Habitat Analysis of Breast Cancer-Enhanced MRI Reflects BRCA1 Mutation Determined by Immunohistochemistry.
Objective. To use habitat analysis (also termed habitat imaging) for classifying untreated breast cancer-enhanced magnetic resonance imaging (MRI) in women. Moreover, we intended to obtain clustering parameters to predict the BReast CAncer gene 1 (BRCA1) gene mutation and to determine the use of MRI...
| Publicado en: | BioMed Research International pp. 1 - 10 |
|---|---|
| Autores principales: | , |
| Formato: | diagnostic images research tables/charts Journal Article |
| Publicado: |
Wiley-Blackwell
3/30/2022
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=156022270&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 156022270 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 23146133 FT2T jtl: BioMed Research International issn: 23146133 maglogo: N pubinfo: dt: 3/30/2022 pid: 480 pub: Wiley-Blackwell place: Malden, Massachusetts artinfo: ui: 156022270 156022270 156022270 10.1155/2022/9623173 156022270 ppf: 1 ppct: 9 formats: fmt: – @attributes: type: T – @attributes: type: P tig: atl: Habitat Analysis of Breast Cancer-Enhanced MRI Reflects BRCA1 Mutation Determined by Immunohistochemistry. aug: au: Du, Tianming Zhao, Haidong affil: Department of Breast Surgery, Second Affiliated Hospital of Dalian Medical University, 467, Zhongshan Road, Shahekou District, Dalian, Liaoning 116023, China sug: subj: Immunohistochemistry Breast Neoplasms Radiography Magnetic Resonance Imaging Evaluation Mutation Breast Neoplasms Diagnosis Human Female Descriptive Statistics Confidence Intervals Odds Ratio Forecasting Algorithms Female ab: Objective. To use habitat analysis (also termed habitat imaging) for classifying untreated breast cancer-enhanced magnetic resonance imaging (MRI) in women. Moreover, we intended to obtain clustering parameters to predict the BReast CAncer gene 1 (BRCA1) gene mutation and to determine the use of MRI as a noninvasive examination tool. Methods. We obtained enhanced MRI data of patients with breast cancer before treatment and selected some sequences as the source of habitat imaging. We used the k -means clustering to classify these images. According to the formed subregions, we calculated several parameters to evaluate the clustering. We used immunohistochemistry to detect BRCA1 mutations. Moreover, we separately determined the ability of these parameters through independent modeling or multiple parameter joint modeling to predict these mutations. Results. Of all extracted values, separation (SP) demonstrated the best prediction performance for a single parameter (area under the receiver operating characteristic curve (AUC), 0.647; 95% confidence interval (CI), 0.557–0.731). Simultaneously, models based on the Calinski-Harabasz Index and sum of square error performed better in the training (AUC, 0.903; 95% CI, 0.831–0.96) and verification (AUC, 0.845; 95% CI, 0.723–0.942) sets for multiparameter joint modeling. Conclusion. Based on the enhanced MRI of breast tumors and the subregions generated according to the habitat imaging theory, the parameters extracted to describe the clustering effect could reflect the BRCA1 status. Differences between clusters, including the general differences of cluster centers and clusters and the similarity of samples within clusters, were the embodiment of this mutation. We propose an algorithm to predict the BRCA1 mutation of a patient according to the enhanced MRI of the breast tumor. pubtype: Academic Journal doctype: diagnostic images research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|