|DementiaBank||ADReSSo 2021 Challenge|
This challenge was organized by Saturnino Luz, Fasih Haider, and Sofia de la Fuente Garcia of the University of Edinburgh and Davida Fromm and Brian MacWhinney of Carnegie Mellon University.
The objective of the ADReSSo-2021 challenge is to make available a benchmark dataset of spontaneous speech, which is acoustically pre-processed and balanced in terms of age and gender, defining a shared task through which different approaches to AD recognition in spontaneous speech can be compared. Our JAD systematic review describes the state of research as background for the challenge.
Dementia is a category of neurodegenerative diseases that entails a long-term and usually gradual decrease of cognitive functioning. The main risk factor for dementia is age, and therefore its greatest incidence is among the elderly. Due to the severity of the situation worldwide, institutions and researchers are investing considerably on dementia prevention and early detection, focusing on disease progression. There is a need for cost-effective and scalable methods for detection of dementia from its most subtle forms, such as the preclinical stage of Subjective Memory Loss (SML), to more severe conditions like Mild Cognitive Impairment (MCI) and Alzheimer's Dementia (AD) itself.
The main features of the ADReSSo (ADReSS, speech only) Challenge are:
The ADReSSo challenge consists of the following tasks:
You may choose to do one or more of these tasks. You will be provided with access to training and test sets.
You must first join as a DementiaBank member.
For the diagnosis task you can use this training data
For the progression tasks you can use this training data
For the diagnosis task you can use this testing data
For the progression tasks you can use this testing data
The training data consists of three folders of data (full enhanced audio, normalised sub-chunks, transcriptions). There are also .csv files with information on age, gender and MMSE scores for Task 1 and Tasks 2 and 3.
The training data are organised into two folders: diagnosis and progression. The diagnosis/train/audio/ad folder contains speech from speakers with Alzheimer's dementia diagnosis. The diagnosis/train/audio/cn folder contains speech from controls. The progression/train/audio/decline folrder contains baseline data from patients who exhibited cognitive dcline between their baseline assessment and their year-2 visit to the clinic. The progression/train/audio/no_decline folder has speech from patients with no decline during that period. Decline is defined as a difference in MMSE score between baseline and year-2 greater than or equal to 5 points.
For the AD/CN diagnosis task and the MMSE predication task, each sub-directory contains compressed (ZIP) archives with recordings of a picture description task ("Cookie Theft" picture from the Boston Diagnostic Aphasia Exam). Those recodings have been acoustically enhanced (noise reduction through spectral subtraction) and normalised. The directory structure and files for the disease progression prediction task are similarly organised. They consists of recordings of a laguage fluency task, also normalised and acoustically enhanced.
The diagnosis task dataset has been balanced with respect to age and gender in order to eliminate potential confunding and bias. We employed a propensity score approach to matching (Rosenbaum & Rubin, 1983; Rubin, 1973; Ho et al., 2007). The dataset was checked for matching according to scores defined in terms of the probability of an instance being treated as AD given covariates age and gender, estimated through logistic regression, and matching instances were selected. All standardized mean differences for the covariates were well below 0.1 and all standardized mean differences for squares and two-way interactions between covariates were well below 0.15, indicating adequate balance for the covariates. The propensity score was estimated using a probit regression of the treatment on the covariates 'age' and 'gender' (probit generated a better balanced than logistic regression).
A full description of the ADReSSo Challenge and its datasets, along with a basic set of baseline results can be found in this paper
The Challenge papers submitted for INTERSPEECH2021 are combined into this PDF.
Several Challenge papers, along with related work, are compiled in this Frontiers Research Topic special issue.
The ADReSSo Challenge acknowledges the support and sponsorship of the European Union's Horizon 2020 research programme, under grant agreement No 769661, towards the SAAM project.