# Sex-specific machine learning classification models improve outcome prediction for abdominal aortic aneurysms

## TL;DR

This study shows that machine learning models tailored to male and female patients separately predict abdominal aortic aneurysm outcomes better than general models, addressing gender disparities in medical predictions.

## Contribution

The novelty lies in demonstrating that sex-specific ML models improve outcome prediction for abdominal aortic aneurysms compared to a general model.

## Key findings

- Sex-specific ML models outperformed general models in predicting patient outcomes for abdominal aortic aneurysms.
- Equalizing sample sizes improved predictions for female patients without reducing model performance for males.
- Sex-specific models revealed differences in feature importance, suggesting the need for sex-based grouping in ML models for AAA prediction.

## Abstract

Abdominal aortic aneurysm (AAA) is an abnormal dilation of the abdominal aorta that carries up to a 90% mortality rate when ruptured. Although male patients experience AAA at a higher rate than females, female patients experience AAA rupture at a rate three- to four-fold higher that of their male counterparts. The current standard clinicians use for determining when to surgically intervene is maximum transverse diameter of the AAA perpendicular to the axis of flow. However, some aneurysms below these diameter thresholds rupture. Machine learning (ML) classification models have been previously shown to predict patient outcomes with more discriminability than the diameter criterion. However, these models do not consider sex-based differences. In this proof-of-concept study, we investigate how creating sex-specific ML models impacts patient outcome prediction as compared to a general model encompassing all patients (sex agnostic).

Computed tomography image sets were acquired from 537 patients (n = 159 female, n = 378 male) at the University of Pittsburgh Medical Center (UPMC) and Mayo Clinic Health Systems. Features used as input to the ML models were categorized as clinical, biomechanical, and morphological data. Prior to ML model training, patient data were randomly split for 20% holdout testing. ML models encompassing all patients (general model), only male patients (male-specific model), and only female patients (female-specific model) were trained and tested.

A female-specific model and male-specific model both had a higher maximum area under the receiver-operating characteristic curve values than a general model for female patients and male patients, respectively. Equalizing the sample sizes of female and male patients in the model led to improved outcomes for female patients without decreasing performance for male patients.

ML classification models show promise in improving predictions of patient outcomes for AAA. The higher AAA prevalence rate for males leads to female patients being underrepresented in AAA datasets. In this proof-of-concept study, we demonstrated that sex-specific models outperformed a general model in predicting patient outcomes. Additionally, equalizing sample sizes within the dataset improved predictions for female patients without compromising overall performance of the model. As ML applications in medicine continue to grow, it is important to consider population representation within datasets to reduce model bias.

The online version contains supplementary material available at 10.1186/s13293-025-00765-w.

Although abdominal aortic aneurysm has a higher prevalence rate in male patients, female patients experience higher rupture rates as well as higher surgical and post-operative mortality rates, indicating a need for sex-based considerations in outcome prediction.This work demonstrates improved outcome predictions when patient models are striated based on sex for female and male patients as well as sex-specific differences in feature importance, suggesting that grouping by sex should be considered in ML models for AAA outcome prediction.As machine learning tools continue to grow in prevalence within the medical environment, accounting for bias within these models is crucial to ensure generalizability.

Although abdominal aortic aneurysm has a higher prevalence rate in male patients, female patients experience higher rupture rates as well as higher surgical and post-operative mortality rates, indicating a need for sex-based considerations in outcome prediction.

This work demonstrates improved outcome predictions when patient models are striated based on sex for female and male patients as well as sex-specific differences in feature importance, suggesting that grouping by sex should be considered in ML models for AAA outcome prediction.

As machine learning tools continue to grow in prevalence within the medical environment, accounting for bias within these models is crucial to ensure generalizability.

The online version contains supplementary material available at 10.1186/s13293-025-00765-w.

Ballooning of the aorta, a major artery that delivers blood from the heart to the body, in the abdomen, or an abdominal aortic aneurysm, is often asymptomatic, but can be deadly if ruptured. However, not all abdominal aortic aneurysms rupture. Predicting if rupture will occur is important, as surgery can prevent rupture, but also comes with its own risks. Clinicians currently use the diameter of the aneurysm to decide when to perform surgery, but diameter does not always accurately predict ruptures. Men get abdominal aortic aneurysms more than women do, but women have a higher rate of these ruptures. There have been some machine learning tools that have been designed to predict when these ruptures occur, but they do not account for differences between men and women in their models. In this work, tools were created that did a better job of predicting whether or not an aneurysm will rupture than diameter. The tools that were sex-specific did a better job than the ones that encompassed all patients.

The online version contains supplementary material available at 10.1186/s13293-025-00765-w.