Clinical Measurements with Calibrated Instance-Dependent Confidence Interval

Rotem Nizhar, Lior Frenkel, Jacob Goldberger
Proceedings of The 8th International Conference on Medical Imaging with Deep Learning, PMLR 301:1227-1237, 2026.

Abstract

Reporting meaningful confidence intervals for the predictions of a regression neural network is critical in medical imaging applications since clinical decisions are based on network predictions. We expect to obtain larger intervals for difficult examples and smaller ones for easier examples to predict. A recently proposed calibration procedure suggests predicting the mean and the variance and scaling the variance on a validation set. Another calibration approach is based on applying conformal prediction to quantile regression. We show that assuming a Gaussian distribution to predict the variance followed by a non-parametric Conformal Prediction technique to scale the estimated variance is the most effective way of achieving a small confidence interval with a coverage guarantee. We report extensive experimental results on various medical imaging datasets and network architectures.

Cite this Paper

BibTeX
@InProceedings{pmlr-v301-nizhar26a, title = {Clinical Measurements with Calibrated Instance-Dependent Confidence Interval}, author = {Nizhar, Rotem and Frenkel, Lior and Goldberger, Jacob}, booktitle = {Proceedings of The 8th International Conference on Medical Imaging with Deep Learning}, pages = {1227--1237}, year = {2026}, editor = {Tasdizen, Tolga and Elhabian, Shireen and Summers, Ronald and Chen, Chen and Koch, Lisa and Zhuang, Yan}, volume = {301}, series = {Proceedings of Machine Learning Research}, month = {09--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v301/main/assets/nizhar26a/nizhar26a.pdf}, url = {https://proceedings.mlr.press/v301/nizhar26a.html}, abstract = {Reporting meaningful confidence intervals for the predictions of a regression neural network is critical in medical imaging applications since clinical decisions are based on network predictions. We expect to obtain larger intervals for difficult examples and smaller ones for easier examples to predict. A recently proposed calibration procedure suggests predicting the mean and the variance and scaling the variance on a validation set. Another calibration approach is based on applying conformal prediction to quantile regression. We show that assuming a Gaussian distribution to predict the variance followed by a non-parametric Conformal Prediction technique to scale the estimated variance is the most effective way of achieving a small confidence interval with a coverage guarantee. We report extensive experimental results on various medical imaging datasets and network architectures.} }
Endnote
%0 Conference Paper %T Clinical Measurements with Calibrated Instance-Dependent Confidence Interval %A Rotem Nizhar %A Lior Frenkel %A Jacob Goldberger %B Proceedings of The 8th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Tolga Tasdizen %E Shireen Elhabian %E Ronald Summers %E Chen Chen %E Lisa Koch %E Yan Zhuang %F pmlr-v301-nizhar26a %I PMLR %P 1227--1237 %U https://proceedings.mlr.press/v301/nizhar26a.html %V 301 %X Reporting meaningful confidence intervals for the predictions of a regression neural network is critical in medical imaging applications since clinical decisions are based on network predictions. We expect to obtain larger intervals for difficult examples and smaller ones for easier examples to predict. A recently proposed calibration procedure suggests predicting the mean and the variance and scaling the variance on a validation set. Another calibration approach is based on applying conformal prediction to quantile regression. We show that assuming a Gaussian distribution to predict the variance followed by a non-parametric Conformal Prediction technique to scale the estimated variance is the most effective way of achieving a small confidence interval with a coverage guarantee. We report extensive experimental results on various medical imaging datasets and network architectures.
APA
Nizhar, R., Frenkel, L. & Goldberger, J.. (2026). Clinical Measurements with Calibrated Instance-Dependent Confidence Interval. Proceedings of The 8th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 301:1227-1237 Available from https://proceedings.mlr.press/v301/nizhar26a.html.

Related Material