InterSliceBoost: Identifying Tissue Layers in Three-dimensional Ultrasound Images for Chronic Lower Back Pain (cLBP) Assessment

2025-03-25

AI Summary

Manual annotation of multi-layer tissue segmentation in 3D ultrasound for chronic low back pain (cLBP) assessment is prohibitively expensive and impractical for full-volume labeling. Method: We propose a cross-slice generative augmentation framework leveraging only partial slice annotations (33% of slices). A novel residual encoder-driven generator models inter-slice differences between adjacent image–mask pairs to synthesize high-fidelity training samples. The generator and segmentation network are jointly optimized, and a partial-labeling–driven cross-slice augmentation strategy is introduced to enforce semantic consistency across slices. Contribution/Results: Evaluated on 76 3D B-mode ultrasound volumes, our method achieves a mean Dice score of 80.84% across six tissue classes, with all layers exceeding 61%—significantly outperforming the fully annotated baseline (p < 0.05). To our knowledge, this is the first approach to achieve cross-slice semantic-consistent augmentation under low-labeling-ratio conditions, establishing a new weakly supervised paradigm for medical image segmentation.

Technology Category

• Computer Vision: Segmentation
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Available studies on chronic lower back pain (cLBP) typically focus on one or a few specific tissues rather than conducting a comprehensive layer-by-layer analysis. Since three-dimensional (3-D) images often contain hundreds of slices, manual annotation of these anatomical structures is both time-consuming and error-prone. We aim to develop and validate a novel approach called InterSliceBoost to enable the training of a segmentation model on a partially annotated dataset without compromising segmentation performance. The architecture of InterSliceBoost includes two components: an inter-slice generator and a segmentation model. The generator utilizes residual block-based encoders to extract features from adjacent image-mask pairs (IMPs). Differential features are calculated and input into a decoder to generate inter-slice IMPs. The segmentation model is trained on partially annotated datasets (e.g., skipping 1, 2, 3, or 7 images) and the generated inter-slice IMPs. To validate the performance of InterSliceBoost, we utilized a dataset of 76 B-mode ultrasound scans acquired on 29 subjects enrolled in an ongoing cLBP study. InterSliceBoost, trained on only 33% of the image slices, achieved a mean Dice coefficient of 80.84% across all six layers on the independent test set, with Dice coefficients of 73.48%, 61.11%, 81.87%, 95.74%, 83.52% and 88.74% for segmenting dermis, superficial fat, superficial fascial membrane, deep fat, deep fascial membrane, and muscle. This performance is significantly higher than the conventional model trained on fully annotated images (p<0.05). InterSliceBoost can effectively segment the six tissue layers depicted on 3-D B-model ultrasound images in settings with partial annotations.

Problem

• Automating tissue layer segmentation in 3D ultrasound for cLBP
• Reducing manual annotation workload in medical image analysis
• Enhancing segmentation accuracy with partial annotated datasets

Innovation

• InterSliceBoost trains models on partially annotated datasets
• Uses inter-slice generator for feature extraction
• Achieves high segmentation accuracy with fewer annotations