Blind Image Quality Assessment by Pairwise Ranking Image Series

Image quality assessment(IQA)is constantly innovating,but there are still three types of stickers that have not been resolved:the“content sticker”-limitation of training set,the“annotation sticker”-subjective instability in opinion scores and the“distortion sticker”-disordered distortion settings.In this paper,a No-Reference Image Quality Assessment(NR IQA)approach is proposed to deal with the problems.For“content sticker”,we introduce the idea of pairwise comparison and generate a largescale ranking set to pre-train the network;For“annotation sticker”,the absolute noise-containing subjective scores are transformed into ranking comparison results,and we design an indirect unsupervised regression based on EigenValue Decomposition(EVD);For“distortion sticker”,we propose a perception-based distortion classification method,which makes the distortion types clear and refined.Experiments have proved that our NR IQA approach Experiments show that the algorithm performs well and has good generalization ability.Furthermore,the proposed perception based distortion classification method would be able to provide insights on how the visual related studies may be developed and to broaden our understanding of human visual system.

Performance Validation and Analysis for Multi-Method Fusion Based Image Quality Metrics in A New Image Database

Considering that there is no single full reference image quality assessment method that could give the best performance in all situations, some multi-method fusion metrics were proposed. Machine learning techniques are often involved in such multi-method fusion metrics so that its output would be more consistent with human visual perceptions. On the other hand, the robustness and generalization ability of these multi-method fusion metrics are questioned because of the scarce of images with mean opinion scores. In order to comprehensively validate whether or not the generalization ability of such multi-method fusion IQA metrics are satisfying, we construct a new image database which contains up to 60 reference images. The newly built image database is then used to test the generalization ability of different multi-method fusion IQA metrics. Cross database validation experiment indicates that in our new image database, the performances of all the multi-method fusion IQA metrics have no statistical significant different with some single-method IQA metrics such as FSIM and MAD. In the end, a thorough analysis is given to explain why the performance of multi-method fusion IQA framework drop significantly in cross database validation.

Investigation on the short-term outcome and prognostic impact of predisposition,and precipitants in inpatients with chronic liver disease from Chinese AcuTe on CHronic LIver FailurE(CATCH-LIFE)cohorts

Aim:The study aimed to investigate the short-term outcomes of hospitalized patients with chronic liver disease(CLDs)and assess the prognostic impact of predisposition and precipitants,which currently remains unclear.Methods:The study included 3970 hospitalized patients with CLDs from two prospective longitudinal multicenter studies(NCT02457637 and NCT03641872)conducted in highly endemic hepatitis B virus(HBV)areas.Competing risk analysis was used to evaluate the effect of predispositions,including the etiology and severity of CLDs and precipitants;on sequential 28,90,and 365-day liver transplantation(LT)-free mortality.Results:Among all enrolled patients,76.8%of adverse outcomes(including death and LT)within one year occurred within 90 days.Compared with alcoholic etiology,the association of HBV etiology with poorer outcomes was remarkably on the 28th day(hazard ratio[HR],1.81;95%confidence interval[CI],1.07-3.06;p=0.026);however,and dimin-ished or became insignificant at 90 days and 365 days.Cirrhosis increased the adjusted risk for 365-day(HR,1.50;CI,1.13-1.99;p=0.004)LT-free mortality when compared with noncirrhosis.In patients with cirrhosis,prior decompensation(PD)independently increased the adjusted risk of 365-day LT-free mortality by 1.25-fold(p=0.021);however,it did not increase the risk for 90-day mortality.Neither the category nor the number of precipitants influenced the adjusted risk of 28 or 90-day LT-free mortality.Conclusions:The 90-day outcome should be considered a significant endpoint for evaluating the short-term prognosis of hospitalized patients with CLD.Predisposing factors,other than etiology,mainly affected the delayed(365-day)outcome.Timely effective therapy for CLD etiology,especially antiviral treatments for HBV,and post-discharge long-term surveillance monitoring in cirrhotic patients undergoing PD are suggested to enhance disease management and reduce mortality.