scholarly journals PET/MR fusion texture analysis for the clinical outcome prediction in soft-tissue sarcoma

2022 ◽  
Vol 22 (1) ◽  
Author(s):  
Wenzhe Zhao ◽  
Xin Huang ◽  
Geliang Wang ◽  
Jianxin Guo

Abstract Background Various fusion strategies (feature-level fusion, matrix-level fusion, and image-level fusion) were used to fuse PET and MR images, which might lead to different feature values and classification performance. The purpose of this study was to measure the classification capability of features extracted using various PET/MR fusion methods in a dataset of soft-tissue sarcoma (STS). Methods The retrospective dataset included 51 patients with histologically proven STS. All patients had pre-treatment PET and MR images. The image-level fusion was conducted using discrete wavelet transformation (DWT). During the DWT process, the MR weight was set as 0.1, 0.2, 0.3, 0.4, …, 0.9. And the corresponding PET weight was set as 1- (MR weight). The fused PET/MR images was generated using the inverse DWT. The matrix-level fusion was conducted by fusing the feature calculation matrix during the feature extracting process. The feature-level fusion was conducted by concatenating and averaging the features. We measured the predictive performance of features using univariate analysis and multivariable analysis. The univariate analysis included the Mann-Whitney U test and receiver operating characteristic (ROC) analysis. The multivariable analysis was used to develop the signatures by jointing the maximum relevance minimum redundancy method and multivariable logistic regression. The area under the ROC curve (AUC) value was calculated to evaluate the classification performance. Results By using the univariate analysis, the features extracted using image-level fusion method showed the optimal classification performance. For the multivariable analysis, the signatures developed using the image-level fusion-based features showed the best performance. For the T1/PET image-level fusion, the signature developed using the MR weight of 0.1 showed the optimal performance (0.9524(95% confidence interval (CI), 0.8413–0.9999)). For the T2/PET image-level fusion, the signature developed using the MR weight of 0.3 showed the optimal performance (0.9048(95%CI, 0.7356–0.9999)). Conclusions For the fusion of PET/MR images in patients with STS, the signatures developed using the image-level fusion-based features showed the optimal classification performance than the signatures developed using the feature-level fusion and matrix-level fusion-based features, as well as the single modality features. The image-level fusion method was more recommended to fuse PET/MR images in future radiomics studies.

2020 ◽  
Vol 50 (10) ◽  
pp. 1168-1174
Author(s):  
Toshihide Hirai ◽  
Hiroshi Kobayashi ◽  
Tomotake Okuma ◽  
Yuki Ishibashi ◽  
Masachika Ikegami ◽  
...  

Abstract Background It is unknown whether sarcopenia influences treatment outcome in patients with soft tissue sarcoma. Herein, we aimed to elucidate the impact of sarcopenia on sarcoma treatment. Methods A total of 163 soft tissue sarcoma patients were included. Skeletal muscle measures were calculated using computed tomography images. Skeletal muscle area (SMA) and density (SMD) at the L3 level were extracted, and SMA was normalized by height as skeletal muscle index (SMI). The skeletal muscle gauge (SMG) was calculated by multiplying SMD × SMI. The relationship of skeletal muscle measures and clinical factors to wound complications and prognosis was evaluated, and classification and regression tree (CART) analysis was used to develop classification models for risk groups of surgical wound complications. Results Thirty-three patients developed wound complications. In univariate analysis, age (P = 0.0022), tumour location of adductor compartment of the thigh (P = 0.0019), operating time (P = 0.010), blood loss (P = 0.030), SMD (P = 0.0004) and SMG (P = 0.0001) were significantly correlated with complications. In multivariate analysis, lower SMG was an independent risk factor (P = 0.031, OR = 3.27). CART analysis classified three risk groups of surgical wound complications by SMG, age, tumour location and operating time, and area under the receiver operating characteristic curve (AUROCC) was 0.75. SMG was not associated with prognosis in univariate analysis (P = 0.15). Conclusions The SMG does not affect overall survival but predicts surgical wound complications.


Cancers ◽  
2021 ◽  
Vol 13 (5) ◽  
pp. 1175
Author(s):  
Yanhong Su ◽  
Panagiotis Tsagkozis ◽  
Andri Papakonstantinou ◽  
Nicholas P. Tobin ◽  
Okan Gultekin ◽  
...  

Checkpoint inhibitors are slowly being introduced in the care of specific sarcoma subtypes such as undifferentiated pleomorphic sarcoma, alveolar soft part sarcoma, and angiosarcoma even though formal indication is lacking. Proper biomarkers to unravel potential immune reactivity in the tumor microenvironment are therefore expected to be highly warranted. In this study, intratumoral spatial cross presentation was investigated as a novel concept where immune cell composition in the tumor microenvironment was suggested to act as a proxy for immune surveillance. Double immunohistochemistry revealed a prognostic role of direct spatial interactions between CD11c+ antigen-presenting cells (APCs) and CD8+ cells in contrast to each marker alone in a soft tissue sarcoma (STS) cohort of 177 patients from the Karolinska University Hospital (MFS p = 0.048, OS p = 0.025). The survival benefit was verified in multivariable analysis (MFS p = 0.012, OS p = 0.004). Transcriptomics performed in the TCGA sarcoma cohort confirmed the prognostic value of combining CD11c with CD8 (259 patients, p = 0.005), irrespective of FOXP3 levels and in a CD274 (PD-LI)-rich tumor microenvironment. Altogether, this study presents a histopathological approach to link immune surveillance and patient survival in STS. Notably, spatial cross presentation as a prognostic marker is distinct from therapy response-predictive biomarkers such as immune checkpoint molecules of the PD-L1/PD1 pathway.


2009 ◽  
Vol 27 (15_suppl) ◽  
pp. e21508-e21508
Author(s):  
G. F. Almeida ◽  
S. A. Siqueira ◽  
G. Castro ◽  
I. L. Snitcovsky ◽  
E. H. Akaishi ◽  
...  

e21508 Background: Ezrin is a member of the ERM (ezrin, radixin, moesin) protein family and links F-actin to the cell membrane, and it is involved in regulating growth and metastatic behaviour of cancer cells. Ezrin expression is associated with tumor progression and metastasis in several cancers, including sarcomas. This study evaluated ezrin expression as a predictor of response to chemotherapy and as a prognostic factor in soft tissue sarcoma (STS) patients (pts) treated in a dose-dense schedule protocol. Methods: 21 chemotherapy-naïve pts diagnosed with high-grade STS, not candidates for a limb-sparing surgery, were enrolled in a prospective phase II study of a sequential and dose-dense regimen consisting of doxorubicin 30 mg/m2 d1–3 q2w and ifosfamide 2.5 g/m2 d1–5 q3w, 3 cycles each, with G-CSF support. Ezrin expression was analyzed by immunohistochemistry on slides from formalin-fixed, paraffin-embedded, primary tumor biopsy blocks, with the anti-ezrin antibody clone AB-1(3C12) (NeoMarkers). Cytoplasmic immunostaining in more than 10% of tumor cells was considered as positive. Ezrin expression was correlated with response rate, progression-free (PFS) and overall survival (OS). Results: Leiomyo-, synovial and sarcoma NOS were the most frequent subtypes (5 pts each). 13 out of 21 pts (62%) presented distant metastasis. Protocol was halted after 3 toxic deaths. Three pts achieved partial response (RR 14%). With a median follow-up of 11.7 mo, median PFS and OS were 8.9 and 20.1 mo, respectively. In univariate analysis, OS was higher for synovial sarcoma pts (not reached vs. 14.2 mo, HR 0.0, 95%CI 0.06–0.82, p=0.02), and for those aged 45 y or less (20.1 vs. 4.2 mo, HR 0.30, 95%CI 0.03–0.88, p=0.04). Ezrin expression was available for 20 pts: it was positive in 9 pts (45%) and negative in 11 pts (55%). Ezrin expression was not related to tumor response (p = 0.40), and no significant association was detected between ezrin expression and median PFS or OS. Conclusions: Expression of ezrin was not a useful marker to predict outcomes in STS pts treated with dose-dense doxorubicin- and ifosfamide-based chemotherapy. No significant financial relationships to disclose.


2013 ◽  
Vol 273 ◽  
pp. 260-263
Author(s):  
Ling Li Jiang ◽  
Hua Kui Yin ◽  
Si Wen Tang

Bearing failure is one of the foremost causes of breakdowns in rotating machinery and such failure can be catastrophic. Fault diagnosis is critical to maintaining the normal operation of the bearings. This paper proposes feature-level fusion method for rolling bearing fault diagnosis. Features are extracted from eight vibration signals to constitute a fusion vector. SVM is used for pattern recognition. The case study results show that the proposed method is useful for rolling bearing fault diagnosis.


Author(s):  
Joanna Szkandera ◽  
Armin Gerger ◽  
Bernadette Liegl-Atzwanger ◽  
Michael Stotz ◽  
Hellmut Samonigg ◽  
...  

AbstractRecent evidence indicates toward a role of uric acid (UA) as a potential antioxidant. Elevated UA levels were shown to be associated with better survival in various malignancies. The aim of the present study was to evaluate the prognostic relevance of pre-operative UA levels on cancer-specific survival (CSS) in soft-tissue sarcoma (STS) patients who underwent curative surgical resection.Three hundred and fifty-seven patients with STS were included in the study. Pre-operative serum UA level was measured using an enzymatic colorimetric assay. The effect of UA levels on CSS was analyzed using Kaplan-Meier curves. To further evaluate the prognostic impact of UA levels, univariate and multivariate Cox proportional models were calculated.Among the 357 STS patients, cancer-related deaths occurred in 20 (24.7%) of 81 patients with a serum UA level <279.6 µmol/L and in 36 (13%) of 276 patients with a UA level ≥279.6 µmol/L. In univariate analysis, elevated UA levels were significantly associated with increased CSS in STS patients [hazard ratio (HR) 0.44, 95% confidence interval (CI) 0.26–0.77, p=0.004]. Furthermore, elevated UA levels remain a significant factor for better CCS in multivariate analysis (HR 0.42, 95% CI 0.23–0.75, p=0.003).Our study is the first one to demonstrate that higher UA levels are associated with positive clinical outcome in STS patients. UA levels are a simple and cost-effective test for the assessment of the prognosis of STS patients.


2021 ◽  
Vol 13 (7) ◽  
pp. 1323
Author(s):  
Yingying Kong ◽  
Biyuan Yan ◽  
Yanjuan Liu ◽  
Henry Leung ◽  
Xiangyang Peng

In terms of land cover classification, optical images have been proven to have good classification performance. Synthetic Aperture Radar (SAR) has the characteristics of working all-time and all-weather. It has more significant advantages over optical images for the recognition of some scenes, such as water bodies. One of the current challenges is how to fuse the benefits of both to obtain more powerful classification capabilities. This study proposes a classification model based on random forest with the conditional random fields (CRF) for feature-level fusion classification using features extracted from polarized SAR and optical images. In this paper, feature importance is introduced as a weight in the pairwise potential function of the CRF to improve the correction rate of misclassified points. The results show that the dataset combining the two provides significant improvements in feature identification when compared to the dataset using optical or polarized SAR image features alone. Among the four classification models used, the random forest-importance_ conditional random fields (RF-Im_CRF) model developed in this paper obtained the best overall accuracy (OA) and Kappa coefficient, validating the effectiveness of the method.


Sign in / Sign up

Export Citation Format

Share Document