Magician’s Corner: 9. Performance Metrics for Machine Learning Models
Open accessThe DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.
Abstract
I the previous articles, we showed how to process images and train classification models, segmentation models, generative adversarial networks, as well as image denoising models. We also provided a guide on how to visualize model training metrics with TensorBoard and how to connect a model to your picture archiving and communication system. In this article, we will show you how to evaluate the performance of your models by delving into some details about classification and segmentation metrics and show how to calculate them. Performance metrics are useful during model training and validation. We also provide an interactive way for you to see how a custom threshold influences these metrics. To follow this guide, open Colab by clicking on this link: https://colab.research.google.com. Make sure to use Google Chrome as your web browser. You should be prompted to open a file when you start it, but otherwise, select the File > Open Notebook menu option, select the “Github” tab at the top, enter “RSNA” into the search field, and then select the repository “RSNA/MagiciansCorner.” Select the entry called “Magicians_Corner_9_ Performance_Metrics_for_Machine_Learning.ipynb” to load the notebook. Click the arrow at the left of the first cell to run cell 1 (you may need to accept that it is not Google code).
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
A modified Hausdorff distance for object matching
Dubuisson, M.-P., & Jain, A. K. (2002) · Proceedings of 12th International Conference on Pattern Recognition · doi:10.1109/icpr.1994.576361 · Cited by 1,464
Cited by
Methods for clinical evaluation of artificial intelligence algorithms for medical diagnosis
Park, S. H., Han, K., Jang, H. Y., Park, J. E., Lee, J.-G., Kim, D. W., & Choi, J. (2023) · Radiology · doi:10.1148/radiol.220182 · Cited by 163
Towards Trustworthy Machine Learning in Production: An Overview of the Robustness in MLOps Approach
Bayram, F., & Ahmed, B. S. (2025) · ACM Computing Surveys · doi:10.1145/3708497 · Cited by 48
“The EU requirements for trustworthy AI with different levels of abstraction are summarized in Figure 3.”
Decoding machine learning in nursing research: A scoping review of effective algorithms
Choi, J., Lee, H., & Kim‐Godwin, Y. (2025) · Journal of Nursing Scholarship · doi:10.1111/jnu.13026 · Cited by 15
Marte, M. J., Carpenter, E., Scimeca, M., Russell-Meill, M., Peñaloza, C., Grasemann, U., Miikkulainen, R., & Kiran, S. (2025) · Stroke · doi:10.1161/strokeaha.124.047867 · Cited by 9
Metadata
- Authors
- Erickson, B. J., & Kitamura, F.
- Venue
- Radiology: Artificial Intelligence
- Published
- 2021-05-01
- Volume
- 3
- Issue
- 3
- Pages
- e200126
- ISSN
- 2638-6100 (ISSN-L)
- OpenAlex
- W3163400780
- Cited by
- 306
- References
- 1
- Type
- journal article
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Magician’s Corner: 9. Performance Metrics for Machine Learning Models. https://app.cochat.ai/papers/10.1148/ryai.2021200126
- The paper
- Erickson, B. J., & Kitamura, F. (2021). Magician’s Corner: 9. Performance Metrics for Machine Learning Models. Radiology: Artificial Intelligence. https://doi.org/10.1148/ryai.2021200126
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.