Prediction Of Lymph Node Metastasis Research Articles

Lymph node metastasis detection using artificial intelligence in T1 colorectal cancer: A comprehensive systematic review.

We systematically reviewed the application of artificial intelligence (AI) in predicting lymph node metastasis (LNM) in T1 colorectal cancer (CRC). Thirteen studies with 8417 patients were included. AI demonstrated high potential in predicting LNM with sensitivity, specificity, and AUC ranging from 0.561 to 1.0, 0.45 to 1.0, and 0.717 to 1.0, respectively, reducing unnecessary surgeries by approximately 70%.

Development and validation of a nomogram for predicting lymph node metastasis in early gastric cancer

Development and validation of a nomogram for predicting lymph node metastasis in early gastric cancer

Ultra-high b-value DWI in rectal cancer: image quality assessment and regional lymph node prediction based on radiomics.

This study aims to evaluate image quality and regional lymph node metastasis (LNM) in patients with rectal cancer (RC) on multi-b-value diffusion-weighted imaging (DWI). This retrospective study included 199 patients with RC who had undergone multi-b-value DWI. Subjective (five-point Likert scale) and objective assessments of quality images were performed on DWIb1000, DWIb2000, and DWIb3000. Patients were randomly divided into a training (n = 140) or validation cohort (n = 59). Radiomics features were extracted within the whole volume tumor on ADC maps (b = 0, 1000 s/mm2), DWIb1000, DWIb2000, and DWIb3000, respectively. Five prediction models based on selected features were developed using logistic regression analysis. The performance of radiomics models was evaluated with a receiver operating characteristic curve, calibration, and decision curve analysis (DCA). The mean signal intensity of the tumor (SItumor), signal-to-noise ratio (SNR), and artifact and anatomic differentiability score gradually weredecreased as the b-value increased. However, the contrast-to-noise (CNR) on DWIb2000 was superior to those of DWIb1000 and DWIb3000 (4.58 ± 0.86, 3.82 ± 0.77, 4.18 ± 0.84, p < 0.001, respectively). The overall image quality score of DWIb2000 was higher than that of DWIb3000 (p < 0.001) and showed no significant difference between DWIb1000 and DWIb2000 (p = 0.059). The area under curve (AUC) value of the radiomics model based on DWIb2000 (0.728) was higher than conventional ADC maps (0.690), DWIb1000 (0.699), and DWIb3000 (0.707), but inferior to multi-b-value DWI (0.739) in predicting LNM. DWIb2000 provides better lesion conspicuity and LNM prediction than DWIb1000 and DWIb3000 in RC. DWIb2000 offers satisfactory visualization of lesions. Radiomics features based on DWIb2000 can be applied for preoperatively predicting regional lymph node metastasis in rectal cancer, thereby benefiting the stratified treatment strategy. Lymph node staging is required to determine the best treatment plan for rectal cancer. DWIb2000 provides superior contrast-to-noise ratio and lesion conspicuity and its derived radiomics best predict lymph node metastasis. DWIb2000 may be recommended as the optimal b-value in rectal MRI protocol.

Predictors of lung cancer subtypes and lymph node status in non-small-cell lung cancer: intravoxel incoherent motion parameters and extracellular volume fraction

ObjectiveTo determine the performance of intravoxel incoherent motion (IVIM) parameters and the extracellular volume fraction (ECV) in distinguishing between different subtypes of lung cancer and predicting lymph node metastasis (LNM) status in patients with non-small-cell lung cancer (NSCLC).MethodsOne hundred sixteen patients with lung cancer were prospectively recruited. IVIM, native, and postcontrast T1 mapping examinations were performed, and the T1 values were measured to calculate the ECV. The differences in IVIM parameters and ECV were compared between NSCLC and small-cell lung cancer (SCLC), adenocarcinoma (Adeno-Ca) and squamous cell carcinoma (SCC), and NSCLC without and with LNM. The assessment of each parameter’s diagnostic performance was based on the area under the receiver operating characteristic curve (AUC).ResultsThe apparent diffusion coefficient (ADC), true diffusion coefficient (D), and ECV values in SCLC were considerably lower compared with NSCLC (all p < 0.001, AUC > 0.887). The D value in SCC was substantially lower compared with Adeno-Ca (p < 0.001, AUC = 0.735). The perfusion fraction (f) and ECV values in LNM patients were markedly higher compared with those without LNM patients (p < 0.01, < 0.001, AUC > 0.708).ConclusionIVIM parameters and ECV can serve as non-invasive biomarkers for assisting in the pathological classification and LNM status assessment of lung cancer patients.Critical relevance statementIVIM parameters and ECV demonstrated remarkable potential in distinguishing pulmonary carcinoma subtypes and predicting LNM status in NSCLC.Key PointsLung cancer is prevalent and differentiating subtype and invasiveness determine the treatment course.True diffusion coefficient and ECV showed promise for subtyping and determining lymph node status.These parameters could serve as non-invasive biomarkers to help determine personalized treatment strategies.Graphical

Can physician judgment enhance model trustworthiness? A case study on predicting pathological lymph nodes in rectal cancer

Explainability is key to enhancing the trustworthiness of artificial intelligence in medicine. However, there exists a significant gap between physicians’ expectations for model explainability and the actual behavior of these models. This gap arises from the absence of a consensus on a physician-centered evaluation framework, which is needed to quantitatively assess the practical benefits that effective explainability should offer practitioners. Here, we hypothesize that superior attention maps, as a mechanism of model explanation, should align with the information that physicians focus on, potentially reducing prediction uncertainty and increasing model reliability. We employed a multimodal transformer to predict lymph node metastasis of rectal cancer using clinical data and magnetic resonance imaging. We explored how well attention maps, visualized through a state-of-the-art technique, can achieve agreement with physician understanding. Subsequently, we compared two distinct approaches for estimating uncertainty: a standalone estimation using only the variance of prediction probability, and a human-in-the-loop estimation that considers both the variance of prediction probability and the quantified agreement. Our findings revealed no significant advantage of the human-in-the-loop approach over the standalone one. In conclusion, this case study did not confirm the anticipated benefit of the explanation in enhancing model reliability. Superficial explanations could do more harm than good by misleading physicians into relying on uncertain predictions, suggesting that the current state of attention mechanisms should not be overestimated in the context of model explainability.

Deep Learning Histology for Prediction of Lymph Node Metastases and Tumor Regression after Neoadjuvant FLOT Therapy of Gastroesophageal Adenocarcinoma.

The aim of this study was to establish a deep learning prediction model for neoadjuvant FLOT chemotherapy response. The neural network utilized clinical data and visual information from whole-slide images (WSIs) of therapy-naïve gastroesophageal cancer biopsies. This study included 78 patients from the University Hospital of Cologne and 59 patients from the University Hospital of Heidelberg used as external validation. After surgical resection, 33 patients from Cologne (42.3%) were ypN0 and 45 patients (57.7%) were ypN+, while 23 patients from Heidelberg (39.0%) were ypN0 and 36 patients (61.0%) were ypN+ (p = 0.695). The neural network had an accuracy of 92.1% to predict lymph node metastasis and the area under the curve (AUC) was 0.726. A total of 43 patients from Cologne (55.1%) had less than 50% residual vital tumor (RVT) compared to 34 patients from Heidelberg (57.6%, p = 0.955). The model was able to predict tumor regression with an error of ±14.1% and an AUC of 0.648. This study demonstrates that visual features extracted by deep learning from therapy-naïve biopsies of gastroesophageal adenocarcinomas correlate with positive lymph nodes and tumor regression. The results will be confirmed in prospective studies to achieve early allocation of patients to the most promising treatment.

Predicting lymph node metastasis in colorectal cancer patients: development and validation of a column chart model.

Lymph node metastasis (LNM) is one of the crucial factors in determining the optimal treatment approach for colorectal cancer. The objective of this study was to establish and validate a column chart for predicting LNM in colon cancer patients. We extracted a total of 83,430 cases of colon cancer from the Surveillance, Epidemiology, and End Results (SEER) database, spanning the years 2010-2017. These cases were divided into a training group and a testing group in a 7:3 ratio. An additional 8545 patients from the years 2018-2019 were used for external validation. Univariate and multivariate logistic regression models were employed in the training set to identify predictive factors. Models were developed using logistic regression, LASSO regression, ridge regression, and elastic net regression algorithms. Model performance was quantified by calculating the area under the ROC curve (AUC) and its corresponding 95% confidence interval. The results demonstrated that tumor location, grade, age, tumor size, T stage, race, and CEA were independent predictors of LNM in CRC patients. The logistic regression model yielded an AUC of 0.708 (0.7038-0.7122), outperforming ridge regression and achieving similar AUC values as LASSO regression and elastic net regression. Based on the logistic regression algorithm, we constructed a column chart for predicting LNM in CRC patients. Further subgroup analysis based on gender, age, and grade indicated that the logistic prediction model exhibited good adaptability across all subgroups. Our column chart displayed excellent predictive capability and serves as a useful tool for clinicians in predicting LNM in colorectal cancer patients.

Preoperative computed tomography semantic features in predicting lymph node metastasis of part-solid nodules in non-small cell lung cancer: a multicenter retrospective study.

Lymph node metastasis (LNM) is the most common route of metastasis for lung cancer, and it is an independent risk factor for long-term survival and recurrence in patients with non-small cell lung cancer (NSCLC). The purpose of this study was to explore the value of preoperative computed tomography (CT) semantic features in the differential diagnosis of LNM in part-solid nodules (PSNs) of NSCLC. A total of 955 patients with NSCLC confirmed by postoperative pathology were retrospectively enrolled from January 2019 to March 2023. The clinical, pathological data and preoperative CT images of these patients were investigated and statistically analyzed in order to identify the risk factors for LNM. Multivariate logistic regression was used to select independent risk factors and establish different prediction models. Ten-fold cross-validation was used for model training and validation. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve was calculated, and the Delong test was used to compare the predictive performance between the models. LNM occurred in 68 of 955 patients. After univariate analysis and adjustment for confounding factors, smoking history, pulmonary disease, solid component proportion, pleural contact type, and mean diameter were identified as the independent risk factors for LNM. The image predictors model established by the four independent factors of CT semantic features, except smoking history, showed a good diagnostic efficacy for LNM. The AUC in the validation group was 0.857, and the sensitivity, specificity, and accuracy of the model were all 77.6%. Preoperative CT semantic features have good diagnostic value for the LNM of NSCLC. The image predictors model based on pulmonary disease, solid component proportion, pleural contact type, and mean diameter demonstrated excellent diagnostic efficacy and can provide non-invasive evaluation in clinical practice.

Occult lymph node metastasis prediction in non-small cell lung cancer based self-supervised pretrained and hyperbolic theory

Predicting occult lymph node metastasis in non-small cell lung cancer (NSCLC) patients is pivotal for tailoring appropriate surgical and therapeutic interventions. This prognostic factor remains underexplored, largely due to the intricate variability of occult lymph node characteristics and the absence of a pathologically confirmed predictive dataset. Addressing this gap, we retrospectively assembled a dataset of occult lymph node metastases (TJ-OLNM) from NSCLC patients who underwent chest Computed Tomography (CT) scans at Tongji Hospital, Tongji University from 2016 to 2021. Utilizing this dataset, we developed a novel self-supervised learning model, the Occult Lymph Node Metastasis Network (OLNM-Net), which leverages hyperbolic metric few-shot learning to enhance the prediction accuracy of occult metastases. Our comprehensive evaluations demonstrate that OLNM-Net significantly outperforms existing models in predicting occult lymph node metastasis, offering new insights into the preoperative assessment of NSCLC and advancing the application of machine learning in medical diagnostics.

Preoperative prediction of central lymph node metastasis in follicular variant of papillary thyroid carcinoma using clinical and ultrasound features.

The most common metastatic site of follicular variant of papillary thyroid carcinoma (FVPTC) is the central lymph nodes, which may be associated with the prognosis and survival of patients. In the present study, we establish a combined model based on preoperative clinical and ultrasound (US) features of FVPTC to predict the risk of central lymph node metastasis (CLNM). From January 2013 to December 2022, 315 patients with FVPTC were enrolled and randomly divided into the training and validation cohorts in a ratio of 7:3. The independent risk factors for CLNM in FVPTC were analysed using univariate and multivariate logistic regression analyses. Then, three different models were established based on clinical and US data. Subsequently, a nomogram was constructed to predict CLNM. Its predictive effect was evaluated via receiver operating characteristic and calibration curve analyses. Backward multivariate regression analysis revealed that age (P=0.001), thyroid peroxidase antibody (TPOAb) (P=0.11), diameter (P=0.047), irregular/lobulated margin (P=0.15), extrathyroidal extension (P=0.001), nodules with macrocalcifications (P=0.009), nodules with microcalcification (P=0.003) and Thyroid Imaging Reporting and Data System (ACR-TI-RADS) category 5 (P=0.33) were independent risk factors for CLNM in FVPTC. The areas under the curve of the matching nomogram in the training (N=221) and validation cohorts (N=94) were 0.841 [95% confidence interval (CI): 0.788-0.895] and 0.735 (95% CI: 0.621-0.872), respectively. Preoperative thyroid US provides useful features for prediction of CLNM. The nomogram constructed based on combining US and clinical features can better predict the risk of CLNM and may facilitate decision-making in clinical settings.

Nomogram for predicting the risk of central lymph node metastasis in papillary thyroid microcarcinoma: a combination of sonographic findings and clinical factors.

A considerable controversy over performing thyroidectomy and central lymph node dissection in patients with papillary thyroid microcarcinoma (PTMC) remained. However, accurate prediction of central lymph node metastasis (CLNM) is crucial for surgical extent and proper management. The aim of this study was to develop and validate a practical nomogram for predicting CLNM in patients with PTMC. A total of 1,029 patients with PTMC who underwent thyroidectomy and central lymph node dissection at Tangdu Hospital (the Second Affiliated Hospital of Air Force Medical University) and Xijing Hospital (the First Affiliated Hospital of Air Force Medical University) were selected. Seven hundred and nine patients were assigned to the training set and 320 patients to the validation set. Data encompassing demographic characteristics, ultrasonography results, and biochemical indicators were obtained. Stepwise backward selection and multiple logistic regression were used to screen the variables and establish the nomogram. Concordance index (C-index), receiver operating characteristic (ROC) curve analysis, and decision curve analysis (DCA) were employed to evaluate the nomogram's distinguishability, accuracy, and clinical utility. Young age, multifocality, bigger tumor, presence of microcalcification, aspect ratio (height divided by width) ≥1, loss of fatty hilum, high free thyroxine (FT4), and lower anti-thyroid peroxidase antibody (TPOAb) were significantly associated with CLNM. The nomogram showed strong predictive capacity, with a C-index and accuracy of 0.784 and 0.713 in the training set and 0.779 and 0.703 in the external validation set, respectively. DCA indicated that the nomogram demonstrated strong clinical applicability. We established a reliable, cost-effective, reproducible, and noninvasive nomogram for predicting CLNM in patients with PTMC. This tool could be a valuable guidance for deciding on management in PTMC.

Application of CT and MRI images based on artificial intelligence to predict lymph node metastases in patients with oral squamous cell carcinoma: a subgroup meta-analysis.

The performance of artificial intelligence (AI) in the prediction of lymph node (LN) metastasis in patients with oral squamous cell carcinoma (OSCC) has not been quantitatively evaluated. The purpose of this study was to conduct a systematic review and meta-analysis of published data on the diagnostic performance of CT and MRI based on AI algorithms for predicting LN metastases in patients with OSCC. We searched the Embase, PubMed (Medline), Web of Science, and Cochrane databases for studies on the use of AI in predicting LN metastasis in OSCC. Binary diagnostic accuracy data were extracted to obtain the outcomes of interest, namely, the area under the curve (AUC), sensitivity, and specificity, and compared the diagnostic performance of AI with that of radiologists. Subgroup analyses were performed with regard to different types of AI algorithms and imaging modalities. Fourteen eligible studies were included in the meta-analysis. The AUC, sensitivity, and specificity of the AI models for the diagnosis of LN metastases were 0.92 (95% CI 0.89-0.94), 0.79 (95% CI 0.72-0.85), and 0.90 (95% CI 0.86-0.93), respectively. Promising diagnostic performance was observed in the subgroup analyses based on algorithm types [machine learning (ML) or deep learning (DL)] and imaging modalities (CT vs. MRI). The pooled diagnostic performance of AI was significantly better than that of experienced radiologists. In conclusion, AI based on CT and MRI imaging has good diagnostic accuracy in predicting LN metastasis in patients with OSCC and thus has the potential for clinical application. https://www.crd.york.ac.uk/PROSPERO/#recordDetails, PROSPERO (No. CRD42024506159).

Multimodal data integration using machine learning to predict the risk of clear cell renal cancer metastasis: a retrospective multicentre study.

To develop and validate a predictive combined model for metastasis in patients with clear cell renal cell carcinoma (ccRCC) by integrating multimodal data. In this retrospective study, the clinical and imaging data (CT and ultrasound) of patients with ccRCC confirmed by pathology from three tertiary hospitals in different regions were collected from January 2013 to January 2023. We developed three models, including a clinical model, a radiomics model, and a combined model. The performance of the model was determined based on its discriminative power and clinical utility. The evaluation indicators included area under the receiver operating characteristic curve (AUC) value, accuracy, sensitivity, specificity, negative predictive value, positive predictive value and decision curve analysis(DCA) curve. A total of 251 patients were evaluated. Patients (n = 166) from Shandong University Qilu Hospital (Jinan) were divided into the training cohort, of which 50 patients developed metastases; patients (n = 37) from Shandong University Qilu Hospital (Qingdao) were used as internal testing, of which 15 patients developed metastases; patients (n = 48) from Changzhou Second People's Hospital were used as external testing, of which 13 patients developed metastases. In the training set, the combined model showed the highest performance (AUC, 0.924) in predicting lymph node metastasis(LNM), while the clinical and radiomics models both had AUCs of 0.845 and 0.870, respectively. In the internal testing, the combined model had the highest performance (AUC, 0.877) for predicting LNM, while the AUCs of the clinical and radiomics models were 0.726 and 0.836, respectively. In the external testing, the combined model had the highest performance (AUC, 0.849) for predicting LNM, while the AUCs of the clinical and radiomics models were 0.708 and 0.804, respectively. The DCA curve showed that the combined model had a significant prediction probability in predicting the risk of LNM in ccRCC patients compared with the clinical model or the radiomics model. The combined model was superior to the clinical and radiomics models in predicting LNM in ccRCC patients.

Identification and Characterization of Metastasis-Initiating Cells in ESCC in a Multi-Timepoint Pulmonary Metastasis Mouse Model.

Metastasis is the biggest obstacle to esophageal squamous cell carcinoma (ESCC) treatment. Single-cell RNA sequencing analyses are applied to investigate lung metastatic ESCC cells isolated from pulmonary metastasis mouse model at multiple timepoints to characterize early metastatic microenvironment. A small population of parental KYSE30 cell line (Cluster S) resembling metastasis-initiating cells (MICs) is identified because they survive and colonize at lung metastatic sites. Differential expression profile comparisons between Cluster S and other subpopulations identified a panel of 7 metastasis-initiating signature genes (MIS), including CD44 and TACSTD2, to represent MICs in ESCC. Functional studies demonstrated MICs (CD44high) exhibited significantly enhanced cell survival (resistances to oxidative stress and apoptosis), migration, invasion, stemness, and in vivo lung metastasis capabilities, while bioinformatics analyses revealed enhanced organ development, stress responses, and neuron development, potentially remodel early metastasis microenvironment. Meanwhile, early metastasizing cells demonstrate quasi-epithelial-mesenchymal phenotype to support both invasion and anchorage. Multiplex immunohistochemistry (mIHC) staining of 4 MISs (CD44, S100A14, RHOD, and TACSTD2) in ESCC clinical samples demonstrated differential MIS expression scores (dMISs) predict lymph node metastasis, overall survival, and risk of carcinothrombosis.

Analysis of risk factors and construction of nomogram model for local lymph node metastasis in salivary gland mucoepidermoid carcinoma

Objective: To analyze the risk factors affecting regional lymph node metastasis in salivary gland mucoepidermoid carcinoma (MEC) and to establish a nomogram model for individually predicting lymph node metastasis in salivary gland MEC. Methods: The clinical data of 2 152 patients with salivary gland MEC from 1975 to 2020 were collected from the Surveillance, Epidemiology, and End Results (SEER) database of the National Cancer Institute. The collected data were divided into training cohort (1 506 cases) and validation cohort (646 cases) according to the ratio of 7∶3. Single-factor regression and multi-factor logistic regression were used to screen factors related to local lymph node metastasis in salivary gland MEC, with constructing of a nomogram. Calibration curve, receiver operating characteristic (ROC) curve, area under the ROC curve (AUC) and decision curve analysis were used to evaluate model performance in the validation cohort and the total cohort. Statistical tests were performed using SPSS (26.0) and R (4.3.0) software. Results: Multivariate logistic regression results showed that M stage [OR(95%CI):12.360(3.295-46.365), P=0.014], pathological grade Ⅱ、Ⅲ、Ⅳ[OR(95%CI): 1.956(1.329-2.879), 9.654(6.309-14.772), 9.298(6.072-14.238), P<0.001], T staging T2, T3, T4[OR(95%CI): 1.706(0.932-3.124), 3.021(1.790-5.096), 3.311(1.925-5.695), P<0.001], and gender [OR(95%CI):0.759(0.593-0.972), P=0.029] were independent factors affecting local lymph node metastasis in salivary gland MEC. Through verification in the validation cohort and the total cohort, the AUC values were greater than 0.8, and the calibration curve was close to the perfect reference line, proving that the constructed nomogram model had good specificity and sensitivity for predicting local lymph node metastasis in salivary gland MEC. Conclusion: M stage, pathological grade, T stage, and gender are risk factors for predicting regional lymph node metastasis and the established-nomogram has good predictive performance for local lymph node metastasis in salivary gland MEC.

Prediction of lymph node metastasis in T1 colorectal cancer based on combination of body composition and vascular invasion

ObjectivesLymph node metastasis (LNM) in colorectal cancer (CRC) patients is not only associated with the tumor’s local pathological characteristics but also with systemic factors. This study aims to assess the feasibility of using body composition and pathological features to predict LNM in early stage colorectal cancer (eCRC) patients.MethodsA total of 192 patients with T1 CRC who underwent CT scans and surgical resection were retrospectively included in the study. The cross-sectional areas of skeletal muscle, subcutaneous fat, and visceral fat at the L3 vertebral body level in CT scans were measured using Image J software. Logistic regression analysis were conducted to identify the risk factors for LNM. The predictive accuracy and discriminative ability of the indicators were evaluated using receiver operating characteristic (ROC) curves. Delong test was applied to compare area under different ROC curves.ResultsLNM was observed in 32 out of 192 (16.7%) patients with eCRC. Multivariate analysis revealed that the ratio of skeletal muscle area to visceral fat area (SMA/VFA) (OR = 0.021, p = 0.007) and pathological indicators of vascular invasion (OR = 4.074, p = 0.020) were independent risk factors for LNM in eCRC patients. The AUROC for SMA/VFA was determined to be 0.740 (p < 0.001), while for vascular invasion, it was 0.641 (p = 0.012). Integrating both factors into a proposed predictive model resulted in an AUROC of 0.789 (p < 0.001), indicating a substantial improvement in predictive performance compared to relying on a single pathological indicator.ConclusionThe combination of the SMA/VFA ratio and vascular invasion provides better prediction of LNM in eCRC.

Expression of CD109 in oral squamous cell carcinoma and its clinical significance.

The purpose of this study was to investigate the expression of CD109 and its clinicopathological significance in oral squamous cell carcinoma. Data from TIMER2.0 and UALCAN were analyzed to assess CD109 mRNA levels in OSCC. The immunohistochemical method was used to investigate the expressions of CD109 in 20 normal oral mucosa and 75 OSCC and analyzed the relationship between the expression of CD109 and the clinical variables. The mRNA levels of CD109 in OSCC tissues were significantly higher than in adjacent normal tissues (p<0.05). Immunohistochemical analysis revealed that CD109 protein expression was increased in OSCC tissues compared to normal tissues, and this difference was statistically significant (P<0.05). The positive rate of CD109 expression was 94% (16/117) in the group with lymph node metastasis, while it was 55% (32/58) in the group without metastasis (P<0.05). Similarly, the positive rate of CD109 expression was 91% (22/23) in the low differentiation group and 59% (26/52) in the high differentiation group (P<0.05). CD109 expression is markedly higher in OSCC, contributes to the pathological grading of OSCC and predicts lymph node metastasis.

A Preoperative Prediction Model for Lymph Node Metastasis in Patients with Gastric Cancer Using a Machine Learning-based Ultrasomics Approach.

This study aimed to develop an ultrasomics model for predicting lymph node metastasis preoperative in patients with gastric cancer (GC). This study enrolled GC patients who underwent preoperative ultrasound examination. Manual segmentation of the region of interest (ROI) was performed by an experienced radiologist to extract radiomics features using the Pyradiomics software. The Z-score algorithm was used for feature normalization, followed by the Wilcoxon test to identify the most informative features. Linear prediction models were constructed using the least absolute shrinkage and selection operator (LASSO). The performance of the ultrasomics model was evaluated using the area under curve (AUC), sensitivity, specificity, and the corresponding 95% confidence intervals (CIs). A total of 464 GC patients (mean age: 60.4 years ±11.3 [SD]; 328 men [70.7%]) were analyzed, of whom 291 had lymph node metastasis. The patients were randomly assigned to either the training (n=324) or test (n=140) sets, using a 7:3 ratio. An ultrasomics model that consisted of 19 radiomics features was developed using Wilcoxon and LASSO algorithms in the training set. Our ultrasomics model showed moderate performance for lymph node metastasis prediction in both the training (AUC: 0.802, 95%CI: 0.752-0.851, P<0.001) and test sets (AUC: 0.802, 95%CI: 0.724-0.879, P<0.001). The calibration curve analysis indicated good agreement between the predicted probabilities of ultrasomics and actual lymph node metastasis status. Our study highlights the potential of a machine learning-based ultrasomics model in predicting lymph node metastasis in GC patients, offering implications for personalized therapy approaches.

The value of dual time-point fluorine-18 fluorodeoxyglucose PET/computed tomography imaging in predicting lymph node metastasis in non-small cell lung cancer patients.

The purpose of this study was to analyze the correlation between specified dual time-point fluorine-18 fluorodeoxyglucose (18F-FDG) PET/computed tomography (CT) imaging parameters and pathological characteristics in non-small cell lung cancer (NSCLC) patients. This study retrospectively analyzed 47 patients with NSCLC. All patients underwent dual time-point 18F-FDG PET/CT imaging. We obtained the metabolic parameters, standardized uptake value (SUV) maximum, SUVmean, delayed standardized uptake value (DSUV) maximum, DSUVmean, delay index standardized uptake value (DISUV) maximum, and DISUVmean, of the primary tumor. The tumor size was measured by CT. All lymph nodes had a definite pathological diagnosis. We next evaluated the status of the lymph node metastases (LNM) and the correlations between metabolic parameters and clinical characteristics. Receiver operating characteristic curves were drawn for the prediction of LNM. We found that the DSUVmax, DISUVmax, DSUVmean, and tumor size were significantly related to LNM (P = 0.036, 0.009, and 0.049, respectively). Multivariate analysis revealed that tumor size and DISUVmax were independent risk factors for LNM in lung cancer patients. According to the receiver operating characteristic curve analysis, the optimal cutoff values for DISUVmax and tumor size were 0.33 and 2.8 cm, respectively. When these two parameters were combined, the area under the curve for predicting LNM in NSCLC was 0.768, and the sensitivity was 95.7% for predicting LNM in lung cancer patients. We further allocated the patients to three groups: the high-risk group (tumor size ≥ 2.8 cm, DISUVmax ≥ 0.33), the moderate-risk group (tumor size ≥ 2.8 cm, DISUVmax < 0.33, or tumor size < 2.8 cm, DISUVmax ≥ 0.33), and the low-risk group (tumor size < 2.8 cm, DISUVmax < 0.33). The rates of LNM were 70, 50, and 0%, respectively. Tumor size and DISUVmax are risk factors for predicting LNM, and they are more useful in combination. Compared with standard PET/CT imaging, dual time-point PET/CT imaging has added value in predicting LNM in NSCLC patients.

A deep learning-based radiomics model for predicting lymph node status from lung adenocarcinoma

ObjectivesAt present, there are many limitations in the evaluation of lymph node metastasis of lung adenocarcinoma. Currently, there is a demand for a safe and accurate method to predict lymph node metastasis of lung cancer. In this study, radiomics was used to accurately predict the lymph node status of lung adenocarcinoma patients based on contrast-enhanced CT.MethodsA total of 503 cases that fulfilled the analysis requirements were gathered from two distinct hospitals. Among these, 287 patients exhibited lymph node metastasis (LNM +) while 216 patients were confirmed to be without lymph node metastasis (LNM-). Using both traditional and deep learning methods, 22,318 features were extracted from the segmented images of each patient's enhanced CT. Then, the spearman test and the least absolute shrinkage and selection operator were used to effectively reduce the dimension of the feature data, enabling us to focus on the most pertinent features and enhance the overall analysis. Finally, the classification model of lung adenocarcinoma lymph node metastasis was constructed by machine learning algorithm. The Accuracy, AUC, Specificity, Precision, Recall and F1 were used to evaluate the efficiency of the model.ResultsBy incorporating a comprehensively selected set of features, the extreme gradient boosting method (XGBoost) effectively distinguished the status of lymph nodes in patients with lung adenocarcinoma. The Accuracy, AUC, Specificity, Precision, Recall and F1 of the prediction model performance on the external test set were 0.765, 0.845, 0.705, 0.784, 0.811 and 0.797, respectively. Moreover, the decision curve analysis, calibration curve and confusion matrix of the model on the external test set all indicated the stability and accuracy of the model.ConclusionsLeveraging enhanced CT images, our study introduces a noninvasive classification prediction model based on the extreme gradient boosting method. This approach exhibits remarkable precision in identifying the lymph node status of lung adenocarcinoma patients, offering a safe and accurate alternative to invasive procedures. By providing clinicians with a reliable tool for diagnosing and assessing disease progression, our method holds the potential to significantly improve patient outcomes and enhance the overall quality of clinical practice.