SALIENT: Frequency-Aware Paired Diffusion for Controllable Long-Tail CT Detection

Yifan Li; Mehrdad Salimitari; Taiyu Zhang; Guang Li; David Dreizin

eess.IVFeb 26, 2026

SALIENT: Frequency-Aware Paired Diffusion for Controllable Long-Tail CT Detection

Authors: Yifan Li, Mehrdad Salimitari, Taiyu Zhang, Guang Li, David Dreizin

Organizations: Trauma Radiology AI Laboratory (TRAIL), Department of Radiology and Nuclear Medicine, University of Maryland School of Medicine · Department of Radiology and Nuclear Medicine, University of Maryland School of Medicine

Abstract

Detection of rare lesions in whole-body CT is fundamentally limited by extreme class imbalance and low target-to-volume ratios, producing precision collapse despite high AUROC. Synthetic augmentation with diffusion models offers promise, yet pixel-space diffusion is computationally expensive, and existing mask-conditioned approaches lack controllable attribute-level regulation and paired supervision for accountable training. We introduce SALIENT, a mask-conditioned wavelet-domain diffusion framework that synthesizes paired lesion-masking volumes for controllable CT augmentation under long-tail regimes. Instead of denoising in pixel space, SALIENT performs structured diffusion over discrete wavelet coefficients, explicitly separating low-frequency brightness from high-frequency structural detail. Learnable frequency-aware objectives disentangle target and background attributes (structure, contrast, edge fidelity), enabling interpretable and stable optimization. A 3D VAE generates diverse volumetric lesion masks, and a semi-supervised teacher produces paired slice-level pseudo-labels for downstream mask-guided detection. SALIENT improves generative realism, as reflected by higher MS-SSIM (0.63 to 0.83) and lower FID (118.4 to 46.5). In a separate downstream evaluation, SALIENT-augmented training improves long-tail detection performance, yielding disproportionate AUPRC gains across low prevalences and target-to-volume ratios. Optimal synthetic ratios shift from 2x to 4x as labeled seed size decreases, indicating a seed-dependent augmentation regime under low-label conditions. SALIENT demonstrates that frequency-aware diffusion enables controllable, computationally efficient precision rescue in long-tail CT detection.

Figures & tables

Figure 1: Overview of the proposed SALIENT synthetic data and classification pipeline. Real CT volumes and lesion masks are first processed by a 3D VAE to generate diverse volumetric masks, which are projected into 2D slice space and used as conditioning signals for the wavelet-domain diffusion model. SALIENT operates on discrete wavelet coefficients to synthesize mask-guided CT slices, which are subsequently pseudo-labeled by a semi-supervised segmentation teacher (UCMT). The resulting synthetic CT–mask pairs augment training for a slice-level mask-guided ResNet-50 classifier. Slice-level predictions are further aggregated into subject-level decisions using an Embedded Vision Transformer (EViT) [ 15 ] .

Figure 2: Architecture of SALIENT in the wavelet domain. A central CT slice and its axial neighbors are transformed into wavelet coefficients (LL, LH, HL, HH). A mask-gated frequency scaling (FSA) module modulates the noisy coefficients before concatenation with a 2.5D conditioning stack. A time-conditioned UNet predicts clean wavelet coefficients at each diffusion step, which are reconstructed into synthetic CT slices via inverse DWT.

Figure 3: Comparison of synthetic CT slices generated by pixel-space MedDDPM and wavelet-domain SALIENT. SALIENT produces sharper anatomical boundaries, reduced high-frequency noise, and improved contrast stability in mediastinal regions.

Figure 4: Quantitative frequency and intensity comparison between Real CT, pixel-space MedDDPM, and SALIENT. (Left) Per-slice LL standard deviation. (Middle) High-frequency variance per slice (LH/HL/HH). (Right) ROI intensity histograms using method-specific masks.

Seed	Prev.	AUPRC ( $0\times$ )	Best AUPRC	Opt. dose	$\Delta$ AUPRC
$n{=}50$	1%	0.8809	0.9414	$2\times$	+0.0605
	2%	0.9062	0.9677	$2\times$	+0.0615
	3%	0.9110	0.9684	$2\times$	+0.0574
	4%	0.9215	0.9745	$2\times$	+0.0530
	5%	0.9223	0.9745	$2\times$	+0.0522
$n{=}25$	1%	0.8169	0.9408	$4\times$	+0.1239

Table 1: Subject-level dose-response of SALIENT augmentation. “Opt. dose” denotes the synthetic-to-real ratio achieving the best AUPRC for each prevalence. SALIENT yields a stable therapeutic dose of

2\times

n{=}50

and a right-shift to

4\times

n{=}25

, with larger gains in the low-label regime.

Figure 5: Saliency alignment improves with paired augmentation. Top: training without mask guidance; middle: without synthetic data; bottom: with SALIENT paired CT–mask augmentation. SALIENT encourages lesion-focused evidence and reduces spurious activations on irrelevant anatomy.

Medical Image Analysis

Medical image analysis uses computational methods to interpret clinical images, measure anatomy and disease, and support diagnosis and treatment.

See all 292 papers

Oct 16, 2026

NV-Generate-Body: 3D Latent Diffusion Model forControllable Anatomy, Tumor, and Demographic Body Mask Synthesis

Curating large labeled CT datasets is expensive and constrained by patient privacy. We present NV-Generate-BodyMask, a 3D latent diffusion model that synthesizes whole-body CT segmentation masks with precise, joint control over organ anatomy, tumor burden, and patient demographics. The model operates in the compressed latent space of a frozen mask autoencoder, generating

256^3

volumetric label maps conditioned on a 19-dimensional vector via Adaptive Group Normalization and cross-attention. A size-adaptive ROI loss with spatial halo weighting and a paired tumor injection pipeline enable accurate tumor size conditioning despite the scarcity of annotated tumor cases. Trained on 8,571 masks across 26 cohorts spanning 132 labels, the model achieves Spearman

ρ= 0.70

0.92

for organ size, 85--100% tumor recall for lung, pancreatic, colon, and bone lesions, and strong demographic alignment (BMI

ρ= 0.78

, sex-conditioned perfect prostate control).

Can Zhao, Pengfei Guo, Dong Yang +8

NVIDIA · National Institutes of Health · University of Arkansas for Medical Sciences

Medical Image Synthesis Conditional Generation 3D Diffusion Models Computed Tomography

Ask PDF

May 12, 2026

Few-Shot Synthetic Data Generation with Diffusion Models for Downstream Vision Tasks

Class imbalance is a persistent challenge in visual recognition, particularly in safety-critical domains where collecting positive examples is expensive and rare events are inherently underrepresented. We propose a lightweight synthetic data augmentation pipeline that fine-tunes a LoRA adapter on as few as 20-50 real images of a rare class and uses a pretrained diffusion model to generate synthetic samples for training. We systematically vary the synthetic-to-real ratio and evaluate the approach across two structurally different domains: chest X-ray pathology classification (NIH ChestX-ray14) and industrial surface crack detection (Magnetic Tile Defect dataset). All evaluations are performed on held-out sets of real images only. Across both domains, synthetic augmentation consistently improves rare-class recall and F1 compared to training with real data alone. Performance improves with moderate synthetic augmentation and shows diminishing returns as the synthetic ratio increases. These results suggest that LoRA-adapted diffusion models provide a simple and scalable mechanism for augmenting rare classes, enabling effective learning in data-scarce scenarios across heterogeneous visual domains.

Daniil Dushenev, Nazariy Karpov, Daniil Zinovjev +2

National University of Science and Technology MISIS, Moscow, Russia

Class-Imbalanced Learning Synthetic Data Augmentation Diffusion Model Fine-Tuning Diffusion-Based Data Augmentation

Ask PDF

May 15, 2026

Flow Matching with Optimized Subclass Priors for Medical Image Augmentation

Rare diseases dominate the diagnostic challenge in medical imaging yet are severely underrepresented in clinical datasets, causing classifiers to fail on exactly the conditions where reliable detection matters most. Generative augmentation can supply the missing tail-class coverage, but coarse disease labels aggregate diverse subtypes and acquisition settings into multi-modal conditionals that bias generators toward dominant submodes, while a shared Gaussian source forces rare subpopulations through disproportionately long transport paths. We propose an offline strategy that introduces informative priors at two levels: first, we partition each coarse label into coherent submodes via Gaussian mixture modeling in the generative model's latent space; second, we learn subclass-conditioned source distributions that re-center and re-scale the starting distribution per submode, shortening trajectories and reducing within-subclass dispersion. To prevent degenerate solutions we impose explicit geometric control, moderately concentrating normalized displacement directions around learnable prototypes while capping path-length outliers. On long-tailed chest X-ray (MIMIC-LT, NIH-LT) and CT slice (CT-RATE) benchmarks the proposed method consistently improves tail-class generation fidelity and diversity (FID, IRS) and is a promising augmentation strategy that reliably improves downstream balanced accuracy and macro-F1 over a non-augmented baseline across modalities.

Felix Nützel, Mischa Dombrowski, Bernhard Kainz

Department of Artificial Intelligence in Biomedical Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg, 91052 Erlangen, Germany · Department of Computing, Imperial College London, London SW7 2AZ, UK

Flow Matching Synthetic Data Augmentation Long-Tail Learning Medical Image Synthesis Conditional Generative Modeling

Ask PDF

Abstract

Medical Image Analysis

Explore similar work

NV-Generate-Body: 3D Latent Diffusion Model forControllable Anatomy, Tumor, and Demographic Body Mask Synthesis

Few-Shot Synthetic Data Generation with Diffusion Models for Downstream Vision Tasks

Flow Matching with Optimized Subclass Priors for Medical Image Augmentation