Accepted for/Published in: JMIR Biomedical Engineering
Date Submitted: Aug 24, 2020
Open Peer Review Period: Aug 24, 2020 - Nov 9, 2020
Date Accepted: Apr 4, 2021
(closed for review but you can still tweet)
Neural network pattern recognition of ultrasound image gray scale intensity histogram of breast lesions to differentiate between benign and malignant lesions : An analytical study
ABSTRACT
Background:
Ultrasound based radiomic features to differentiate between benign and malignant breast lesion with the help of machine learning is being researched currently. Mean echogenicity ratio has been used for diagnosis of malignant breast lesions. However grey scale intensity histogram values as a single radiomic feature for detection of malignant breast lesions using machine learning algorithms has not been explored yet.
Objective:
Assessing the utility of a simple convoluted neural network in classifying benign and malignant breast lesion using grey scale intensity values of the lesion.
Methods:
Open access online dataset of 200 USG (Ultrasonogram) breast lesion were collected and ROI (Region of interest) was drawn over the lesions. The grey scale intensity values of the lesions were extracted. An input file containing the values and an output file consisting of the breast lesions’ diagnosis were created. The CNN (Convoluted neural network) was trained using the files and tested on the whole dataset.
Results:
The results were impressive. The trained CNN had an accuracy of 94.5% and a precision of 94%. The sensitivity and specificity were 94.9% and 94.1% respectively.
Conclusions:
Simple neural networks which are cheap and easy to use can be applied to diagnose malignant breast lesion with grey scale intensity values obtained from USG images in low resource settings with minimal manpower.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.