Autism spectrum disorder (ASD) is an abstruse brain disorder in neuroscience research, which lead by challenges of social interactions, speech and nonverbal communication, and repetitive behaviors. Diagnosis of ASD is mostly based on behavioral analysis, which is time-consuming and depends on patient cooperation and examiner expertise. Recently, machine learning(ML) based algorithms are applied for ASD identification. However, there are some limitations using ML approaches such as investigation with functional magnetic resonance imaging (fMRI), depend on region-based analysis, and handling with the big dataset. In this work, we proposed a novel architecture based on the Bag-of-Features model for ASD classification to overcome these challenges. Firstly, we preprocess the images to extract the speeded-up robust features (SURF) from the selected feature point locations. The Bag-of-Feature extraction procedures include feature concatenation, select the most robust feature, and convert to feature vector. After that, we employ K-Means clustering to create a word visual vocabulary from the SURF vector. Then, we encode the Bag-of-Features by adopting coding and quantization techniques to get each class's indexed database. We prefer most tremendous machine learning methods SVM as ASD classifier. Finally, we independently evaluate our proposed architecture's performance using three different datasets from different fields, including ABIDE fMRI preprocessed images and subject's face images. In our experiments, weigh against other state-of-the-art methods that our ML classifiers with Bag-of-Feature extractors reinforce in medication and clinical purposes of ASD.
In this paper, a linear discriminant analysis method with L1 norm (LDA-L1) for palmprint recognition is proposed. The traditional linear discriminant analysis method based on L2 norm is very sensitive to outliers, but the L1 specification can overcome this problem very well. In the LDA-L1 method, a series of projection vectors are obtained by the iterative method, which can maximize the inter class dispersion and minimize the L1 norm based on within class dispersion. We tested the performance of our approach in the PolyU palmprint database. The experimental results show that LDA-L1 has robustness to outliers.