Abstract 4035
Background
Breast cancer is one of the most common malignant disease for women. Mammography is the preferred method for breast cancer detection. The purpose is to investigate the feasibility and accuracy of texture features extracted from digital mammograms at predicting benign and malignant breast mass using Radiomics.
Methods
494 digital mammograms data who diagnosed as breast masses (Benign: 251 Malignant: 243) by mammography were enrolled. Enrol criteria: breast masses classified as BI-RADS 3, 4, and 5 and at last confirmed by histopathology. Lesion area was marked with a rectangular frame on the Cranio-Caudal (CC) and MedioLateral Oblique (MLO) images at the 5M workstation. The rectangular regions of interest (ROI) was segmented and 456 radiomics features were extracted from every ROI. Extracted features were dimensioned by Maximum Relevance Minimum Redundancy (MRMR) and Lasso algorithm. Post-dimension features were classified using Support Vector Machine (SVM). 70% of the data as a training set and the other 30% as a testing set. The reliability of the Classifier was evaluated by the 10-fold cross-validation. The classification accuracy was evaluated by the accuracy and sensitivity and AUC.
Results
Both the MRMR and Lasso screened 30 radiomics features respectively. 10-fold cross-validation showed that their accuracy were 88.70% and 86.71%, respectively. In testing sets, Through the MRMR algorithm, the classifier achieves an accuracy of 92.00% and a sensitivity of 91.10% and AUC of 95.10%. Through the lasso dimension reduction algorithm, the classifier achieves an accuracy of 83.26% and a sensitivity of 75.90% and AUC of 89.38%.
Conclusions
Radiomics texture features from digital mammograms may be used for benign and malignant prediction. This method offer better accuracy and sensitivity. It is expected to provide an auxiliary diagnosis for the imaging doctors.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.
Resources from the same session
5760 - Landscape of PD-L1 expression status in Chinese solid tumor patients.
Presenter: Yi Zhong
Session: Poster Display session 3
Resources:
Abstract
3733 - Anti-cancer and immunomodulatory effects of cobimetinib in triple negative breast cancer
Presenter: Chun-Yu Liu
Session: Poster Display session 3
Resources:
Abstract
4426 - Differential expression of immunoregulatory molecules and highly-associated cancer genes may provide novel insights into strategic trial design for therapeutics
Presenter: Jacob Adashek
Session: Poster Display session 3
Resources:
Abstract
2752 - Insights into the Tumor Immune Microenvironment using Tissue Phenomics to Drive Cancer Immunotherapy
Presenter: Martin Groher
Session: Poster Display session 3
Resources:
Abstract
5713 - Immune competent somatic mosaic model of colorectal cancer
Presenter: Stefania Napolitano
Session: Poster Display session 3
Resources:
Abstract
1898 - Genomic correlates of response to anti-PDL1 Atezolizumab in non-small-cell lung cancer OAK and POPLAR trials
Presenter: Hari Singhal
Session: Poster Display session 3
Resources:
Abstract
3246 - Erdafitinib (erda) versus available therapies in advanced urothelial cancer: A matching adjusted indirect comparison
Presenter: Yohann Loriot
Session: Poster Display session 3
Resources:
Abstract
3311 - High level of activity of Nivolumab anti-PD-1 immunotherapy and favorable outcome in metastatic/refractory MSI-H non-colorectal cancer: Results of the MSI cohort from the French AcSé program
Presenter: Christophe Tournigand
Session: Poster Display session 3
Resources:
Abstract
2314 - TP53 and ATM Co-mutation Predicts Response to Immune Checkpoint Inhibitors in Non-Small Cell Lung Cancer
Presenter: Yu Chen
Session: Poster Display session 3
Resources:
Abstract
4692 - Immune cell biomarkers on neo-adjuvant chemo-immunotherapy treatment for resectable stage IIIA NSCLC patients
Presenter: Raquel Laza-Briviesca
Session: Poster Display session 3
Resources:
Abstract