# MAISI-v2: Accelerated 3D High-Resolution Medical Image Synthesis with Rectified Flow and Region-specific Contrastive Loss arXiv: https://arxiv.org/abs/2508.05772 · HTML: https://mlsanity.com/paper/2508.05772/ ## Key facts - Cited by 20 papers in the map (data through 14 Oct 2026). - Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models. - Builds on 16 of its 41 references that are in the map. - First posted on arXiv 7 Aug 2025. ## Placed in - Diffusion models / [Medical imaging and segmentation](https://mlsanity.com/diffusion/medical-imaging-and-segmentation/) ## Builds on (most cited first, 16 of 16) - [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020) - [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020) - [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024) - [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022) - [MONAI: An open-source framework for deep learning in healthcare](https://mlsanity.com/paper/2211.02701/) (2022) - [AbdomenCT-1K: Is Abdominal Organ Segmentation A Solved Problem?](https://mlsanity.com/paper/2010.14808/) (2020) - [Can AI help in screening Viral and COVID-19 pneumonia?](https://mlsanity.com/paper/2003.13145/) (2020) - [Medical Image Synthesis with Context-Aware Generative Adversarial Networks](https://mlsanity.com/paper/1612.05362/) (2016) - [MAISI: Medical AI for Synthetic Imaging](https://mlsanity.com/paper/2409.11169/) (2024) - [GenerateCT: Text-Conditional Generation of 3D Chest CT Volumes](https://mlsanity.com/paper/2305.16037/) (2023) - [Hierarchical Amortized Training for Memory-efficient High Resolution 3D GAN](https://mlsanity.com/paper/2008.01910/) (2020) - [WDM: 3D Wavelet Diffusion Models for High-Resolution Medical Image Synthesis](https://mlsanity.com/paper/2402.19043/) (2024) - [Deep Learning to Segment Pelvic Bones: Large-scale CT Datasets and Baseline Models](https://mlsanity.com/paper/2012.08721/) (2020) - [Make-A-Volume: Leveraging Latent Diffusion Models for Cross-Modality 3D Brain MRI Synthesis](https://mlsanity.com/paper/2307.10094/) (2023) - [3D MedDiffusion: A 3D Medical Latent Diffusion Model for Controllable and High-quality Medical Image Generation](https://mlsanity.com/paper/2412.13059/) (2024) - [Flow Diverse and Efficient: Learning Momentum Flow Matching via Stochastic Velocity Field Sampling](https://mlsanity.com/paper/2506.08796/) (2025) ## Built on by (most cited first, 20 of 20) - [Diffusion-Driven Generation of Minimally Preprocessed Brain MRI](https://mlsanity.com/paper/2510.26834/) (2025) - [NodMAISI: Nodule-Oriented Medical AI for Synthetic Imaging](https://mlsanity.com/paper/2512.18038/) (2025) - [Auditing Sybil: Explaining Deep Lung Cancer Risk Prediction Through Generative Interventional Attributions](https://mlsanity.com/paper/2602.02560/) (2026) - [Mitigating Surgical Data Imbalance with Dual-Prediction Video Diffusion Model](https://mlsanity.com/paper/2510.07345/) (2025) - [Exploiting Completeness Perception with Diffusion Transformer for Unified 3D MRI Synthesis](https://mlsanity.com/paper/2602.18400/) (2026) - [Any2Any 3D Diffusion Models with Knowledge Transfer: A Radiotherapy Planning Study](https://mlsanity.com/paper/2605.09622/) (2026) - [CRAFT: Clinical Reward-Aligned Finetuning for Medical Image Synthesis](https://mlsanity.com/paper/2605.12650/) (2026) - [Wavelet-Fusion Diffusion Model for Multimodal Brain MRI Synthesis with Modality and Metadata Conditioning](https://mlsanity.com/paper/2606.00689/) (2026) - [KVAE: Family of Tokenizers for Multimodal Generative Models](https://mlsanity.com/paper/2608.05798/) (2026) - [Knowledge-Guided 3D CT Generation: A Conditioning-Centric Taxonomy](https://mlsanity.com/paper/2608.09992/) (2026) - [Foundation Models for Medical Imaging: Status, Challenges, and Directions](https://mlsanity.com/paper/2602.15913/) (2026) - [UnSCAR: Universal, Scalable, Controllable, and Adaptable Image Restoration](https://mlsanity.com/paper/2603.07406/) (2026) - [Distilling Photon-Counting CT into Routine Chest CT through Clinically Validated Degradation Modeling](https://mlsanity.com/paper/2604.07329/) (2026) - [Weakly-Supervised Lung Nodule Segmentation via Training-Free Guidance of 3D Rectified Flow](https://mlsanity.com/paper/2604.08313/) (2026) - [Controllable Lung Nodule Synthesis via Histogram-Regularized Latent Diffusion Models](https://mlsanity.com/paper/2605.30631/) (2026) - [MedSyn2: Flexible Control of 3D CT Generation via Text and Semantically-Defined Segmentation Prompts](https://mlsanity.com/paper/2606.00967/) (2026) - [3D-GlioPREDICT: 3D Latent Diffusion for Post-Radiotherapy Brain MRI Prediction in Patients with Glioma](https://mlsanity.com/paper/2606.05113/) (2026) - [VoxStruct3D: Structure-Leading Flow Matching for Voxel-Space 3D MRI Synthesis](https://mlsanity.com/paper/2608.04557/) (2026) - [Representation-driven Endoscopic Visual Embedding Alignment for Latent Generation](https://mlsanity.com/paper/2608.07176/) (2026) - [THA-Flow Generative Model: Prosthesis Geometry Prediction from Preoperative CT](https://mlsanity.com/paper/2608.25845/) (2026) ## Abstract (arXiv) Medical image synthesis is an important topic for both clinical and research applications. Recently, diffusion models have become a leading approach in this area. Despite their strengths, many existing methods struggle with (1) limited generalizability that only work for specific body regions or voxel spacings, (2) slow inference, which is a common issue for diffusion models, and (3) weak alignment with input conditions, which is a critical issue for medical imaging. MAISI, a previously proposed framework, addresses generalizability issues but still suffers from slow inference and limited condition consistency. In this work, we present MAISI-v2, the first accelerated 3D medical image synthesis framework that integrates rectified flow to enable fast and high quality generation. To further enhance condition fidelity, we introduce a novel region-specific contrastive loss to enhance the sensitivity to region of interest. Our experiments show that MAISI-v2 can achieve SOTA image quality with 33 × acceleration for latent diffusion model. We also conducted a downstream segmentation experiment to show that the synthetic images can be used for data augmentation. We release our code, training details, model weights, and a GUI demo to facilitate reproducibility and promote further development within the community.