Demographic-aware deep learning for multi-organ segmentation: Mitigating gender and age biases in CT images

Background: Deep learning algorithms have shown promising results for automated organ-at-risk (OAR) segmentation in medical imaging. However, their performance is frequently compromised by demographic bias. This limitation becomes pronounced when conventional models fail to account for Complex 3D anatomical variations across diverse groups, as they often overlook critical factors such as age and gender. Consequently, this oversight can lead to inaccurate segmentations, thereby posing significant risks to clinical safety in radiotherapy. Purpose: To address this challenge, in this work, we develop a demographic-aware deep learning framework for multi-organ segmentation in computed tomography (CT) images. Our approach is designed to explicitly mitigate age- and gender-specific biases by incorporating demographic prompts and adaptive attention mechanisms, enabling the capture of multi-view anatomical features across diverse groups. Methods: We propose the Demographic-Aware Network (DA-Net), a novel framework trained on a unified dataset of 489 adult (AMOS2022) and 370 pediatric (Pediatric CT-SEG) CT scans, covering 30 organs and including 355 female scans. To robustly learn group-specific anatomical characteristics, DA-Net integrates the Demographic-Aware Hyper-Convolution (DA-HyperConv) module that dynamically adapts convolutional kernels based on demographic prompts. Additionally, an Adaptive Triplet Attention Block (ATAB) is embedded to further leverage multi-view features and enhance segmentation accuracy. We validate the generalizability and effectiveness of our framework on an external dataset (WORD, 150 adults, 62 females). The framework is evaluated quantitatively using the Dice Similarity Coefficient (DICE) and Normalized Surface Dice (NSD). Results: DA-Net surpasses state-of-the-art (SOTA) methods across both the general group and specific demographic subgroups. In the AMOS2022 dataset (mean age 52.8 (Formula presented.) 16.1 years), DA-Net achieves the highest average DICE of 88.6% and NSD of 76.3% for adults. On the Pediatric CT-SEG (mean age 6.9 (Formula presented.) 4.5 years), it achieves top performance with an average DICE of 75.3% (Formula presented.) 20.4% and NSD of 54.8% (Formula presented.) 20.9%. Notably, our proposed framework achieves substantial DICE improvements of 11% to 30% for gender-specific organs, significantly reducing performance disparities. Robustness and generalizability are further supported by consistent results on external validation using the WORD dataset. Compared with the SOTA methods, the performance improvement of our approach is of substantial importance in both the WORD dataset and the Pediatric CT-SEG. Conclusions: In this work, we propose DA-Net, a segmentation network that explicitly incorporates age and gender attributes to mitigate performance disparities between pediatric and adult groups while combining multiple views of anatomic features to improve performance. By leveraging demographic information, DA-Net enhances segmentation accuracy, especially for gender-specific organs. The proposed framework highlights the necessity of developing fair and personalized models tailored to clinical applications, providing a foundation for building more equitable artificial intelligence systems in medical imaging.