Introduction The depiction of features in discourse production promotes accurate diagnosis and helps to establish the therapeutic intervention in cognitive impairment and dementia. We aimed to identify alterations in the macrolinguistic aspects of discourse using a new computational tool. Methods Sixty individuals, aged 60 years and older, were distributed in three different groups: mild Alzheimer's disease (mAD), amnestic mild cognitive impairment, and healthy controls. A narrative created by individuals was analyzed through the Coh-Metrix-Dementia program, extracting the features of interest automatically. Results mAD showed worse overall performance compared to the other groups: less informative discourse, greater impairment in global coherence, greater modalization, and inferior narrative structure. It was not possible to discriminate between amnestic mild cognitive impairment and healthy controls. Discussion Our results are in line with the literature, verifying a pathological change in the macrostructure of discourse in mAD.

Mild Cognitive Impairment (MCI) is a mental disorder difficult to diagnose. Linguistic features, mainly from parsers, have been used to detect MCI, but this is not suitable for large-scale assessments. MCI disfluencies produce nongrammatical speech that requires manual or high precision automatic correction of transcripts. In this paper, we modeled transcripts into complex networks and enriched them with word embedding (CNE) to better represent short texts produced in neuropsychological assessments. The network measurements were applied with well-known classifiers to automatically identify MCI in transcripts, in a binary classification task. A comparison was made with the performance of traditional approaches using Bag of Words (BoW) and linguistic features for three datasets: DementiaBank in English, and Cinderella and Arizona-Battery in Portuguese. Overall, CNE provided higher accuracy than using only complex networks, while Support Vector Machine was superior to other classifiers. CNE provided the highest accuracies for DementiaBank and Cinderella, but BoW was more efficient for the Arizona-Battery dataset probably owing to its short narratives. The approach using linguistic features yielded higher accuracy if the transcriptions of the Cinderella dataset were manually revised. Taken together, the results indicate that complex networks enriched with embedding is promising for detecting MCI in large-scale assessments.

This paper describes the results of NILC team at CWI 2018. We developed solutions following three approaches: (i) a feature engineering method using lexical, n-gram and psycholinguistic features, (ii) a shallow neural network method using only word embeddings, and (iii) a Long Short-Term Memory (LSTM) language model, which is pre-trained on a large text corpus to produce a contextualized word vector. The feature engineering method obtained our best results for the classification task and the LSTM model achieved the best results for the probabilistic classification task. Our results show that deep neural networks are able to perform as well as traditional machine learning methods using manually engineered features for the task of complex word identification in English. * The opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Itaú-Unibanco.

No abstract