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
4097 - Targeting NRG1-fusions in multiple tumour types: Afatinib as a novel potential treatment option
Presenter: Stephen V Liu
Session: Poster Display session 3
Resources:
Abstract
1129 - Aspirin and Ticagrelor for the prevention of tumour cell induced platelet aggregation
Presenter: Meera Chauhan
Session: Poster Display session 3
Resources:
Abstract
4514 - Pharmacokinetic/ pharmacodynamic (PK/PD) exposure-response characterization of GSK3359609 (GSK609) from INDUCE-1, a phase I open-label study
Presenter: Michele Maio
Session: Poster Display session 3
Resources:
Abstract
5169 - In vitro functional interrogation of viable Circulating Tumor Associated Cells (C-TACs) for evaluating Platin resistance.
Presenter: Stefan Schuster
Session: Poster Display session 3
Resources:
Abstract
5827 - Targeting ARG2 as a novel therapeutic approach for cancer
Presenter: Marcin Grzybowski
Session: Poster Display session 3
Resources:
Abstract
3129 - MPS1 and PLK1 as new therapy targets in TP53 mutated solid tumors
Presenter: Balazs Gyorffy
Session: Poster Display session 3
Resources:
Abstract
2129 - The Tumor Static Exposure (TSE) concept & utility: application to combination treatment of radiation and radiosensitizing agent in tumor xenograft experiments
Presenter: Samer El Bawab
Session: Poster Display session 3
Resources:
Abstract
1814 - General Methodology to Optimize Tumor Treating Fields Delivery Utilizing Numerical Simulations
Presenter: Noa Urman
Session: Poster Display session 3
Resources:
Abstract
3010 - The Australian Exceptional Responders Program: a National collaboration
Presenter: Megan Barnet
Session: Poster Display session 3
Resources:
Abstract
4489 - A Window of Opportunity Trial of Atorvastatin Targeting p53 Mutant Malignancies
Presenter: Joaquina Baranda
Session: Poster Display session 3
Resources:
Abstract