Multiple Myeloma Lesion Segmentation on Whole-Body Diffusion-Weighted Imaging via Efficient Anatomical Anticipation and Multimodal Confirmation

Mengmeng Zhang; Shengqian Huang; Junde Zhou; Xiaoping Wu; Hao Luog; Jing Wanga; Yicheng Sun; Jiao Li

cs.CVSep 5, 2026

Multiple Myeloma Lesion Segmentation on Whole-Body Diffusion-Weighted Imaging via Efficient Anatomical Anticipation and Multimodal Confirmation

Authors: Mengmeng Zhang, Shengqian Huang, Junde Zhou, Xiaoping Wu, Hao Luog, Jing Wanga, Yicheng Sun, Jiao Li,

Organizations: the State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, Beijing, China · the School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, 101408, Beijing, China · the Department of Radiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, 100730, Beijing, China · the DAMO Academy, Alibaba Group, Hangzhou, Zhejiang, China ·

Abstract

Whole-body diffusion-weighted imaging (WB-DWI) is widely used for multiple myeloma (MM) assessment, yet automated lesion segmentation remains challenging due to limited anatomical delineation and the low specificity of marrow hyperintensity. Existing studies have introduced bone region-of-interest (ROI) information and apparent diffusion coefficient (ADC) maps to mitigate these ambiguities, but practical limitations remain. Bone ROI construction often relies on costly manual annotation, image registration, or dedicated bone models, while ADC is usually incorporated only through simple channel fusion, limiting its ability to provide complementary structural and lesion-discriminative cues. To address these limitations, we propose a two-stage framework for MM lesion segmentation on WB-DWI. In the first stage, we train a bone ROI generation model from ADC images without dedicated bone labels, providing an efficient and practical anatomical prior for lesion analysis. In the second stage, we propose Anatomy-guided Multimodal U-Net (AMU-Net), which leverages ADC in a manner consistent with clinical lesion assessment rather than treating it as a generic auxiliary modality. Extensive experiments demonstrate the effectiveness and practicality of the proposed method. It achieves the best overall performance among the evaluated methods, with a mean Dice score of 76.2%.

Figures & tables

Figure 1: Motivation of the proposed framework. In clinical reading, radiologists first anticipate anatomically plausible marrow regions as skeletal priors and then jointly interpret WB-DWI and ADC to confirm MM lesions. Prior work addresses these cues only partially, relying on manual or atlas-based bone ROI generation and simple ADC channel fusion. Inspired by this strategy, our method adopts a two-stage design: Stage 1 learns a bone ROI generation model from ADC without dedicated bone labels, and Stage 2 uses AMU-Net for anatomy-guided multimodal lesion segmentation.

Figure 2: Overview of the proposed two-stage framework for MM lesion segmentation on WB-DWI. Stage I generates bone ROI maps from ADC images using a 3D nnUNet ensemble. Stage II uses AMU-Net to perform ROI-guided lesion segmentation from WB-DWI with ADC assistance, via encoder-level ROI-gated cross-attention and a decoder-level asymmetric gate.

Figure 3: Demographic and clinical characteristics of the internal and external cohorts: (a) age distribution by center, (b) sex distribution by center, (c) subtype distribution by center, and (d) age distribution stratified by subtype and center. Blue and green denote the internal and external cohorts, respectively. Violin plots show density distributions with embedded boxplots.

Figure 4: Task-related data characteristics of the internal cohort: (a) lesion burden, (b) lesion sparsity, (c) DWI signal distributions by tissue type, and (d) ADC distributions by tissue type. Green and blue denote focal and diffuse cases in (a) and (b), while colors in (c) and (d) indicate lesion, marrow, and background tissues. Violin plots show density distributions with embedded boxplots, and dots represent individual observations.

Model	Type	Params	Dice (%) $\uparrow$	IoU (%) $\uparrow$	HD95 $\downarrow$	Precision (%) $\uparrow$	Recall (%) $\uparrow$
nnU-Net [ 16 ]	3D CNN	$\sim$ 30M	$73.13\pm 1.5$	$63.24\pm 1.3$	$11.8\pm 1.5$	$\mathbf{80.0}$	76.3
VISTA3D [ 13 ]		$\sim$ 30M	$72.40\pm 0.4$	$61.43\pm 0.7$	$12.6\pm 0.5$	73.0	$\mathbf{82.3}$
SegResNet [ 21 ]		$\sim$ 19M	$67.04\pm 0.7$	$56.00\pm 0.7$	$16.3\pm 0.8$	67.3	76.7
U-Net [ 25 ]		$\sim$ 5M	$64.92\pm 1.4$	$52.93\pm 1.5$	$16.5\pm 0.5$	68.3	73.5
DynUNet [ 7 ]		$\sim$ 5M	$63.81\pm 1.7$	$51.93\pm 1.9$	$16.7\pm 1.3$	65.8	74.0
AttentionUNet [ 23 ]		$\sim$ 5M	$63.29\pm 0.5$	$51.45\pm 0.8$	$16.2\pm 0.7$	65.5	76.1

Table 1: Internal comparison with baseline models under the same 5-fold evaluation protocol. Best results are shown in bold , and second-best results are shown in blue .

Model	Type	Params	Dice (%) $\uparrow$	IoU (%) $\uparrow$	HD95 $\downarrow$	Precision (%) $\uparrow$	Recall (%) $\uparrow$
nnU-Net [ 16 ]	3D CNN	$\sim$ 30M	$28.70\pm 1.3$	$18.59\pm 1.1$	70.29 $\pm$ 0.82	32.31	35.16
VISTA3D [ 13 ]		$\sim$ 30M	$30.83\pm 1.1$	$21.26\pm 0.9$	61.49 $\pm$ 0.79	26.44	48.78
SegResNet [ 21 ]		$\sim$ 19M	$40.65\pm 0.9$	$29.96\pm 0.8$	53.41 $\pm$ 5.02	44.79	44.93
U-Net [ 25 ]		$\sim$ 5M	$39.64\pm 2.7$	$28.97\pm 2.5$	63.98 $\pm$ 3.02	36.99	53.28
DynUNet [ 7 ]		$\sim$ 5M	$39.42\pm 2.4$	$28.55\pm 2.2$	62.20 $\pm$ 4.46	35.43	56.46
AttentionUNet [ 23 ]		$\sim$ 5M	$35.99\pm 1.4$	$25.58\pm 1.2$	69.05 $\pm$ 6.72	32.46	$\mathbf{57.80}$

Table 2: Zero-shot external validation on an independent cohort. Models trained using the same 5-fold protocol were directly tested on the external dataset without any fine-tuning or adaptation. Best results are shown in bold , and second-best results are shown in blue .

Figure 5: ADC distributions of lesion and normal marrow from pooled and per-case perspectives. (a) Pooled across all cases, lesion and normal marrow ADC values substantially overlap, suggesting that absolute ADC values alone are insufficient for disease assessment. (b) Per-case mean and range show that lesion-related abnormalities are more meaningfully assessed relative to the patient’s own marrow background. This supports the role of bone ROI guidance in constraining DWI–ADC interpretation to the skeletal marrow space.

Figure 6: Visualization of ADC-based bone ROI generation. Representative sparse/dense and focal/diffuse cases are shown to illustrate ROI generation across heterogeneous patients. In all cases, the predicted ROI forms anatomically coherent skeletal structures while covering a substantial portion of lesion regions. Although trained with lesion-derived supervision, the ROI generator does not simply fit sparse lesion labels, but instead learns stable bone-related structures for downstream segmentation.

Figure 7: Visual comparison of lesion segmentation results. Representative diffuse and focal cases show that AMU-Net achieves more complete lesion coverage and fewer obvious misses than the compared methods. Zoomed-in views are provided for the boxed regions.

Figure 8: Qualitative comparison between a registration pipeline and the proposed direct bone ROI generation method. The registration-based method often shows spatial mismatch after mapping Dixon-derived bone masks into DWI space, whereas our method produces bone ROIs that better align with lesion-bearing skeletal regions.

Variant	Dice (%) $\uparrow$	IoU (%) $\uparrow$	HD95 $\downarrow$
Structural ablation
Full	76.24 $\pm$ 0.5	65.93 $\pm$ 1.0	10.1 $\pm$ 0.7
w/o Enc-XAttn	73.45 $\pm$ 0.8	63.52 $\pm$ 1.1	12.5 $\pm$ 0.5
w/o Dec-Gate	74.87 $\pm$ 0.7	64.74 $\pm$ 0.9	11.0 $\pm$ 0.5
w/o Both	73.13 $\pm$ 1.5	63.24 $\pm$ 1.3	11.8 $\pm$ 1.5
ROI guidance ablation

Table 3: Ablation results of Stage-II components in AMU-Net. Enc-XAttn: encoder-level ROI-gated cross-attention; Dec-Gate: decoder-level asymmetric gate.

Metric	Registration Method	Our Generation Method
Speed per case	$\sim$ 43.5 s	$\sim$ 4.2 s
Bone–Lesion Dice	$23.04\pm 13.07$	62.34 $\pm$ 26.96
Lesion recall	22.6%	70.9%

Table 4: Comparison between a registration pipeline and the proposed direct ADC-based bone ROI generation method. Best results are shown in bold .

Medical Image Segmentation

Medical image segmentation delineates organs, lesions, and other structures in scans to support measurement, diagnosis, treatment planning, and research.

See all 786 papers

End-to-end prostate lesion segmentation and detection from undersampled diffusion-weighted MRI
Oct 24, 2026Lukas T. Rotkopf, Marco Schlimbach, Kevin Kröninger +4Medical Image Analysis Medical Image Segmentation
Institute for AI in Medicine (IKIM), University Hospital Essen, Girardetstraße 2, 45131, Essen, Germany. · Department of Physics, Technical University Dortmund, Otto-Hahn-Straße 4a, 44227, Dortmund, Germany. · Division of Radiology, German Cancer Research Center (DKFZ), 69120, Heidelberg, Germany. +6
Evaluating ADC-only deep learning pipelines for breast cancer detection and segmentation using standalone diffusion-weighted MRI
Aug 18, 2026Pablo García Marcos, Paula Puerta Gonzĺez, Guillermo Lorenzo +4Breast Cancer Medical Image Segmentation
Computer Sciences Department, University of Oviedo, Asturias, Spain · Biomedical Engineering Center (BME), University of Oviedo, Asturias, Spain · Electrical Engineering Department, University of Oviedo, Asturias, Spain +6
Deep Learning Segmentation of Diffusion-Weighted MRI Acute Ischaemic Stroke: A Pragmatic Evaluation Across Three Datasets
Aug 26, 2026Atle Bjørnerud, Till Schellhorn, Thor H. Skattør +4Medical Image Segmentation U-Net
Computational Radiology & Artificial Intelligence Unit, Department of Physics and Computational Radiology, Oslo University Hospital, Oslo, Norway · Center for Lifespan Changes in Brain and Cognition, Department of Psychology, University of Oslo, Oslo, Norway · Division of Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway +5

Oct 24, 2026J/K move · Enter open · S save

End-to-end prostate lesion segmentation and detection from undersampled diffusion-weighted MRI

Lukas T. Rotkopf, Marco Schlimbach, Kevin Kröninger, Stefan O. Schoenberg+3

Institute for AI in Medicine (IKIM), University Hospital Essen, Girardetstraße 2, 45131, Essen, Germany. · Department of Physics, Technical University Dortmund, Otto-Hahn-Straße 4a, 44227, Dortmund, Germany. · Division of Radiology, German Cancer Research Center (DKFZ), 69120, Heidelberg, Germany.+6

Medical Image Analysis Medical Image Segmentation Diffusion MRI Accelerated MRI Reconstruction

Open paper AskPDF arXiv

Automated prostate lesion detection from diffusion-weighted MR imaging (DWI) conventionally relies on reconstructed images and does not fully use all acquisition information. We present an end-to-end network that uses complex multicoil DWI data and coil sensitivity information for prostate lesion segmentation. We included 269 fastMRI prostate examinations with additional voxel-level lesion annotations. Of these, 64 examinations contained PIRADS 4 or 5 lesions. The proposed network, VarNet-Seg, consists of a k-space variational network (VarNet) with six variational cascades and an image-space 3D segmentation U-Net. Comparator models used the same segmentation architecture with fully coil-combined images (CC-Seg), fixed root-sum-of-squares reconstruction (RSS-Seg), or fixed SENSE reconstruction (SENSE-Seg). We performed ablations with end-to-end training and data consistency (DC) enabled or disabled. We evaluated lesion segmentation using Dice and patient-level detection using AUROC in five-fold cross-validation with full sampling, simulated undersampling, added noise, or both. VarNet-Seg achieved the highest Dice at every undersampling level. At eight-fold acceleration, Dice was 0.22 for VarNet-Seg, 0.14 for SENSE-Seg and 0.16 for VarNet-Seg (no DC). The paired difference from SENSE-Seg was 0.08 (95% CI, 0.04-0.11). With full sampling, VarNet-Seg and RSS-Seg performed similarly (0.40 vs 0.39). Patient-level AUROC decreased from 0.75 with full sampling to 0.63 at eight-fold acceleration; corresponding RSS-Seg AUROCs were 0.76 and 0.64. The proposed multicoil network improved prostate lesion segmentation under simulated undersampling compared to conventional models. Matched ablations supported contributions from both end-to-end training and data consistency. Prospective evaluation is needed to establish whether this approach can be used clinically in accelerated prostate DWI.

Abstract

Medical Image Segmentation

Explore similar work

End-to-end prostate lesion segmentation and detection from undersampled diffusion-weighted MRI

Evaluating ADC-only deep learning pipelines for breast cancer detection and segmentation using standalone diffusion-weighted MRI

Deep Learning Segmentation of Diffusion-Weighted MRI Acute Ischaemic Stroke: A Pragmatic Evaluation Across Three Datasets