An Informative Review of Radiomics Studies on Cancer Imaging: The Main Findings, Challenges and Limitations of the Methodologies

Current Oncology, Journal Year: 2024, Volume and Issue: 31(1), P. 403 - 424

Published: Jan. 10, 2024

The aim of this informative review was to investigate the application radiomics in cancer imaging and summarize results recent studies support oncological with particular attention breast cancer, rectal primitive secondary liver cancer. This also aims provide main findings, challenges limitations current methodologies. Clinical published last four years (2019–2022) were included review. Among 19 analyzed, none assessed differences between scanners vendor-dependent characteristics, collected images individuals at additional points time, performed calibration statistics, represented a prospective study registered database, conducted cost-effectiveness analysis, reported on clinical application, or multivariable analysis non-radiomics features. Seven reached high radiomic quality score (RQS), seventeen earned by using validation steps considering two datasets from distinct institutes open science data domains (radiomics features calculated set representative ROIs are source). potential is increasingly establishing itself, even if there still several aspects be evaluated before passage into routine practice. There challenges, including need for standardization across all stages workflow cross-site real-world heterogeneous datasets. Moreover, multiple centers more samples that add inter-scanner characteristics will needed future, as well collecting time points, reporting statistics performing database.

Language: Английский

---

Breast cancer staging with contrast-enhanced imaging. The benefits and drawbacks of MRI, CEM, and dedicated breast CT

European Journal of Radiology, Journal Year: 2025, Volume and Issue: 185, P. 112013 - 112013

Published: Feb. 28, 2025

Language: Английский

---

Vision transformer-based multimodal fusion network for classification of tumor malignancy on breast ultrasound: A retrospective multicenter study

International Journal of Medical Informatics, Journal Year: 2025, Volume and Issue: 196, P. 105793 - 105793

Published: Jan. 21, 2025

Language: Английский

---

Development of a Prognostic Model for Preoperative Stage I-III Breast Cancer Using Machine Learning with Integrated Cone-Beam Breast Computed Tomography Data in the Context of 3P Medicine

Research Square (Research Square), Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 3, 2025

Background/Objectives: The lack of reliable prognostic predictors in breast cancer undermines the efficacy its prediction, prevention, and personalized medicine (PPPM/3PM) approach. This study aimed to develop an integrated model based on cone-beam computed tomography (CBBCT) hematological indicators predict prognosis preoperative stage I-III cancer. Methods:A retrospective analysis was performed 243 patients with pathologically confirmed A new machine learning framework for feature selection 10 algorithms their 101 combinations. After selection, patient risk score calculated construct a nomogram prognosis. evaluated using receiver operating characteristic (ROC) curve calibration curve. Univariate multivariate logistic regression analyses verified screened features determined independent factors. Results: A computational combinations selected 12 overall survival (OS) 18 disease-free survivals (DFS) from 37 CBBCT features. entire achieved AUC value 0.837 training dataset 0.813 validation dataset, which is superior clinical without regarding OS prediction performance. Similarly, sets DFS 0.996 0.732. Molecular typing, Enhancement types, Morphology were factors associated model. Calcification factor DFS. We constructed combining above Conclusions: Our prognostic-related features, showed satisfactory predictive It can be incorporated into PPPM help clinicians make more accurate treatment decisions.

Language: Английский

---

Cone-beam Breast CT Features Associated With Lymphovascular Invasion in Patients With Breast Cancer

Academic Radiology, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 1, 2025

Language: Английский

---

Prediction of sentinel lymph node status in patients with early breast cancer using breast imaging as an alternative to surgical staging – A systematic review and meta-analysis

Research Square (Research Square), Journal Year: 2025, Volume and Issue: unknown

Published: May 15, 2025

Background Prediction models for sentinel lymph node status could offer an alternative to surgical axillary staging in patients with early breast cancer. Several imaging modalities have been used various approaches feature engineering. This systematic review and meta-analysis aimed evaluate prediction (SLN) using cancer summarize the current evidence identify areas requiring additional research. Methods The literature search strategy was based on following Population, Intervention, Comparison, Outcome (PICO): P: female clinically node-negative invasive scheduled undergo primary surgery; I: imaging; C: upfront biopsy; O: model performance regarding SLN status. conducted PubMed, Embase, Web of Science, Cochrane, Cumulative Index Nursing Allied Health Literature databases were searched March 2024. screening records, data collection, bias assessments performed independently by two reviewers. risk assessed Quality Assessment Diagnostic Accuracy Studies 2 (QUADAS-2) tool Model Study Risk Bias Tool. A a random-effects assess heterogeneity overall subgroups. Results resulted inclusion 32 articles review. Assessments QUADAS-2 revealed four studies high bias, which excluded from meta-analysis. subgroups, except magnetic resonance (MRI)-based studies, pooled area under curve 0.85 (95% confidence interval 0.82–0.87). Meta-regression analyses indicated that MRI, including only one modality, calibration assessment upon validation contributed heterogeneity. Conclusions imaging, particularly be noninvasive results illustrate between need high-quality studies. Systematic registration PROSPERO CRD42022301852, available at https://www.crd.york.ac.uk/PROSPERO

Language: Английский

---

Emerging Clinical Applications for Cone Beam Breast CT: Changing the Breast Imaging Paradigm

Current Breast Cancer Reports, Journal Year: 2024, Volume and Issue: 16(2), P. 134 - 141

Published: March 18, 2024

Abstract Purpose of Review Since its approval by the Unites States Food and Drug Administration (FDA) in 2015, cone-beam breast computed tomography (CBBCT) has gained acceptance among radiologists for cancer imaging. This review aims to highlight advancements benefits CBBCT diagnostic workup disease. It showcases how CBBCT, including both non-contrast (NC-CBBCT) contrast-enhanced (CE-CBBCT) protocols, complements often surpasses performance more traditional imaging modalities such as mammography magnetic resonance (MRI). Recent Findings Studies clinical settings have shown CBBCT’s efficacy detecting characterizing lesions differing morphologies, non-mass enhancement calcifications—tasks that previously required use multiple modalities. In addition, significantly enhances patient comfort efficiency, offering quick acquisition times without discomfort compression. The technology can be utilized guiding biopsies, planning surgical interventions, assessing density tumor characteristics, evidence supporting integration into practice. Summary holds potential shift paradigm care, indicating a promising future modality terms enhancing accuracy, improving experience, influencing treatment outcomes.

Language: Английский

---

An unsupervised learning model based on CT radiomics features accurately predicts axillary lymph node metastasis in breast cancer patients—diagnostic study

International Journal of Surgery, Journal Year: 2024, Volume and Issue: 110(9), P. 5363 - 5373

Published: June 7, 2024

Background: The accuracy of traditional clinical methods for assessing the metastatic status axillary lymph nodes (ALNs) is unsatisfactory. In this study, authors propose use radiomic technology and three-dimensional (3D) visualization to develop an unsupervised learning model predicting node metastasis in patients with breast cancer (BC), aiming provide a new method assessment disease. Methods: we retrospectively analyzed data 350 invasive BC who underwent lung-enhanced computed tomography (CT) dissection surgery at Department Breast Surgery Second Xiangya Hospital Central South University. used 3D create atlas ALNs identified region interest nodes. Radiomic features were subsequently extracted selected, prediction was constructed using K-means algorithm. To validate model, prospectively collected from 128 clinically evaluated as negative our center. Results: Using technology, selected total 36 CT radiomics features. categorized 1737 unlabeled into two groups, analysis between these groups indicated potential differences status. Further validation 1397 labeled demonstrated that had good predictive ability status, area under curve 0.847 (0.825–0.869). Additionally, model’s excellent performance confirmed cohort (cN0) positive (cN+) cohort, which correct classification rates 86.72 87.43%, respectively, significantly greater than those methods. Conclusions: created accurately predicts ALNs. This approach offers novel solution precise BC.

Language: Английский

---

Development and Preliminary Validation of a Novel Convolutional Neural Network Model for Predicting Treatment Response in Patients with Unresectable Hepatocellular Carcinoma Receiving Hepatic Arterial Infusion Chemotherapy

Deleted Journal, Journal Year: 2024, Volume and Issue: unknown

Published: Feb. 23, 2024

The goal of this study was to evaluate the performance a convolutional neural network (CNN) with preoperative MRI and clinical factors in predicting treatment response unresectable hepatocellular carcinoma (HCC) patients receiving hepatic arterial infusion chemotherapy (HAIC). A total 191 HCC who underwent HAIC our hospital between May 2019 March 2022 were retrospectively recruited. We selected InceptionV4 from three representative CNN models, AlexNet, ResNet, InceptionV4, according cross-entropy loss (CEL). subsequently developed fuse information qualified pretreatment data patient factors. Radiomic evaluated based on several constant sequences, including enhanced T1-weighted sequences (with arterial, portal, delayed phases), T2 FSE dual-echo sequences. cross-validated training cohort (n = 127) internally validated an independent 64), comparisons against single important radiologists terms receiver operating characteristic (ROC) curves. Class activation mapping used visualize model. model achieved AUC 0.871 (95% confidence interval [CI] 0.761–0.981) cross-validation 0.826 CI 0.682–0.970) internal validation cohort; these two models performed better than did other methods (AUC ranges 0.783–0.873 0.708–0.806 for cross- validations, respectively; P < 0.01). present model, which integrates radiomic factors, helps predict treatment.

Language: Английский

---

Performance evaluation of ML models for preoperative prediction of HER2-low BC based on CE-CBBCT radiomic features: A prospective study

Medicine, Journal Year: 2024, Volume and Issue: 103(24), P. e38513 - e38513

Published: June 14, 2024

To explore the value of machine learning (ML) models based on contrast-enhanced cone-beam breast computed tomography (CE-CBBCT) radiomics features for preoperative prediction human epidermal growth factor receptor 2 (HER2)-low expression cancer (BC). Fifty-six patients with HER2-negative invasive BC who underwent CE-CBBCT were prospectively analyzed. Patients randomly divided into training and validation cohorts at approximately 7:3. A total 1046 quantitative radiomic extracted from images normalized using z-scores. The Pearson correlation coefficient recursive feature elimination used to identify optimal features. Six ML constructed selected features: linear discriminant analysis (LDA), random forest (RF), support vector (SVM), logistic regression (LR), AdaBoost (AB), decision tree (DT). evaluate performance these models, receiver operating characteristic curves area under curve (AUC) used. Seven as constructing models. In cohort, AUC values SVM, LDA, RF, LR, AB, DT 0.984, 0.981, 1.000, 0.970, respectively. 0.859, 0.880, 0.781, 0.750, 0.713, Among all LDA LR demonstrated best performance. DeLong test showed that there no significant differences among in cohort (P > .05); however, between AUCs statistically = .037, .003, .046). .023, .005, .030). Nevertheless, observed when compared other achieved excellent HER2-low could potentially serve an effective tool assist precise personalized targeted therapy.

Language: Английский

---