert independent reproduction. ### Proposed Network (Score Fusion) Accuracy: 83.06. Result ID: 69979d2f5698665ec4f31678. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input":"2.5D (Slice + Max Stacks)","View":"Axial","Aggregation Method":"If Any"}. This record does not assert independent reproduction. ### Proposed Network (Score Fusion) Accuracy: 83.06. Result ID: 69979d2f5698665ec4f31679. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input":"2.5D (Slice + Max Stacks)","View":"Axial","Aggregation Method":"Adaptive Average"}. This record does not assert independent reproduction. ### Proposed Method (Score Fusion) Accuracy: 83.06. Result ID: 69979d2f5698665ec4f31685. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Test Set Size":"2019 Patients"}. This record does not assert independent reproduction. ### Proposed Network Accuracy: 82.67. Result ID: 69979d2f5698665ec4f31676. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input":"2.5D (Slice Stacks)","View":"Axial","Aggregation Method":"If Any"}. This record does not assert independent reproduction. ### Proposed Method Accuracy: 82.67. Result ID: 69979d2f5698665ec4f31684. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Test Set Size":"2019 Patients","Cross Validation":"true"}. This record does not assert independent reproduction. ### Proposed Network Accuracy: 81.18. Result ID: 69979d2f5698665ec4f31677. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input":"2.5D (Max Projection Stacks)","View":"Axial","Aggregation Method":"If Any"}. This record does not assert independent reproduction. ### Mahajan et al. Accuracy: 76. Result ID: 69979d2f5698665ec4f31682. [Reporting paper]() Recorded evaluation configuration: {"Dataset Split":"Full"}. This record does not assert independent reproduction. ### DenseNet121 Accuracy: 66.67. Result ID: 69979d2f5698665ec4f31671. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input":"2D (whole)","View":"Coronal + Sagittal","Aggregation Method":"Score Fusion"}. This record does not assert independent reproduction. ### EfficientNetB0 Accuracy: 62. Result ID: 69979d2f5698665ec4f31675. [Reporting paper]() Recorded evaluation configuration: {"Input":"3D","Aggregation Method":"If Any"}. This record does not assert independent reproduction. ### DenseNet121 Accuracy: 61.42. Result ID: 69979d2f5698665ec4f31672. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input":"2D","View":"Axial","Aggregation Method":"If Any"}. This record does not assert independent reproduction. ### DenseNet121 Accuracy: 60.99. Result ID: 69979d2f5698665ec4f3166f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input":"2D (whole)","View":"Coronal","Aggregation Method":"None"}. This record does not assert independent reproduction. ### DenseNet121 Accuracy: 59.26. Result ID: 69979d2f5698665ec4f31670. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input":"2D (whole)","View":"Sagittal","Aggregation Method":"None"}. This record does not assert independent reproduction. ### DenseNet121 Accuracy: 48.34. Result ID: 69979d2f5698665ec4f31674. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input":"2D","View":"Sagittal","Aggregation Method":"If Any"}. This record does not assert independent reproduction. ### DenseNet121 Accuracy: 47.6. Result ID: 69979d2f5698665ec4f31673. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input":"2D","View":"Coronal","Aggregation Method":"If Any"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.