# Deep Learning Discovery of Demographic Biomarkers in Echocardiography

arXiv: https://arxiv.org/abs/2207.06421 · HTML: https://mlsanity.com/paper/2207.06421/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 33 references that are in the map.
- First posted on arXiv 13 Jul 2022.

## Builds on (most cited first, 8 of 8)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [Reading Race: AI Recognises Patient's Racial Identity In Medical Images](https://mlsanity.com/paper/2107.10356/) (2021)
- [CheXclusion: Fairness gaps in deep chest X-ray classifiers](https://mlsanity.com/paper/2003.00827/) (2020)
- [Predicting Cardiovascular Risk Factors from Retinal Fundus Photographs using Deep Learning](https://mlsanity.com/paper/1708.09843/) (2017)
- [High-Throughput Precision Phenotyping of Left Ventricular Hypertrophy with Cardiovascular Deep Learning](https://mlsanity.com/paper/2106.12511/) (2021)

## Built on by (most cited first, 12 of 12)

- [Exploring scalable medical image encoders beyond text supervision](https://mlsanity.com/paper/2401.10815/) (2024)
- [Mind the Graph When Balancing Data for Fairness or Robustness](https://mlsanity.com/paper/2406.17433/) (2024)
- [Are demographically invariant models and representations in medical imaging fair?](https://mlsanity.com/paper/2305.01397/) (2023)
- [Using generative AI to investigate medical imagery models and datasets](https://mlsanity.com/paper/2306.00985/) (2023)
- [Regulating AI Adaptation: An Analysis of AI Medical Device Updates](https://mlsanity.com/paper/2407.16900/) (2024)
- [A data- and compute-efficient chest X-ray foundation model beyond aggressive scaling](https://mlsanity.com/paper/2602.22843/) (2026)
- [AI-based association analysis for medical imaging using latent-space geometric confounder correction](https://mlsanity.com/paper/2311.12836/) (2023)
- [Visualizing chest X-ray dataset biases using GANs](https://mlsanity.com/paper/2305.00147/) (2023)
- [X-ray Recognition: Patient identification from X-rays using a contrastive objective](https://mlsanity.com/paper/2305.00149/) (2023)
- [RetinaGuard: Obfuscating Retinal Age in Fundus Images for Biometric Privacy Preserving](https://mlsanity.com/paper/2509.06142/) (2025)
- [Beyond Metadata: CAPRA for Hidden Subgroup Analysis under Missing Metadata in Medical Imaging](https://mlsanity.com/paper/2607.09102/) (2026)
- [Is invariance all you need for algorithmic fairness? Removing demographic information can create new bias](https://mlsanity.com/paper/2609.32004/) (2026)

## Abstract (arXiv)

Deep learning has been shown to accurately assess 'hidden' phenotypes and predict biomarkers from medical imaging beyond traditional clinician interpretation of medical imaging. Given the black box nature of artificial intelligence (AI) models, caution should be exercised in applying models to healthcare as prediction tasks might be short-cut by differences in demographics across disease and patient populations. Using large echocardiography datasets from two healthcare systems, we test whether it is possible to predict age, race, and sex from cardiac ultrasound images using deep learning algorithms and assess the impact of varying confounding variables. We trained video-based convolutional neural networks to predict age, sex, and race. We found that deep learning models were able to identify age and sex, while unable to reliably predict race. Without considering confounding differences between categories, the AI model predicted sex with an AUC of 0.85 (95% CI 0.84 - 0.86), age with a mean absolute error of 9.12 years (95% CI 9.00 - 9.25), and race with AUCs ranging from 0.63 - 0.71. When predicting race, we show that tuning the proportion of a confounding variable (sex) in the training data significantly impacts model AUC (ranging from 0.57 to 0.84), while in training a sex prediction model, tuning a confounder (race) did not substantially change AUC (0.81 - 0.83). This suggests a significant proportion of the model's performance on predicting race could come from confounding features being detected by AI. Further work remains to identify the particular imaging features that associate with demographic information and to better understand the risks of demographic identification in medical AI as it pertains to potentially perpetuating bias and disparities.