Yunzhi Wang1, Faranak Aghaei1, Ali Zarafshani1, Yuchen Qiu1, Wei Qian2,3, Bin Zheng1. 1. School of Electrical and Computer Engineering, University of Oklahoma, Norman, OK, USA. 2. Department of Electrical and Computer Engineering, University of Texas, El Paso, TX, USA. 3. Sino-Dutch Biomedical and Information Engineering School, Northeastern University, Shenyang, China.
Abstract
PURPOSE: To develop a new computer-aided diagnosis (CAD) scheme that computes visually sensitive image features routinely used by radiologists to develop a machine learning classifier and distinguish between the malignant and benign breast masses detected from digital mammograms. METHODS: An image dataset including 301 breast masses was retrospectively selected. From each segmented mass region, we computed image features that mimic five categories of visually sensitive features routinely used by radiologists in reading mammograms. We then selected five optimal features in the five feature categories and applied logistic regression models for classification. A new CAD interface was also designed to show lesion segmentation, computed feature values and classification score. RESULTS: Areas under ROC curves (AUC) were 0.786±0.026 and 0.758±0.027 when to classify mass regions depicting on two view images, respectively. By fusing classification scores computed from two regions, AUC increased to 0.806±0.025. CONCLUSION: This study demonstrated a new approach to develop CAD scheme based on 5 visually sensitive image features. Combining with a "visual aid" interface, CAD results may be much more easily explainable to the observers and increase their confidence to consider CAD generated classification results than using other conventional CAD approaches, which involve many complicated and visually insensitive texture features.
PURPOSE: To develop a new computer-aided diagnosis (CAD) scheme that computes visually sensitive image features routinely used by radiologists to develop a machine learning classifier and distinguish between the malignant and benign breast masses detected from digital mammograms. METHODS: An image dataset including 301 breast masses was retrospectively selected. From each segmented mass region, we computed image features that mimic five categories of visually sensitive features routinely used by radiologists in reading mammograms. We then selected five optimal features in the five feature categories and applied logistic regression models for classification. A new CAD interface was also designed to show lesion segmentation, computed feature values and classification score. RESULTS: Areas under ROC curves (AUC) were 0.786±0.026 and 0.758±0.027 when to classify mass regions depicting on two view images, respectively. By fusing classification scores computed from two regions, AUC increased to 0.806±0.025. CONCLUSION: This study demonstrated a new approach to develop CAD scheme based on 5 visually sensitive image features. Combining with a "visual aid" interface, CAD results may be much more easily explainable to the observers and increase their confidence to consider CAD generated classification results than using other conventional CAD approaches, which involve many complicated and visually insensitive texture features.
Entities:
Keywords:
Computer-aided diagnosis (CAD); classification of mammographic masses; quantification of visually sensitive image features; quantitative image feature selection in CAD
Authors: Morteza Heidari; Seyedehnafiseh Mirniaharikandehei; Wei Liu; Alan B Hollingsworth; Hong Liu; Bin Zheng Journal: IEEE Trans Med Imaging Date: 2019-10-09 Impact factor: 10.048
Authors: Abolfazl Zargari; Yue Du; Morteza Heidari; Theresa C Thai; Camille C Gunderson; Kathleen Moore; Robert S Mannel; Hong Liu; Bin Zheng; Yuchen Qiu Journal: Phys Med Biol Date: 2018-08-06 Impact factor: 3.609
Authors: Rebecca Sawyer Lee; Jared A Dunnmon; Ann He; Siyi Tang; Christopher Ré; Daniel L Rubin Journal: J Biomed Inform Date: 2020-12-11 Impact factor: 6.317
Authors: Morteza Heidari; Sivaramakrishnan Lakshmivarahan; Seyedehnafiseh Mirniaharikandehei; Gopichandh Danala; Sai Kiran R Maryada; Hong Liu; Bin Zheng Journal: IEEE Trans Biomed Eng Date: 2021-08-19 Impact factor: 4.756