Medical image quality assessment (MIQA) is highly related to content interpretation and disease diagnosis in medical community. However, a few metrics have been developed. On the contrary, massive models have been designed for natural image quality assessment (NIQA) in the field of computer vision. Connect both sides of MIQA and NIQA is useful and challenging. This study explores signal-to-noise ratio (SNR) as the intermediate metric to bridge the gap between MIQA and NIQA and consequently, models for NIQA can be employed or modified for MIQA applications. A number of 411 images from 4 magnetic resonance (MR) imaging sequences are collected. First, the consistency of SNR in MIQA is validated which involves inter-rater and intra-rater (inter-session) reliability analysis. Then, 4 NIQA models (BIQI, BLIINDS-II, BRISQUE and NIQE) are evaluated on these MR images. After that, the correlation between SNR values and NIQA results are analyzed. Statistical analysis indicates that SNR measurement shows reliability regard to different raters in each sequence. Moreover, BLIINDS-II and BRISQUE have the potential for automated MIQA tasks. This study attempts to use SNR bridging the gap between MIQA and NIQA, and a large-scale experiment should be further conducted to verify the conclusion.
Our previous study proposed a shallow convolutional neural network (CNN) to quantify the sharpness of natural images. The network utilized a multilayer perceptron (MLP) as its regression function in the full-connection layer. In this paper, we make use of a polynomial mapping (the logistic map, LM) as the regression function in the natural image sharpness assessment (NISA). First, the coefficient of logistic map is experimentally determined based on the database of LIVE-II. Then, the prediction performance is evaluated on Gaussian blurred images from CSIQ and TID2013. After that, three regression functions, LM, BCF (the basic cubical function) and MLP, are evaluated with Pearson linear correlation coefficient (PLCC) and Spearman rank-order correlation coefficient (SROCC). In addition, eleven state-of-the-art NISA models are compared. Based on the same shallow CNN architecture, experimental results indicate that MLP achieves the best performance, followed by BCF and LM. Furthermore, its performance is rival to or better than other NISA models. Conclusively, in comparison to LM and BCF, MLP is relatively better as a regression function for automatic network optimization and numerical regression. Meanwhile, it achieves the state-of-the-art performance in NISA task.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.