Multi-modal radiomics model based on four imaging modalities for predicting pathological complete response to neoadjuvant treatment in breast cancer DOI Creative Commons

Yuwen Liang,

Haonan Xu, Jie Lin

и другие.

BMC Cancer, Год журнала: 2025, Номер 25(1)

Опубликована: Июнь 2, 2025

The radiomics model based on single imaging modality has been demonstrated as a promising approach for predicting the response to neoadjuvant treatment (NAT) in breast cancer. However, whether integrating multiple modalities improve performance of is undetermined. This study aims develop multi-modal four modalities, including ultrasound (US), mammography (MM), computed tomography (CT), and magnetic resonance (MRI), pathological complete (pCR) cancer after NAT. Patients who underwent surgery NAT from January 2019 July 2023 were retrospectively studied. Univariate multivariate analyses performed identify independent clinical risk factors pCR. radiomic features extracted volume interest modalities. least absolute shrinkage selection operator was used developing signatures. developed by combining combined A nomogram visualize model. Model internally validated using five-fold cross-validation. In total, 89 patients included, with pCR rate 31.5% (28/89). Multivariate identified PR status (OR = 4.450, 95% confidence interval [CI], 1.228-18.063, P 0.028), HER2 9.95, CI, 1.525-201.894, 0.044) T stage 0.253, 0.076-0.753, 0.016) AUCs brier scores signatures US, MM, CT, MRI 0.702 (95% CI: 0.583-0.821), 0.762 0.660-0.865), 0.814 0.725-0.903), 0.787 0.685-0.889) 0.198, 0.177, 0.165, 0.170 respectively. superior all an AUC 0.904 0.838-0.970) score 0.111. After adding factors, further improved, achieving 0.943 0.893-0.992) 0.082. showed potential value. could accurately predict NAT, which Incorporating may muti-modal model, provide valuable information guiding decisions.

Язык: Английский

Evaluating AI and Machine Learning Models in Breast Cancer Detection: A Review of Convolutional Neural Networks (CNN) and Global Research Trends DOI

Mutaz Abdel Wahed,

Muhyeeddin Alqaraleh, Mowafaq Salem Alzboon

и другие.

LatIA, Год журнала: 2024, Номер 3, С. 117 - 117

Опубликована: Окт. 18, 2024

Numerous studies have highlighted the significance of artificial intelligence (AI) in breast cancer diagnosis. However, systematic reviews AI applications this field often lack cohesion, with each study adopting a unique approach. The aim is to provide detailed examination AI's role diagnosis through citation analysis, helping categorize key areas that attract academic attention. It also includes thematic analysis identify specific research topics within category. A total 30,200 related and AI, published between 2015 2024, were sourced from databases such as IEEE, Scopus, PubMed, Springer, Google Scholar. After applying inclusion exclusion criteria, 32 relevant identified. Most these utilized classification models for prediction, high accuracy being most commonly reported performance metric. Convolutional Neural Networks (CNN) emerged preferred model many studies. findings indicate both quantity quality AI-based algorithms are increases given years. increasingly seen complement healthcare sector clinical expertise, target enhancing accessibility affordability worldwide.

Язык: Английский

Процитировано

24
Enhancing Diagnostic Efficiency: A Radiomics Approach for Distinguishing Benign and Malignant Breast Lesions Using BI-RADS Features from Ultrasound Imaging DOI

Runqiu Cai,

Man Wang, Yan Yu

и другие.

Clinical Breast Cancer, Год журнала: 2025, Номер unknown

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

0
Comparative Evaluation of Machine Learning-Based Radiomics and Deep Learning for Breast Lesion Classification in Mammography DOI Creative Commons
Alessandro Stefano, Fabiano Bini,

Eleonora Giovagnoli

и другие.

Diagnostics, Год журнала: 2025, Номер 15(8), С. 953 - 953

Опубликована: Апрель 9, 2025

Background: Breast cancer is the second leading cause of cancer-related mortality among women, accounting for 12% cases. Early diagnosis, based on identification radiological features, such as masses and microcalcifications in mammograms, crucial reducing rates. However, manual interpretation by radiologists complex subject to variability, emphasizing need automated diagnostic tools enhance accuracy efficiency. This study compares a radiomics workflow machine learning (ML) with deep (DL) approach classifying breast lesions benign or malignant. Methods: matRadiomics was used extract features from mammographic images 1219 patients CBIS-DDSM public database, including 581 cases 638 masses. Among ML models, linear discriminant analysis (LDA) demonstrated best performance both lesion types. External validation conducted private dataset 222 evaluate generalizability an independent cohort. Additionally, EfficientNetB6 model employed comparison. Results: The LDA achieved mean AUC 68.28% 61.53% In external validation, values 66.9% 61.5% were obtained, respectively. contrast, superior performance, achieving 81.52% 76.24% masses, highlighting potential DL improved accuracy. Conclusions: underscores limitations ML-based diagnosis. Deep proves be more effective approach, offering enhanced supporting clinicians improving patient management.

Язык: Английский

Процитировано

0
Revolutionizing breast cancer care: the synergy of AI-powered diagnostics, haptic-based biopsy simulators, and advanced surgical techniques DOI

Tanyaradzwa Roselyn Tichiwangana,

Qing Ji, Xingqi Fan

и другие.

Expert Review of Medical Devices, Год журнала: 2025, Номер unknown

Опубликована: Май 29, 2025

In 2022, a report by the World Health Organization revealed 2.3 million new breast cancer cases and 670,000 related deaths, which represented 11.7% of all worldwide. Early screening biopsy for can provide more effective minimally invasive treatment options. As options evolve, surgery ensure cure rate aesthetics after surgery. This review article examines latest advancements in care, highlighting integration artificial intelligence (AI) diagnostics, development haptic-based simulators, innovative surgical techniques. AI-driven diagnostic systems have significantly improved accuracy effectiveness with precision comparable to that experienced radiologists. Furthermore, simulators are revolutionizing training providing practitioners realistic safe environment refine their techniques skills. Concurrently, procedures, often augmented AI virtual reality (VR) simulations, transforming treatment, facilitate practice complex techniques, potentially resulting specialized procedures. Collectively, these innovations improving screening, diagnosis, results patients.

Язык: Английский

Процитировано

0
Multi-modal radiomics model based on four imaging modalities for predicting pathological complete response to neoadjuvant treatment in breast cancer DOI Creative Commons

Yuwen Liang,

Haonan Xu, Jie Lin

и другие.

BMC Cancer, Год журнала: 2025, Номер 25(1)

Опубликована: Июнь 2, 2025

The radiomics model based on single imaging modality has been demonstrated as a promising approach for predicting the response to neoadjuvant treatment (NAT) in breast cancer. However, whether integrating multiple modalities improve performance of is undetermined. This study aims develop multi-modal four modalities, including ultrasound (US), mammography (MM), computed tomography (CT), and magnetic resonance (MRI), pathological complete (pCR) cancer after NAT. Patients who underwent surgery NAT from January 2019 July 2023 were retrospectively studied. Univariate multivariate analyses performed identify independent clinical risk factors pCR. radiomic features extracted volume interest modalities. least absolute shrinkage selection operator was used developing signatures. developed by combining combined A nomogram visualize model. Model internally validated using five-fold cross-validation. In total, 89 patients included, with pCR rate 31.5% (28/89). Multivariate identified PR status (OR = 4.450, 95% confidence interval [CI], 1.228-18.063, P 0.028), HER2 9.95, CI, 1.525-201.894, 0.044) T stage 0.253, 0.076-0.753, 0.016) AUCs brier scores signatures US, MM, CT, MRI 0.702 (95% CI: 0.583-0.821), 0.762 0.660-0.865), 0.814 0.725-0.903), 0.787 0.685-0.889) 0.198, 0.177, 0.165, 0.170 respectively. superior all an AUC 0.904 0.838-0.970) score 0.111. After adding factors, further improved, achieving 0.943 0.893-0.992) 0.082. showed potential value. could accurately predict NAT, which Incorporating may muti-modal model, provide valuable information guiding decisions.

Язык: Английский

Процитировано

0