RESUMO
OBJECTIVE: Early diagnosis of osteoporosis is crucial to prevent osteoporotic vertebral fracture and complications of spine surgery. We aimed to conduct a hybrid transformer convolutional neural network (HTCNN)-based radiomics model for osteoporosis screening in routine CT. METHODS: To investigate the HTCNN algorithm for vertebrae and trabecular segmentation, 92 training subjects and 45 test subjects were employed. Furthermore, we included 283 vertebral bodies and randomly divided them into the training cohort (n = 204) and test cohort (n = 79) for radiomics analysis. Area receiver operating characteristic curves (AUCs) and decision curve analysis (DCA) were applied to compare the performance and clinical value between radiomics models and Hounsfield Unit (HU) values to detect dual-energy X-ray absorptiometry (DXA) based osteoporosis. RESULTS: HTCNN algorithm revealed high precision for the segmentation of the vertebral body and trabecular compartment. In test sets, the mean dice scores reach 0.968 and 0.961. 12 features from the trabecular compartment and 15 features from the entire vertebral body were used to calculate the radiomics score (rad score). Compared with HU values and trabecular rad-score, the vertebrae rad-score suggested the best efficacy for osteoporosis and non-osteoporosis discrimination (training group: AUC = 0.95, 95%CI 0.91-0.99; test group: AUC = 0.97, 95%CI 0.93-1.00) and the differences were significant in test group according to the DeLong test (p < 0.05). CONCLUSIONS: This retrospective study demonstrated the superiority of the HTCNN-based vertebrae radiomics model for osteoporosis discrimination in routine CT.
Assuntos
Osteoporose , Fraturas por Osteoporose , Humanos , Absorciometria de Fóton , Densidade Óssea , Vértebras Lombares/diagnóstico por imagem , Redes Neurais de Computação , Osteoporose/diagnóstico por imagem , Fraturas por Osteoporose/diagnóstico por imagem , Radiômica , Estudos Retrospectivos , Tomografia Computadorizada por Raios X , Distribuição AleatóriaRESUMO
Lumbar vertebral body cancellous bone location and segmentation is crucial in an automated lumbar spine processing pipeline. Accurate and reliable analysis of lumbar spine image is expected to advantage practical medical diagnosis and population-based analysis of bone strength. However, the design of automated algorithms for lumbar spine processing is demanding due to significant anatomical variations and scarcity of publicly available data. In recent years, convolutional neural network (CNN) and vision transformers (Vits) have been the de facto standard in medical image segmentation. Although adept at capturing global features, the inherent bias of locality and weight sharing of CNN constrains its capacity to model long-range dependency. In contrast, Vits excel at long-range dependency modeling, but they may not generalize well with limited datasets due to the lack of inductive biases inherent to CNN. In this paper, we propose a deep learning-based two-stage coarse-to-fine solution to address the problem of automatic location and segmentation of lumbar vertebral body cancellous bone. Specifically, in the first stage, a Swin-transformer based model is applied to predict the heatmap of lumbar vertebral body centroids. Considering the characteristic anatomical structure of lumbar spine, we propose a novel loss function called LumAnatomy loss, which enforces the order and bend of the predicted vertebral body centroids. To inherit the excellence of CNN and Vits while preventing their respective limitations, in the second stage, we propose an encoder-decoder network to segment the identified lumbar vertebral body cancellous bone, which consists of two parallel encoders, i.e., a Swin-transformer encoder and a CNN encoder. To enhance the combination of CNNs and Vits, we propose a novel multi-scale attention feature fusion module (MSA-FFM), which address issues that arise when fusing features given at different encoders. To tackle the issue of lack of data, we raise the first large-scale lumbar vertebral body cancellous bone segmentation dataset called LumVBCanSeg containing a total of 185 CT scans annotated at voxel level by 3 physicians. Extensive experimental results on the LumVBCanSeg dataset demonstrate the proposed algorithm outperform other state-of-the-art medical image segmentation methods. The data is publicly available at: https://zenodo.org/record/8181250. The implementation of the proposed method is available at: https://github.com/sia405yd/LumVertCancNet.
Assuntos
Osso Esponjoso , Corpo Vertebral , Vértebras Lombares/diagnóstico por imagem , Algoritmos , Região Lombossacral , Processamento de Imagem Assistida por ComputadorRESUMO
Background: Ossification of the posterior longitudinal ligament (OPLL) and ossification of the ligamentum flavum (OLF) are multifactor diseases characterized by progressively ectopic ossification in the spinal ligament. However, the shared ossification mechanism of OPLL and OLF remains to be elucidated. The study aims to investigate the common biomarkers related to ectopic ossification and the potential molecular regulatory mechanism. Methods: Microarray and RNA-seq datasets were obtained from Gene Expression Omnibus (GEO) database. The differentially expressed genes (DEGs) from OPLL and OLF were identified to construct the protein-protein interaction (PPI) network. Furthermore, the hub intersection genes were screened and the diagnostic performance was assessed in the external OLF and OPLL cohorts. We also depicted the landscape of immune cell infiltration and m6A modification meanwhile further estimating the relationship with BMP4. Results: A total of nine up-regulated DEGs and 11 down-regulated DEGs were identified to construct the PPI networks. The integrative bioinformatic analysis defined five hub genes (BMP4, ADAMTS4, HBEGF, IL11, and HAS2) as the common risk biomarkers. Among them, BMP4 was the core target. ROC analysis demonstrated a high diagnostic value of the hub genes. Moreover, activated B cells were recognized as shared differential immune infiltrating cells and significantly associated with BMP4 in OPLL and OLF. Meanwhile, a strong correlation was detected between the expression pattern of the m6A regulator METTL3 and BMP4. Conclusion: This study first identified BMP4 as the shared core biomarker in the development of OPLL and OLF. Activated B cells and m6A writer METTL3 might be involved in the osteogenesis process mediated by BMP4. Our findings provide insights into the pathogenesis in the ossification of the spinal ligament and unveil the potential therapeutic targets.
RESUMO
Background: The relationship between CILP (1184T>C) and IL-1α(+889C/T) polymorphisms and intervertebral disc degeneration (IDD) have been explored in several studies but the results were conflicting. The aim of the study was to evaluate and synthesize the currently available data on the association between CILP (1184T>C) and IL-1α(+889C/T) polymorphisms and susceptibility of phenotype-dependent radiologic IDD (RIDD) and symptomatic intervertebral disk herniation (SIDH). Methods: A computerized literature search was in PubMed, Cochrane Library, Embase, China National Knowledge Infrastructure database, and Web of Science. The pooled results were presented as odds ratios (ORs) with 95% confidence intervals (CIs). Moreover, the false-positive report probability (FPRP) test and trial sequential analysis (TSA) were applied to estimate the significant results. Results: Our evidence demonstrated that IL-1α(+889C/T) was significant associated with RIDD (allele model: OR = 1.34, 95%CI 1.03-1.74, p = 0.029) and SIDH (allele model: OR = 1.28, 95% CI 1.03-1.60, p = 0.028). However, the results were not noteworthy under the FPRP test and TSA analysis. Additionally, CILP (1184T>C) polymorphism was significantly associated with RIDD with adequate evidence (allele model: OR = 1.27, 95% CI 1.09-1.48, p = 0.002) instead of SIDH. Conclusion: The current meta-analysis illustrated firm evidence that CILP (1184T>C) polymorphism was significantly associated with the susceptibility of RIDD. However, the significant associations between IL-1α(+889C/T) and RIDD and SIDH were less credible. Thus, more multi-center studies with diverse populations were required to verify the results.