Subnetwork representation learning for discovering network biomarkers in predicting lymph node metastasis in early oral cancer.
Ontology highlight
ABSTRACT:
SUBMITTER: Kim M
PROVIDER: S-EPMC8671417 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundThe aim of this study is to design a deep learning (DL) model to preoperatively predict the occurrence of central lymph node metastasis (CLNM) in patients with papillary thyroid microcarcinoma (PTMC).MethodsThis research collected preoperative ultrasound (US) images and clinical factors of 611 PTMC patients. The clinical factors were analyzed using multivariate regression. Then, a DL model based on US images and clinical factors was developed to preoperatively predict CLNM. The model's efficacy was evaluated using the receiver operating characteristic (ROC) curve, along with accuracy, sensitivity, specificity, and the F1 score.ResultsThe multivariate analysis indicated an independent correlation factors including age ≥55 (OR = 0.309, p < 0.001), tumor diameter (OR = 2.551, p = 0.010), macrocalcifications (OR = 1.832, p = 0.002), and capsular invasion (OR = 1.977, p = 0.005). The suggested DL model utilized US images achieved an average area under the curve (AUC) of 0.65, slightly outperforming the model that employed traditional clinical factors (AUC = 0.64). Nevertheless, the model that incorporated both of them did not enhance prediction accuracy (AUC = 0.63).ConclusionsThe suggested approach offers a reference for the treatment and supervision of PTMC. Among three models used in this study, the deep model relied generally more on image modalities than the data modality of clinic records when making the predictions.
| S-EPMC10984175 | biostudies-literature
Project description:Oral microbiota can alter cancer susceptibility and progression by modulating metabolism and inflammation. We assessed the association between the oral microbiome and lymph node (LN) metastasis in oral squamous cell carcinoma (OSCC). We collected a total of 54 saliva samples from patients with OSCC before surgery. LN metastasis was assessed based on postoperative pathological examination. We used QIIME2, linear discriminant analysis effect size (LEfSe), and PICRUSt2 methods to analyze microbial dysbiosis. A random forest classifier was used to assess whether the oral microbiome could predict LN metastasis. Among the 54 OSCC samples, 20 had LN metastasis, and 34 had no evidence of metastasis. There was a significant difference in β-diversity between the metastasis and no metastasis groups. Through LEfSe analysis, the metastasis group was enriched in the genera Prevotella, Stomatobaculum, Bifidobacterium, Peptostreptococcaceae, Shuttleworthia and Finegoldia. Pathways related to signal peptidase II were predominant in the no metastasis group. The RF model showed a modestly high accuracy for predicting metastasis. Differences in microbial community composition and functions were observed in the oral microbiome of patients with OSCC with and without LN metastasis. However, the finding that specific taxa may be associated with LN metastasis should be verified in a further prospective study.
| S-EPMC8633319 | biostudies-literature
Project description:ObjectivesTo develop and validate a deep learning (DL)-based primary tumor biopsy signature for predicting axillary lymph node (ALN) metastasis preoperatively in early breast cancer (EBC) patients with clinically negative ALN.MethodsA total of 1,058 EBC patients with pathologically confirmed ALN status were enrolled from May 2010 to August 2020. A DL core-needle biopsy (DL-CNB) model was built on the attention-based multiple instance-learning (AMIL) framework to predict ALN status utilizing the DL features, which were extracted from the cancer areas of digitized whole-slide images (WSIs) of breast CNB specimens annotated by two pathologists. Accuracy, sensitivity, specificity, receiver operating characteristic (ROC) curves, and areas under the ROC curve (AUCs) were analyzed to evaluate our model.ResultsThe best-performing DL-CNB model with VGG16_BN as the feature extractor achieved an AUC of 0.816 (95% confidence interval (CI): 0.758, 0.865) in predicting positive ALN metastasis in the independent test cohort. Furthermore, our model incorporating the clinical data, which was called DL-CNB+C, yielded the best accuracy of 0.831 (95%CI: 0.775, 0.878), especially for patients younger than 50 years (AUC: 0.918, 95%CI: 0.825, 0.971). The interpretation of DL-CNB model showed that the top signatures most predictive of ALN metastasis were characterized by the nucleus features including density (p = 0.015), circumference (p = 0.009), circularity (p = 0.010), and orientation (p = 0.012).ConclusionOur study provides a novel DL-based biomarker on primary tumor CNB slides to predict the metastatic status of ALN preoperatively for patients with EBC.
| S-EPMC8551965 | biostudies-literature
Project description:BackgroundPresence of lymph node metastasis (LNM) influences prognosis and clinical decision-making in colorectal cancer. However, detection of LNM is variable and depends on a number of external factors. Deep learning has shown success in computational pathology, but has struggled to boost performance when combined with known predictors.MethodsMachine-learned features are created by clustering deep learning embeddings of small patches of tumor in colorectal cancer via k-means, and then selecting the top clusters that add predictive value to a logistic regression model when combined with known baseline clinicopathological variables. We then analyze performance of logistic regression models trained with and without these machine-learned features in combination with the baseline variables.ResultsThe machine-learned extracted features provide independent signal for the presence of LNM (AUROC: 0.638, 95% CI: [0.590, 0.683]). Furthermore, the machine-learned features add predictive value to the set of 6 clinicopathologic variables in an external validation set (likelihood ratio test, p < 0.00032; AUROC: 0.740, 95% CI: [0.701, 0.780]). A model incorporating these features can also further risk-stratify patients with and without identified metastasis (p < 0.001 for both stage II and stage III).ConclusionThis work demonstrates an effective approach to combine deep learning with established clinicopathologic factors in order to identify independently informative features associated with LNM. Further work building on these specific results may have important impact in prognostication and therapeutic decision making for LNM. Additionally, this general computational approach may prove useful in other contexts.
| S-EPMC10125969 | biostudies-literature
Project description:BackgroundThe authors aimed to create a novel model to predict lymphatic metastasis in thymic epithelial tumors.MethodsData of 1018 patients were collected from the Surveillance, Epidemiology, and End Results database from 2004 to 2015. To construct a nomogram, the least absolute shrinkage and selection operator (LASSO) regression model was used to select candidate features of the training cohort from 2004 to 2013. A simple model called the Lymphatic Node Metastasis Risk Scoring System (LNMRS) was constructed to predict lymphatic metastasis. Using patients from 2014 to 2015 as the validation cohort, the predictive performance of the model was determined by receiver operating characteristic (ROC) curves.ResultsThe LASSO regression model showed that age, extension, and histology type were significantly associated with lymph node metastasis, which were used to construct the nomogram. Through analysis of the area under the curve (AUC), the nomogram achieved a AUC value of 0.80 (95 % confidence interval [Cl] 0.75-0.85) in the training cohort and 0.82 (95 % Cl 0.70-0.93) in the validation cohort, and had closed calibration curves. Based on the nomogram, the authors constructed the LNMRS model, which had an AUC of 0.80 (95 % Cl 0.75-0.85) in the training cohort and 0.82 (95% Cl 0.70-0.93) in the validation cohort. The ROC curves indicated that the LNMRS had excellent predictive performance for lymph node metastasis.ConclusionThis study established a nomogram for predicting lymph node metastasis. The LNMRS model, constructed to predict lymphatic involvement of patients, was more convenient than the nomogram.
| S-EPMC8677650 | biostudies-literature
Project description:Lymph nodes are important peripheral immune organs in which numerous important immune responses occur. During the process of lymphatic metastasis, lymph nodes are also sites through which tumor cells must pass. Therefore, it is essential to develop a drug delivery system that can specifically transfer immunostimulatory medicine into lymph nodes to block lymphatic metastasis. Here, we developed a nucleic acid drug delivery system containing cationic agarose (C-agarose) and CpG oligodeoxynucleotides. C-agarose has a high affinity for Siglec-1 on the surface of lymph node sinus macrophages, which have a high specificity for targeting lymph nodes. Subcutaneous implantation of C-agarose+CpG gel caused the accumulation of CpG in the lymph node sinus macrophages and generated antitumor immune responses in the lymph node. C-agarose+CpG gel treatment decreased the metastasis size in the tumor-draining lymph node (TDLN) and lung metastatic nodules and suppressed tumor growth in both a mouse 4T1 breast cancer model and a B16F10 melanoma model. On this basis, this study proposes a nonsurgical invasive lymph node targeting immunotherapy concept that may provide a new approach for antitumor metastasis.
| S-EPMC6529739 | biostudies-literature
Project description:BackgroundHepatic alveolar echinococcosis (AE) is a severe zoonotic parasitic disease, and accurate preoperative prediction of lymph node (LN) metastasis in AE patients is crucial for disease management, but it remains an unresolved challenge. The aim of this study was to establish a radiomics model for the preoperative prediction of LN metastasis in hepatic AE patients.MethodsA total of 100 hepatic AE patients who underwent hepatectomy and hepatoduodenal ligament LN dissection at Qinghai Provincial People's Hospital between January 2016 and August 2023 were included in the study. The patients were randomly divided into a training set and a validation set at an 8:2 ratio. Radiomic features were extracted from three-dimensional images of the hepatoduodenal ligament LNs delineated on arterial phase computed tomography (CT) scans of hepatic AE patients. Least absolute shrinkage and selection operator (LASSO) regression was applied for data dimensionality reduction and feature selection. Multivariate logistic regression analysis was performed to develop a prediction model, and the predictive performance of the model was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA).ResultsA total of 7 radiomics features associated with LN status were selected using LASSO regression. The classification performances of the training set and validation set were consistent, with area under the operating characteristic curve (AUC) values of 0.928 and 0.890, respectively. The model also demonstrated good stability in subsequent validation.ConclusionIn this study, we established and evaluated a radiomics-based prediction model for LN metastasis in patients with hepatic AE using CT imaging. Our findings may provide a valuable reference for clinicians to determine the occurrence of LN metastasis in hepatic AE patients preoperatively, and help guide the implementation of individualized surgical plans to improve patient prognosis.
| S-EPMC11304587 | biostudies-literature
Project description:Lymph node metastasis is the most reliable indicator of a poor prognosis for patients with oral tongue cancers. Currently, there are no biomarkers to predict whether a cancer will spread in the future if it has not already spread at the time of diagnosis. The aim of this study was to quantitatively profile the proteomes of extracellular vesicles (EVs) isolated from blood samples taken from patients with oral tongue squamous cell carcinoma with and without lymph node involvement and non-cancer controls. EVs were enriched using size exclusion chromatography (SEC) from pooled plasma samples of patients with non-nodal and nodal oral tongue squamous cell carcinoma (OTSCC) and non-cancer controls. Protein cargo was quantitatively profiled using isobaric labelling (iTRAQ) and two-dimensional high-performance liquid chromatography followed by tandem mass spectrometry. We identified 208 EV associated proteins and, after filtering, generated a short list of 136 proteins. Over 85% of the EV-associated proteins were associated with the GO cellular compartment term "extracellular exosome". Comparisons between non-cancer controls and oral tongue squamous cell carcinoma with and without lymph node involvement revealed 43 unique candidate EV-associated proteins with deregulated expression patterns. The shortlisted EV associated proteins described here may be useful discriminatory biomarkers for differentiating OTSCC with and without nodal disease or non-cancer controls.
| S-EPMC8468562 | biostudies-literature
Project description:Lymph node involvement increases the risk of breast cancer recurrence. An accurate non-invasive assessment of nodal involvement is valuable in cancer staging, surgical risk, and cost savings. Radiomics has been proposed to pre-operatively predict sentinel lymph node (SLN) status; however, radiomic models are known to be sensitive to acquisition parameters. The purpose of this study was to develop a prediction model for preoperative prediction of SLN metastasis using deep learning-based (DLB) features and compare its predictive performance to state-of-the-art radiomics. Specifically, this study aimed to compare the generalizability of radiomics vs DLB features in an independent test set with dissimilar resolution. Dynamic contrast-enhancement images from 198 patients (67 positive SLNs) were used in this study. Of these subjects, 163 had an in-plane resolution of 0.7 × 0.7 mm2, which were randomly divided into a training set (approximately 67%) and a validation set (approximately 33%). The remaining 35 subjects with a different in-plane resolution (0.78 × 0.78 mm2) were treated as independent testing set for generalizability. Two methods were employed: (1) conventional radiomics (CR), and (2) DLB features which replaced hand-curated features with pre-trained VGG-16 features. The threshold determined using the training set was applied to the independent validation and testing dataset. Same feature reduction, feature selection, model creation procedures were used for both approaches. In the validation set (same resolution as training), the DLB model outperformed the CR model (accuracy 83% vs 80%). Furthermore, in the independent testing set of the dissimilar resolution, the DLB model performed markedly better than the CR model (accuracy 77% vs 71%). The predictive performance of the DLB model outperformed the CR model for this task. More interestingly, these improvements were seen particularly in the independent testing set of dissimilar resolution. This could indicate that DLB features can ultimately result in a more generalizable model.
| S-EPMC8901808 | biostudies-literature
Project description:ObjectivesTo determine whether ultrasound radiomics can be used to distinguish axillary lymph nodes (ALN) metastases in breast cancer based on ALN imaging.MethodsA total of 147 breast cancer patients with 41 non-metastatic lymph nodes and 109 metastatic lymph nodes were divided into a training set (105 ALN) and a validation set (45 ALN). Radiomics features were extracted from ultrasound images and a radiomics signature (RS) was built. The Intraclass correlation coefficients (ICCs), Spearman correlation analysis, and least absolute shrinkage and selection operator (LASSO) methods were used to select the ALN status-related features. All images were assessed by two radiologists with at least 10 years of experience in ALN ultrasound examination. The performance levels of the model and radiologists in the training and validation subgroups were then evaluated and compared.ResultRadiomics signature accurately predicted the ALN status, achieved an area under the receiver operator characteristic curve of 0.929 (95%CI, 0.881-0.978) and area under curve(AUC) of 0.919 (95%CI, 95%CI, 0.841-0.997) in training and validation cohorts respectively. The radiomics model performed better than two experts' prediction of ALN status in both cohorts (P<0.05). Besides, prediction in subgroups based on baseline clinicopathological information also achieved good discrimination performance, with an AUC of 0.937, 0.918, 0.885, 0.930, and 0.913 in HR+/HER2-, HER2+, triple-negative, tumor sized ≤ 3cm and tumor sized>3 cm, respectively.ConclusionThe radiomics model demonstrated a good ability to predict ALN status in patients with breast cancer, which might provide essential information for decision-making.
| S-EPMC10641324 | biostudies-literature