{"doi":"10.1002/mp.17931","title":"Intersliceboost: Identifying tissue layers in three‐dimensional ultrasound images for chronic lower back pain (cLBP) assessment","abstract":"BACKGROUND: 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. OBJECTIVES: 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. MATERIALS AND METHODS: 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. An ultrasound operator annotated six anatomical layers across all image slices (n = 18 986), including dermis, superficial fat, superficial fascial membrane (SFM), deep fat, deep fascial membrane (DFM), and muscle. The mean Dice coefficient across all tissue layer was used as the primary performance metric. A paired sample t-test with Bonferroni correction was conducted to assess the significance of performance differences. RESULTS: 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, SFM, deep fat, DFM, and muscle. This performance is significantly higher than the conventional model trained on fully annotated images (p < 0.05). CONCLUSION: InterSliceBoost can effectively segment the six tissue layers depicted on 3-D B-model ultrasound images in settings with partial annotations. CLINICAL RELEVANCE STATEMENT: InterSliceBoost enables tissue layer segmentation with fewer manual annotations. By comprehensively examining each tissue layer, this approach offers deeper insights into the structural and functional characteristics that may contribute to the onset and progression of cLBP.","journal":"Medical Physics","year":2025,"id":570397,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":0,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9161,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2025-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1463595,"name":"M. Cartier","orcid":null,"position":1,"is_corresponding":false},{"id":1321578,"name":"Xiaoyan Zhao","orcid":"0009-0002-7369-9898","position":2,"is_corresponding":false},{"id":1475688,"name":"Pengyu Chen","orcid":"0000-0001-6060-9764","position":3,"is_corresponding":false},{"id":1034648,"name":"Xin Meng","orcid":"0000-0001-7727-6046","position":4,"is_corresponding":false},{"id":818312,"name":"Zhiyu Sheng","orcid":"0000-0002-4209-5939","position":5,"is_corresponding":false},{"id":1463596,"name":"Maryam Satarpour","orcid":null,"position":6,"is_corresponding":false},{"id":1087473,"name":"John M. Cormack","orcid":"0000-0003-2403-6043","position":7,"is_corresponding":false},{"id":820071,"name":"Allison Bean","orcid":"0000-0003-0133-9316","position":8,"is_corresponding":false},{"id":556513,"name":"Ryan Nussbaum","orcid":null,"position":9,"is_corresponding":false},{"id":1463381,"name":"Maya Maurer","orcid":"0000-0002-6483-2872","position":10,"is_corresponding":false},{"id":1463597,"name":"Emily Landis‐Walkenhorst","orcid":null,"position":11,"is_corresponding":false},{"id":274449,"name":"Kang Kim","orcid":"0000-0003-1966-1135","position":12,"is_corresponding":false},{"id":226968,"name":"Ajay D. Wasan","orcid":"0000-0002-6394-6077","position":13,"is_corresponding":false},{"id":288540,"name":"Jiantao Pu","orcid":"0000-0003-2127-5313","position":14,"is_corresponding":false},{"id":1276321,"name":"Zixue Zeng","orcid":null,"position":0,"is_corresponding":true}],"reference_count":50,"raw_metadata":null,"created_at":"2026-07-19T02:57:07.857542Z","pmid":"40664617","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}