# PANDA - Prototype-Anchored Alignment for Partially Unpaired Multimodal Learning, with Applications to Alzheimers MRI and TCGA Pathology

arXiv: https://arxiv.org/abs/2608.25970 · HTML: https://mlsanity.com/paper/2608.25970/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 47 references that are in the map.
- First posted on arXiv 26 Aug 2026.

## Builds on (most cited first, 25 of 25)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [Prototypical Networks for Few-shot Learning](https://mlsanity.com/paper/1703.05175/) (2017)
- [Supervised Contrastive Learning](https://mlsanity.com/paper/2004.11362/) (2020)
- [Attention-based Deep Multiple Instance Learning](https://mlsanity.com/paper/1802.04712/) (2018)
- [Multimodal Machine Learning: A Survey and Taxonomy](https://mlsanity.com/paper/1705.09406/) (2017)
- [Contrastive Learning of Medical Visual Representations from Paired Images and Text](https://mlsanity.com/paper/2010.00747/) (2020)
- [On Variational Bounds of Mutual Information](https://mlsanity.com/paper/1905.06922/) (2019)
- [MONAI: An open-source framework for deep learning in healthcare](https://mlsanity.com/paper/2211.02701/) (2022)
- [Med3D: Transfer Learning for 3D Medical Image Analysis](https://mlsanity.com/paper/1904.00625/) (2019)
- [SMIL: Multimodal Learning with Severely Missing Modality](https://mlsanity.com/paper/2103.05677/) (2021)
- [Sliced Wasserstein Discrepancy for Unsupervised Domain Adaptation](https://mlsanity.com/paper/1903.04064/) (2019)
- [Self-Supervision with Superpixels: Training Few-shot Medical Image Segmentation without Annotation](https://mlsanity.com/paper/2007.09886/) (2020)
- [Convolutional Neural Networks for Classification of Alzheimer's Disease: Overview and Reproducible Evaluation](https://mlsanity.com/paper/1904.07773/) (2019)
- [HeMIS: Hetero-Modal Image Segmentation](https://mlsanity.com/paper/1607.05194/) (2016)
- [A unifying mutual information view of metric learning: cross-entropy vs. pairwise losses](https://mlsanity.com/paper/2003.08983/) (2020)
- [Missing MRI Pulse Sequence Synthesis using Multi-Modal Generative Adversarial Network](https://mlsanity.com/paper/1904.12200/) (2019)
- [Knowledge distillation from multi-modal to mono-modal segmentation networks](https://mlsanity.com/paper/2106.09564/) (2021)
- [Machine Learning with Multi-Site Imaging Data: An Empirical Study on the Impact of Scanner Effects](https://mlsanity.com/paper/1910.04597/) (2019)
- [MoME: Mixture of Multimodal Experts for Cancer Survival Prediction](https://mlsanity.com/paper/2406.09696/) (2024)
- [Cross-Modal Prototype based Multimodal Federated Learning under Severely Missing Modality](https://mlsanity.com/paper/2401.13898/) (2024)
- [HyperFusion: A Hypernetwork Approach to Multimodal Integration of Tabular and Medical Imaging Data for Predictive Modeling](https://mlsanity.com/paper/2403.13319/) (2024)
- [DiaMond: Dementia Diagnosis with Multi-Modal Vision Transformers Using MRI and PET](https://mlsanity.com/paper/2410.23219/) (2024)
- [A Progressive Single-Modality to Multi-Modality Classification Framework for Alzheimer's Disease Sub-type Diagnosis](https://mlsanity.com/paper/2407.18466/) (2024)

## Abstract (arXiv)

Multimodal medical prediction often faces incomplete pairing: auxiliary modalities with complementary signal are available for only a subset of subjects (or none) and cannot be assumed at deployment. We introduce PANDA (Prototype Anchored Data Alignment), a two-stage framework that transfers auxiliary information to a primary-modality model without auxiliary inputs at inference. Stage 1 learns a shared embedding from the paired subset and estimates class prototypes from auxiliary modalities; Stage 2 trains the primary encoder on all subjects using cross-entropy plus alignment to the frozen prototypes. Because supervision is defined at the class-prototype level, PANDA accommodates arbitrary pairing rates, including zero subject overlap. We evaluate PANDA on two applications. On a 1,021-subject multi-scanner ADNI cohort, we perform AD/CN classification with three auxiliary modalities at distinct pairing rates: tabular scores (44.8%), FDG-PET (18.7%), and external handwriting kinematics (0% overlap). Relative to the same-backbone MRI-only baseline, PANDA attains AUC 0.868 +-0.020 (+7.9pp) and reduces 1.5T CN false positives by 24.3pp; on a fully trainable Conv5-FC3 backbone it reaches AUC 0.893 (best overall). A pairing-rate ablation shows that the joint anchor remains within seed noise from 75% to 5% pairing. On TCGA-Lung survival prediction from whole-slide images with RNA-seq as auxiliary data, PANDA improves over WSI-only on 2-year OS (AUC +3.5pp) and Cox PH (C-index +9.0pts) and outperforms full-fusion training, which underperforms WSI-only, while requiring no RNA at inference; wide confidence intervals on this smaller cohort keep the gains below conventional significance. Overall, PANDA provides a deployment-oriented mechanism for leveraging incomplete auxiliary modalities to improve primary-modality prediction.