Search

• Relatively large sample sizes of low-function autistic children showed differences in EEG power, entropy, coherence and bicoherence than TD children. • Multi-features were used to distinguish low-function autistic and TD children accurately. • The satisfied classification accuracy is 95.67%. Autism spectrum disorder (ASD) is a very serious neurodevelopmental disorder and diagnosis mainly depends on the clinical scale, which has a certain degree of subjectivity. It is necessary to make accurate evaluation by objective indicators. In this study, we enrolled 96 children aged from 3 to 6 years: 48 low-function autistic children (38 males and 10 females; mean ± SD age: 4.9 ± 1.1 years) and 48 typically developing (TD) children (38 males and 10 females; mean ± SD age: 4.9 ± 1.2 years) to participate in our experiment. We investigated to fuse multi-features (entropy, relative power, coherence and bicoherence) to distinguish low-function autistic children and TD children accurately. Minimum redundancy maximum correlation algorithm was used to choose the features and support vector machine was used for classification. Ten-fold cross validation was used to test the accuracy of the model. Better classification result was obtained. We tried to provide a reliable basis for clinical evaluation and diagnosis for ASD. [ABSTRACT FROM AUTHOR]

Glaucoma is the leading cause of irreversible blindness, and the early detection and timely treatment are essential for glaucoma management. However, due to the interindividual variability in the characteristics of glaucoma onset, a single feature is not yet sufficient for monitoring glaucoma progression in isolation. There is an urgent need to develop more comprehensive diagnostic methods with higher accuracy. In this study, we proposed a multi- feature deep learning (MFDL) system based on intraocular pressure (IOP), color fundus photograph (CFP) and visual field (VF) to classify the glaucoma into four severity levels. We designed a three-phase framework for glaucoma severity diagnosis from coarse to fine, which contains screening, detection and classification. We trained it on 6,131 samples from 3,324 patients and tested it on independent 240 samples from 185 patients. Our results show that MFDL achieved a higher accuracy of 0.842 (95 % CI, 0.795-0.888) than the direct four classification deep learning (DFC-DL, accuracy of 0.513 [0.449-0.576]), CFP-based single-feature deep learning (CFP-DL, accuracy of 0.483 [0.420-0.547]) and VF-based single-feature deep learning (VF-DL, accuracy of 0.725 [0.668-0.782]). Its performance was statistically significantly superior to that of 8 juniors. It also outperformed 3 seniors and 1 expert, and was comparable with 2 glaucoma experts (0.842 vs 0.854, p = 0.663; 0.842 vs 0.858, p = 0.580). With the assistance of MFDL, junior ophthalmologists achieved statistically significantly higher accuracy performance, with the increased accuracy ranged from 7.50 % to 17.9 %, and that of seniors and experts were 6.30 % to 7.50 % and 5.40 % to 7.50 %. The mean diagnosis time per patient of MFDL was 5.96 s. The proposed model can potentially assist ophthalmologists in efficient and accurate glaucoma diagnosis that could aid the clinical management of glaucoma. [ABSTRACT FROM AUTHOR]