32 citations · 86 across the 23 of their papers we have counts for
25 papers
Bayesian additive regression trees for evaluating treatment benefit predictors using observational data
Yuan Xia, Mohsen Sadatsafavi, Jeenat Mehareen +1
A treatment benefit predictor (TBP) is an algorithm that maps a patient's characteristics to their putative benefit from a given treatment, which can be used to inform treatment de…
Value-of-Information Analysis for External Validation of Risk Prediction Models in Multicenter Studies and Systematic Reviews
Laure Wynants, Kim Zhipei Wang, Sabine Grimm +5
External validation studies have finite sample sizes, creating uncertainty about whether a prediction model's Net Benefit (NB) exceeds default strategies' NB. The expected value of…
Progression to the mean: A comparison of Bayesian clinical prediction models outputting the posterior mean versus conventional plug-in predictions
Mohsen Sadatsafavi, Richard D. Riley
Clinical prediction models provide predictions for individuals, typically expressed as point estimates derived from a deterministic function, such as a logistic regression equation…
What Medicine Taught Us About Fairness and What It Missed: Lessons from Reconsidering Race-Specific Lung Function Reference Algorithms
Amin Adibi, Mohsen Sadatsafavi
Since 2019, medical societies have reconsidered race-specific clinical equations often in parallel to and largely independent from algorithmic fairness research. Focusing on lung f…
Non-parametric assessment of the calibration of individualized treatment effects
Mohsen Sadatsafavi, Jeroen Hoogland, Thomas P. A. Debray +1
An important aspect of the performance of algorithms that predict individualized treatment effects (ITE) is moderate calibration, i.e., the average treatment effect among individua…
Transportability of Prognostic Markers: Rethinking Common Practices through a Sufficient-Component-Cause Perspective
Mohsen Sadatsafavi, Gavin Pereira, Wenjia Chen
Transportability, the ability to maintain performance across populations, is a desirable property of markers of clinical outcomes. However, empirical findings indicate that markers…
Sequential sample size calculations and learning curves safeguard the robust development of a clinical prediction model for individuals
Amardeep Legha, Joie Ensor, Rebecca Whittle +7
When prospectively developing a new clinical prediction model (CPM), fixed sample size calculations are typically conducted before data collection based on sensible assumptions. Bu…
Bayesian sample size calculations for external validation studies of risk prediction models
Mohsen Sadatsafavi, Paul Gustafson, Solmaz Setayeshgar +2
Contemporary sample size calculations for external validation of risk prediction models require users to specify fixed values of assumed model performance metrics alongside target…
A general sample size framework for developing or updating a clinical prediction model
Richard D Riley, Rebecca Whittle, Mohsen Sadatsafavi +4
Aims: To propose a general sample size framework for developing or updating a clinical prediction model using any statistical or machine learning method, based on drawing samples f…
Recent Advances, Applications and Open Challenges in Machine Learning for Health: Reflections from Research Roundtables at ML4H 2024 Symposium
Amin Adibi, Xu Cao, Zongliang Ji +39
The fourth Machine Learning for Health (ML4H) symposium was held in person on December 15th and 16th, 2024, in the traditional, ancestral, and unceded territories of the Musqueam,…
A decomposition of Fisher's information to inform sample size for developing fair and precise clinical prediction models -- Part 2: time-to-event outcomes
Richard D Riley, Gary S Collins, Lucinda Archer +9
Background: When developing a clinical prediction model using time-to-event data, previous research focuses on the sample size to minimise overfitting and precisely estimate the ov…
Expected value of sample information calculations for risk prediction model development
Abdollah Safari, Paul Gustafson, Mohsen Sadatsafavi
Risk prediction models are often advertised as deterministic functions that map covariates to predicted risks. However, they are typically trained using finite samples, and as such…
Identification of distributions for risks based on the first moment and c-statistic
Mohsen Sadatsafavi, Tae Yoon Lee, John Petkau
We show that for any family of distributions with support on [0,1] with strictly monotonic cumulative distribution function that has no jumps and is quantile-identifiable (i.e., an…
Evaluating Treatment Benefit Predictors using Observational Data: Contending with Identification and Confounding Bias
Yuan Xia, Mohsen Sadatsafavi, Paul Gustafson
A treatment benefit predictor (TBP) is a function that maps patient characteristics to an estimate of the treatment benefit for that patient. Such predictors support optimizing ind…
The expected value of sample information calculations for external validation of risk prediction models
Mohsen Sadatsafavi, Andrew J Vickers, Tae Yoon Lee +2
In designing external validation studies of clinical prediction models, contemporary sample size calculation methods are based on the frequentist inferential paradigm. One of the w…
Non-parametric inference on calibration of predicted risks
Mohsen Sadatsafavi, John Petkau
Moderate calibration, the expected event probability among observations with predicted probability z being equal to z, is a desired property of risk prediction models. Current grap…
Value of Information Analysis for External Validation of Risk Prediction Models
Mohsen Sadatsafavi, Tae Yoon Lee, Laure Wynants +2
Background: Before being used to inform patient care, a risk prediction model needs to be validated in a representative sample from the target population. The finite size of the va…
Methodological concerns about 'concordance-statistic for benefit' as a measure of discrimination in treatment benefit prediction
Yuan Xia, Paul Gustafson, Mohsen Sadatsafavi
Prediction algorithms that quantify the expected benefit of a given treatment conditional on patient characteristics can critically inform medical decisions. Quantifying the perfor…
Closed-Form Solution of the Unit Normal Loss Integral in Two-Dimensions
Tae Yoon Lee, Paul Gustafson, Mohsen Sadatsafavi
In Value of Information (VoI) analysis, the unit normal loss integral (UNLI) frequently emerges as a solution for the computation of various VoI metrics. However, one limitation of…
Programmable Interface for Statistical & Simulation Models (PRISM): Towards Greater Accessibility of Clinical and Healthcare Decision Models
Amin Adibi, Stephanie Harvard, Mohsen Sadatsafavi
Background: Increasingly, decision-making in healthcare relies on computer models, be it clinical prediction models at point of care or decision-analytic models at the policymaking…
Minding non-collapsibility of odds ratios when recalibrating risk prediction models
Mohsen Sadatsafavi, Hamid Tavakoli, Abdollah Safari
In clinical prediction modeling, model updating refers to the practice of modifying a prediction model before it is used in a new setting. In the context of logistic regression for…
Uncertainty and Value of Information in Risk Prediction Modeling
Mohsen Sadatsafavi, Tae Yoon Lee, Paul Gustafson
Background: Due to the finite size of the development sample, predicted probabilities from a risk prediction model are inevitably uncertain. We apply Value of Information methodolo…
Model-based ROC (mROC) curve: examining the effect of case-mix and model calibration on the ROC plot
Mohsen Sadatsafavi, Paramita Saha-Chaudhuri, John Petkau
The performance of risk prediction models is often characterized in terms of discrimination and calibration. The Receiver Operating Characteristic (ROC) curve is widely used for ev…
Concentration of Benefit index: A threshold-free summary metric for quantifying the capacity of covariates to yield efficient treatment rules
Mohsen Sadatsafavi, Mohammad Ali Mansournia, Paul Gustafson
When data on treatment assignment, outcomes, and covariates from a randomized trial are available, a question of interest is to what extent covariates can be used to optimize treat…