# Dementia Etiology Diagnosis via Collaborative Meta Knowledge Enhancement

arXiv: https://arxiv.org/abs/2607.22770 · HTML: https://mlsanity.com/paper/2607.22770/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 53 references that are in the map.
- First posted on arXiv 24 Jul 2026.

## Builds on (most cited first, 19 of 19)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [DSelect-k: Differentiable Selection in the Mixture of Experts with Applications to Multi-Task Learning](https://mlsanity.com/paper/2106.03760/) (2021)
- [Independent Component Alignment for Multi-Task Learning](https://mlsanity.com/paper/2305.19000/) (2023)
- [A Lifelong Learning Approach to Brain MR Segmentation Across Scanners and Protocols](https://mlsanity.com/paper/1805.10170/) (2018)
- [Task Adaptive Parameter Sharing for Multi-Task Learning](https://mlsanity.com/paper/2203.16708/) (2022)
- [Large-scale Long-tailed Disease Diagnosis on Radiology Images](https://mlsanity.com/paper/2312.16151/) (2023)
- [A Foundation Model for Brain Lesion Segmentation with Mixture of Modality Experts](https://mlsanity.com/paper/2405.10246/) (2024)
- [Prefix Conditioning Unifies Language and Label Supervision](https://mlsanity.com/paper/2206.01125/) (2022)
- [Exploring Training on Heterogeneous Data with Mixture of Low-rank Adapters](https://mlsanity.com/paper/2406.09679/) (2024)
- [SMART: Towards Pre-trained Missing-Aware Model for Patient Health Status Prediction](https://mlsanity.com/paper/2405.09039/) (2024)
- [A Cross-Stitch Architecture for Joint Registration and Segmentation in Adaptive Radiotherapy](https://mlsanity.com/paper/2004.08122/) (2020)
- [Reprogramming Distillation for Medical Foundation Models](https://mlsanity.com/paper/2407.06504/) (2024)
- [Deep Convolutional Neural Network based Classification of Alzheimer's Disease using MRI data](https://mlsanity.com/paper/2101.02876/) (2021)
- [LoRKD: Low-Rank Knowledge Decomposition for Medical Foundation Models](https://mlsanity.com/paper/2409.19540/) (2024)
- [MRI-based Multi-task Decoupling Learning for Alzheimer's Disease Detection and MMSE Score Prediction: A Multi-site Validation](https://mlsanity.com/paper/2204.01708/) (2022)
- [Multi-Modal Mamba Modeling for Survival Prediction (M4Survive): Adapting Joint Foundation Model Representations](https://mlsanity.com/paper/2503.10057/) (2025)

## Abstract (arXiv)

Although artificial intelligence (AI) has shown promising performance in several medical tasks, accurate dementia etiology diagnosis with AI remains challenging due to complex overlapping symptoms among diseases. Scaling up the dataset size by combining the cross-center samples may bring a gain in the pursuit of performance, while the inherent data heterogeneity across centers or populations induces the conflict. Conventional multi-task learning paradigms offer a promising framework; however, they fail to consider critical meta information (e.g., site-specific acquisition and modality availability) to combat the heterogeneity. To address this challenge, we propose a Collaborative Meta Knowledge Enhancement (COME) framework for dementia etiology diagnosis, which injects multi-center acquisition semantics, source identifiers, and modality indicators as heterogeneity-aware embeddings into a unified Transformer architecture for scale-up training, enabling explicit modeling of heterogeneity. Besides, a trust-region constrained optimization scheme is designed to regularize the model from spurious correlations during training through a reference model. Across seven independent cohorts, our method achieves state-of-the-art in-domain performance with a mean macro-averaged AUC of 85.62% and a 4.29-point gain over the strongest baseline, while maintaining superior out-of-domain generalization under both cross-center and cross-sequence evaluations. Extensive validation also confirms the alignment between model predictions and established biomarkers (amyloid, tau) and clinical severity, highlighting the potential of COME to enable robust and interpretable dementia diagnostics in real-world settings.