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
2344 - Lung Cancer in Europe: strengthening policy responses to address unmet needs
Presenter: Mary Bussell
Session: Poster Display session 3
Resources:
Abstract
1359 - Curative treatment timelines for breast, colorectal, lung and prostate cancer: Implications for medical leave coverage
Presenter: Selina Wong
Session: Poster Display session 3
Resources:
Abstract
4433 - Acute Diagnostic Oncology Clinic: A Unique Primary Care-Oncology Service
Presenter: Abhijit Gill
Session: Poster Display session 3
Resources:
Abstract
3506 - THE NEW MUTATIONAL MODEL IN ONCOLOGY. What changes in welfare, clinical practice, research, and regulatory procedures
Presenter: Nicola Normanno
Session: Poster Display session 3
Resources:
Abstract
3350 - Selection of a set of quality indicators (QI) for oncological clinical pathway
Presenter: Aude Fourcade
Session: Poster Display session 3
Resources:
Abstract
4400 - Sustainable drug prices at market launch: policy proposals and their empirical evidence
Presenter: Nora Fanzen
Session: Poster Display session 3
Resources:
Abstract
4118 - Impact of financial considerations on French physicians’ prescription choices for advanced non-small cell lung cancer (NSCLC)
Presenter: Nathalie Olympios
Session: Poster Display session 3
Resources:
Abstract
1340 - The direct medical cost of breast cancer in a Belgian hospital
Presenter: Hannan Lemhouer
Session: Poster Display session 3
Resources:
Abstract
1863 - Does the healthcare system approaches cancer patients for using private services during diagnostic process?
Presenter: Karolina Osowiecka
Session: Poster Display session 3
Resources:
Abstract
2637 - Measuring financial toxicity of cancer in the Italian health care system: initial results of the patient reported outcome for Fighting Financial Toxicity of cancer project (proFFiT).
Presenter: Silvia Riva
Session: Poster Display session 3
Resources:
Abstract