Application of machine learning with multiparametric dual-energy computed tomography of the breast to differentiate between benign and malignant lesions

Xiao, Lan; Wang, Xiaoxia; Qi, Jun; Chen, Huifang; Zeng, Xiangfei; Shi, Jinfang; Liu, Dai‐Hong; Shen, Hesong; Zhang, Jiuquan

doi:10.21037/qims-21-39

Cited by 6 publications

(10 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the test set, the AUC, accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and F1 score were estimated. The best prediction model was selected by evaluating the largest AUC [ 47 , 48 ]. We applied the Shapley Additive Explanation (SHAP) to explain the best-performing predictive model.…”

Section: Methodsmentioning

confidence: 99%

Machine learning-based models for the prediction of breast cancer recurrence risk

Zuo,

Yang,

Jin

et al. 2023

BMC Med Inform Decis Mak

View full text Add to dashboard Cite

Breast cancer is the most common malignancy diagnosed in women worldwide. The prevalence and incidence of breast cancer is increasing every year; therefore, early diagnosis along with suitable relapse detection is an important strategy for prognosis improvement. This study aimed to compare different machine algorithms to select the best model for predicting breast cancer recurrence. The prediction model was developed by using eleven different machine learning (ML) algorithms, including logistic regression (LR), random forest (RF), support vector classification (SVC), extreme gradient boosting (XGBoost), gradient boosting decision tree (GBDT), decision tree, multilayer perceptron (MLP), linear discriminant analysis (LDA), adaptive boosting (AdaBoost), Gaussian naive Bayes (GaussianNB), and light gradient boosting machine (LightGBM), to predict breast cancer recurrence. The area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and F1 score were used to evaluate the performance of the prognostic model. Based on performance, the optimal ML was selected, and feature importance was ranked by Shapley Additive Explanation (SHAP) values. Compared to the other 10 algorithms, the results showed that the AdaBoost algorithm had the best prediction performance for successfully predicting breast cancer recurrence and was adopted in the establishment of the prediction model. Moreover, CA125, CEA, Fbg, and tumor diameter were found to be the most important features in our dataset to predict breast cancer recurrence. More importantly, our study is the first to use the SHAP method to improve the interpretability of clinicians to predict the recurrence model of breast cancer based on the AdaBoost algorithm. The AdaBoost algorithm offers a clinical decision support model and successfully identifies the recurrence of breast cancer.

show abstract

Section: Methodsmentioning

confidence: 99%

Machine learning-based models for the prediction of breast cancer recurrence risk

Zuo,

Yang,

Jin

et al. 2023

BMC Med Inform Decis Mak

View full text Add to dashboard Cite

show abstract

“…10 • Quantitative parameters of DECT may allow estimating the benign or malignant nature of breast masses, such as NIC, λ Hu , nZeff. 10,15,17,38 • When determining the difference between benign and malignant breast lesions or metastatic lymph nodes, venous phase λ Hu has the highest diagnostic value. 12,17,19 To provide as a non-invasive alternate method for staging lymph nodes…”

Section: Application Value Examples Of Clinical Applicationsmentioning

confidence: 99%

“…10,15,17,38 • When determining the difference between benign and malignant breast lesions or metastatic lymph nodes, venous phase λ Hu has the highest diagnostic value. 12,17,19 To provide as a non-invasive alternate method for staging lymph nodes…”

Section: Application Value Examples Of Clinical Applicationsmentioning

confidence: 99%

“…NIC, λ Hu , nZeff and attenuation of malignant breast lesions are higher than those of benign lesions. In the study of DECT quantitative parameters to distinguish benign and malignant breast tumors, Lan et al 17 found that except for nZeff, 11 quantitative parameters of malignant tumours were higher than those of benign tumours. In univariate analysis, the venous phase λ Hu is the most effective in differentiating between benign and malignant breast masses.…”

Section: Dect In Breast Cancer Diagnosismentioning

confidence: 99%

See 1 more Smart Citation

Dual‐Energy CT in Breast Cancer: Current Applications and Future Outlooks

Guo,

Liu,

et al. 2023

Precision Radiation Oncology

View full text Add to dashboard Cite

Breast cancer is the most prevalent cancerous tumor in women, characterized by different subtypes and varying responses to treatment. The continued evolution of breast cancer diagnosis and management has resulted in a transition from a one‐size‐fits‐all approach to a new era of personalized treatment plans. Therefore, it is essential to accurately identify the biological characteristics of breast tissue in order to minimize unnecessary biopsies of benign lesions and improve the overall clinical process, leading to reduced expenses and complications associated with invasive biopsy procedures. Challenges for future research include finding ways to predict the response of breast cancer patients to adjuvant systemic treatment.Dual‐energy CT (DECT) is a new imaging technology integrating functional imaging and molecular imaging. Over the past decade, DECT has gained relevancy, especially in oncological radiology. This article proposed a literature review of the application and research status of DECT in breast cancer treatment strategy determination and prognosis prediction.

show abstract

“…As an imaging technology, dual-energy computed tomography (DECT) or called spectrum CT has the advantages of providing multiple quantitative information for various materials inside the tissue ( 4 ), to clarify benign and malignant tumors ( 5 , 6 ). DECT can provide richer qualitative and quantitative diagnostic information by taking the advantage of covering a wider spectrum of energy parameters to reflect the state of blood supply and angiogenesis in a lesion directly and accurately.…”

Section: Introductionmentioning

confidence: 99%

Quantitative energy spectrum CT in differential diagnosis of aldosterone-producing adenoma and cortisol-producing adenoma

Chen¹,

Zhang²,

Shi³

et al. 2023

Quant Imaging Med Surg

View full text Add to dashboard Cite

Background Patients with aldosterone-producing adenoma (APA) and cortisol-producing adenoma (CPA) show some similar clinical symptoms, and a large overlap of conventional imaging manifestations, which make the differentiation difficult. The purpose of our study was to explore the value of gemstone spectral imaging (GSI) dual-energy computed tomography (DECT) in differential diagnosis of APA and CPA, screen out meaningful energy spectral indicators and provide theoretical basis for the differential diagnosis of the two. Methods We retrospectively analyzed the imaging and clinical data of 30 patients with APA and 27 patients with CPA who underwent GSI DECT in The First Affiliated Hospital of Zhengzhou University (a tertiary care institution). Patients were consecutively enrolled in this study, and the quantitative DECT parameters were compared between the APA and CPA groups by two-sample test. The diagnostic efficacies were evaluated by receiver operating characteristic (ROC) analysis. Results DECT parameters including CT (computed tomography) values at 40–70 keV in the arterial phase, concentrations of I (H 2 O) and fat (I) in the arterial phase, and the effective atomic number in the venous phase, were significantly different between the APA and CPA groups (all P<0.001), and the area under the curve (AUC) values are 0.80, 0.79, 0.88, 0.76, 0.82, 0.87, and 0.86. Conclusions DECT quantitative parameters can effectively identify APA and CPA, the CT values at 40 and 60 keV in the arterial phase, the normalized CT value at 60 keV, the I (H 2 O), fat (I) concentration in the arterial phase and the effective atomic number parameter in the venous phase had valuable diagnostic performance.

show abstract

Application of machine learning with multiparametric dual-energy computed tomography of the breast to differentiate between benign and malignant lesions

Cited by 6 publications

References 33 publications

Machine learning-based models for the prediction of breast cancer recurrence risk

Machine learning-based models for the prediction of breast cancer recurrence risk

Dual‐Energy CT in Breast Cancer: Current Applications and Future Outlooks

Quantitative energy spectrum CT in differential diagnosis of aldosterone-producing adenoma and cortisol-producing adenoma

Contact Info

Product

Resources

About