Ultrasound-Base Radiomics for Discerning Lymph Node Metastasis in Thyroid Cancer: A Systematic Review and Meta-analysis

Academic Radiology, Journal Year: 2024, Volume and Issue: 31(8), P. 3118 - 3130

Published: March 29, 2024

Language: Английский

---

Dual-modal radiomics nomogram based on contrast-enhanced ultrasound to improve differential diagnostic accuracy and reduce unnecessary biopsy rate in ACR TI-RADS 4–5 thyroid nodules

Cancer Imaging, Journal Year: 2024, Volume and Issue: 24(1)

Published: Jan. 23, 2024

Abstract Background American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS, TR) 4 5 thyroid nodules (TNs) demonstrate much more complicated overlapping risk characteristics than TR1-3 have a rather wide range malignancy possibilities (> 5%), which may cause overdiagnosis or misdiagnosis. This study was designed to establish validate dual-modal ultrasound (US) radiomics nomogram integrating B-mode (BMUS) contrast-enhanced (CEUS) imaging improve differential diagnostic accuracy reduce unnecessary fine needle aspiration biopsy (FNAB) rates in TR 4–5 TNs. Methods A retrospective dataset 312 pathologically confirmed TR4-5 TNs from 269 patients collected for our study. were randomly divided into training 219 validation 93 Radiomics derived the BMUS CEUS images. After feature reduction, scores (Rad-score) built. multivariate logistic regression analysis conducted incorporating both Rad-scores clinical/US data, subsequently developed. The performance evaluated using calibration, discrimination, clinical usefulness, FNAB rate also calculated. Results Rad-score, age, shape, margin, enhancement direction significant independent predictors associated with malignant involving six variables exhibited excellent calibration discrimination cohorts, an AUC 0.873 (95% CI, 0.821–0.925) 0.851 0.764–0.938), respectively. marked improvements net reclassification index integrated discriminatory improvement suggested that could be valuable indicators distinguishing benign Decision curve demonstrated developed instrumental tool decision-making. Using nomogram, decreased 35.3 14.5% cohort 41.5 17.7% cohorts compared ACR TI-RADS. Conclusion US revealed superior considerably It guide further examination treatment options.

Language: Английский

---

RadWise: A Rank-Based Hybrid Feature Weighting and Selection Method for Proteomic Categorization of Chemoirradiation in Patients with Glioblastoma

Cancers, Journal Year: 2023, Volume and Issue: 15(10), P. 2672 - 2672

Published: May 9, 2023

Glioblastomas (GBM) are rapidly growing, aggressive, nearly uniformly fatal, and the most common primary type of brain cancer. They exhibit significant heterogeneity resistance to treatment, limiting ability analyze dynamic biological behavior that drives response resistance, which central advancing outcomes in glioblastoma. Analysis proteome aimed at signal change over time provides a potential opportunity for non-invasive classification examination treatment by identifying protein biomarkers associated with interventions. However, data acquired using large proteomic panels must be more intuitively interpretable, requiring computational analysis identify trends. Machine learning is increasingly employed, however, it requires feature selection has critical considerable effect on machine problems when applied large-scale reduce number parameters, improve generalization, find essential predictors. In this study, 7k generated from serum obtained 82 patients GBM pre- post-completion concurrent chemoirradiation (CRT), we select discriminative features define alteration result administering CRT. Thus, present novel rank-based weighting method (RadWise) relevant parameters two popular methods, least absolute shrinkage operator (LASSO) minimum redundancy maximum relevance (mRMR). The results show proposed yields outstanding very few selected features, higher accuracy rate performance than methods do not employ process. While identified several signals identical clinical intuitive (heuristic approach), heuristically were while other heuristic approach carry prognostic only emerged method. promising results, reducing 7 value 93.921%, comparing favorably techniques selection.

Language: Английский

---

The application of a clinical-multimodal ultrasound radiomics model for predicting cervical lymph node metastasis of thyroid papillary carcinoma

Frontiers in Oncology, Journal Year: 2025, Volume and Issue: 14

Published: Jan. 17, 2025

PTC (papillary thyroid cancer) is a lymphotropic malignancy associated with cervical lymph node metastasis (CLNM, including central and lateral LNM), which compromises the effect of treatment prognosis patients. Accurate preoperative identification will provide valuable reference information for formulation diagnostic strategies. The aim this study was to develop validate clinical-multimodal ultrasound radiomics model predicting CLNM PTC. One hundred sixty-four patients who underwent at our hospital between March 2016 December 2021 were included in study. grouped into training cohort (n=115) validation (n=49). Radiomic features extracted from conventional (US), contrast-enhanced (CEUS) strain elastography-ultrasound (SE-US) images Multivariate logistic regression analysis used identify independent risk factors. FAE software radiomic feature extraction construction different prediction models. performance each evaluated compared terms area under curve (AUC), sensitivity, specificity, accuracy, negative predictive value (NPV) positive (PPV). RStudio decision assess clinical model. developed can successfully detect A total 3720 (930 per modality) ROIs multimodal images, 15 representative ultimately screened. combined showed best both cohorts, AUCs 0.957 (95% CI: 0.918-0.987) 0.932 0.822-0.984), respectively. Decision revealed that superior other constructed factors has favorable potential high

Language: Английский

---

Drawing as a Way of Knowing: How a Mapping Model Assists Preoperative Evaluation of Patients with Thyroid Carcinoma

Journal of Clinical Medicine, Journal Year: 2024, Volume and Issue: 13(5), P. 1389 - 1389

Published: Feb. 28, 2024

Background: Effective pre-surgical planning is crucial for achieving successful outcomes in endocrine surgery: it essential to provide patients with a personalized plan minimize operative and postoperative risks. Methods: Preoperative lymph node (LN) mapping structured high-resolution ultrasonography examination performed the presence of two endocrinologists operating surgeon before intervention produce reliable “anatomical guide”. Our aim was propose preoperative complete model that non-invasive, avoids overdiagnosis thyroid microcarcinomas, reduces medical expenses. Results: The use ‘preoperative echography mapping’ has been shown be successful, particularly suspected or confirmed neoplastic malignancy. Regarding prognosis, positive have observed both post-surgery terms recurrence rates. We collected data on parameters such as biological sex, age, BMI, results from cytologic tests needle aspiration, examined whether these predict tumor malignancy aggressiveness, calculated using multivariate analysis (MVA). Conclusions: A standard multidisciplinary approach evaluating neck nodes pre-operation proven an improved diagnostic tool.

Language: Английский

---

Radiomics analysis based on CT for predicting lymph node metastasis and prognosis in duodenal papillary carcinoma

Insights into Imaging, Journal Year: 2024, Volume and Issue: 15(1)

Published: June 20, 2024

Abstract Objectives Radiomics has been demonstrated to be strongly associated with TNM stage and patient prognosis. We aimed develop a model for predicting lymph node metastasis (LNM) survival. Methods For radiomics texture selection, 3D Slicer 5.0.3 software the least absolute shrinkage selection operator (LASSO) algorithm were used. Subsequently, model, computed tomography (CT) image, clinical risk compared. The performance of three models was evaluated using receiver operating characteristic (ROC) curves, decision curve analysis (DCA), calibration plots, impact curves (CICs). Results LNM prediction 224 patients information used construct that applied predict LNM. According CT data characteristics, we constructed imaging model. evaluating status showed excellent discrimination in training cohort (AUC = 0.926, 95% CI 0.869–0.982) validation 0.872, 0.802–0.941). DeLong’s test difference among significant. Similarly, DCA CIC better utility than Our also exhibited good survival—in line findings built factors. Conclusions predictive based on characteristics had comparative Critical relevance statement survival duodenal papillary carcinoma (DPC). Key Points determines most appropriate treatment DPC. DPC performed excellently. high sensitivity specificity survival, exhibiting great value. Graphical

Language: Английский

---

Predictive Values of Clinical Features and Multimodal Ultrasound for Central Lymph Node Metastases in Papillary Thyroid Carcinoma

Diagnostics, Journal Year: 2024, Volume and Issue: 14(16), P. 1770 - 1770

Published: Aug. 14, 2024

Papillary thyroid carcinoma (PTC), the predominant pathological type among malignancies, is responsible for sharp increase in cancer. Although PTC an indolent tumor with good prognosis, 60–70% of patients still have early cervical lymph node metastasis, typically central compartment. Whether there metastasis (CLNM) or not directly affects formulation preoperative surgical procedures, given that such metastases been tied to compromised overall survival and local recurrence. However, detecting CLNM before operation can be challenging due limited sensitivity approaches. Prophylactic dissection (PCLND) absence clinical evidence poses additional risks. This study aims provide a comprehensive review risk factors related patients. A key focus on utilizing multimodal ultrasound (US) accurate prognosis highlight distinctive role US-based characteristics predicting CLNM.

Language: Английский

---

Integration of ultrasound-based radiomics with clinical features for predicting cervical lymph node metastasis in postoperative patients with differentiated thyroid carcinoma

Endocrine, Journal Year: 2023, Volume and Issue: 84(3), P. 999 - 1012

Published: Dec. 22, 2023

Abstract Objective The primary objective was to establish a radiomics model utilizing longitudinal +cross-sectional ultrasound (US) images of lymph nodes (LNs) predict cervical node metastasis (CLNM) following differentiated thyroid carcinoma (DTC) surgery. Methods A retrospective collection 211 LNs from postoperative DTC patients who underwent neck US with suspicious LN fine needle aspiration cytopathology findings at our institution conducted between June 2021 and April 2023. Conventional clinicopathological information were gathered. Based on the pathological results, categorized into CLNM non-CLNM groups. database randomly divided training cohort ( n = 147) test 64) 7:3 ratio. least absolute shrinkage selection operator algorithm applied screen most relevant radiomic features + cross-sectional images, constructed. Univariate multivariate analyses used assess significance features. Subsequently, combined for predicting constructed by integrating radiomics, conventional US, presented as nomogram. Results area under curves (AUCs) models 0.846 0.801 in sets, respectively, outperforming single p < 0.05). In testing cohort, AUC 0.901, surpassing that (AUC, 0.731) 0.801). Conclusions US-based exhibits potential accurately surgery, thereby enhancing diagnostic accuracy.

Language: Английский

---

Implications of a Ultrasomics Signature for Predicting Malignancy in Thyroid Nodules with Hashimoto’s Thyroiditis

Academic Radiology, Journal Year: 2024, Volume and Issue: 31(11), P. 4386 - 4395

Published: May 24, 2024

Language: Английский

---

The clinical value of predicting lymphovascular invasion in patients with invasive lung adenocarcinoma based on the intratumoral and peritumoral CT radiomics models

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 26, 2024

Purpose: To investigate the clinical value of predicting lymphovascular invasion(LVI) in patients with invasive lung adenocarcinoma(LUAD)based on intratumoral and peritumoral CT radiomics models. Materials Methods: The 384 LUAD from Institution 1 were randomly divided into training (n=268) internal validation (n=116) sets a ratio 7:3, 251 2 used as external set. Altogether, 1226 features extracted tumor gross (GT), peritumor (GPT), peritumor(PT), respectively. Clinical independent predictors for LVI screened using univariate multivariate logistic regression, combined model that included optimal Rad-score was constructed , nomogram drawn. Results: The GPT showed better predictive efficacy than GT PT models, area under curve (AUC) 0.83, 0.79, 0.75 training, validation, sets, In model, preoperative carcinoembryonic antigen (CEA) level, diameter, spiculation predictors. containing GPT-Radscore significantly predicted LUAD, AUCs 0.84, 0.82, 0.77 three cohorts, Conclusion: scan-based which including can effectively predict LUAD，and is further improved by combining clinically

Language: Английский

---