Predicting adult Attention Deficit Hyperactivity Disorder (ADHD) using vocal acoustic features

Polier, Georg G. von; Ahlers, Eike; Amunts, Julia; Langner, J; Patil, Kiran Raosaheb; Eickhoff, SB; Helmhold, Florian; Langner, Daina

doi:10.1101/2021.03.18.21253108

Cited by 5 publications

(6 citation statements)

References 90 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To highlight the danger of confound-leakage on this clinically relevant task, we analyzed a dataset with speech-derived features with the task to distinguish individuals with ADHD from controls. Our version of the dataset is a balanced subsample of the dataset described by von Polier et al [ 3 ].…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Confound-leakage: confound removal in machine learning leads to leakage

Hamdan,

Love,

von Polier

et al. 2022

GigaScience

Self Cite

View full text Add to dashboard Cite

Background Machine learning (ML) approaches are a crucial component of modern data analysis in many fields, including epidemiology and medicine. Nonlinear ML methods often achieve accurate predictions, for instance, in personalized medicine, as they are capable of modeling complex relationships between features and the target. Problematically, ML models and their predictions can be biased by confounding information present in the features. To remove this spurious signal, researchers often employ featurewise linear confound regression (CR). While this is considered a standard approach for dealing with confounding, possible pitfalls of using CR in ML pipelines are not fully understood. Results We provide new evidence that, contrary to general expectations, linear confound regression can increase the risk of confounding when combined with nonlinear ML approaches. Using a simple framework that uses the target as a confound, we show that information leaked via CR can increase null or moderate effects to near-perfect prediction. By shuffling the features, we provide evidence that this increase is indeed due to confound-leakage and not due to revealing of information. We then demonstrate the danger of confound-leakage in a real-world clinical application where the accuracy of predicting attention-deficit/hyperactivity disorder is overestimated using speech-derived features when using depression as a confound. Conclusions Mishandling or even amplifying confounding effects when building ML models due to confound-leakage, as shown, can lead to untrustworthy, biased, and unfair predictions. Our expose of the confound-leakage pitfall and provided guidelines for dealing with it can help create more robust and trustworthy ML models.

show abstract

Section: Resultsmentioning

confidence: 99%

“…Imagine building a diagnostic classifier for attention-deficit/hyperactivity disorder (ADHD) based on speech patterns. This will be a useful clinical tool aiding objective diagnosis [ 3 ]. However, like most disorders, ADHD has comorbidity, for instance, with depression.…”

Section: Introductionmentioning

confidence: 99%

Confound-leakage: confound removal in machine learning leads to leakage

Hamdan,

Love,

von Polier

et al. 2022

GigaScience

Self Cite

View full text Add to dashboard Cite

show abstract

“…We also used one clinical dataset, a balanced subsample of the ADHD speech dataset described by von Polier et al [1] includes 126 individuals with 6016 speech-related features, the binary target describing ADHD status (ADHD or control) and contains four confounds: gender, education level, age and, depression score measured using the Beck's depression inventory (BDI). For more information on the datasets see Supplementary Table S1.…”

Section: Datamentioning

confidence: 99%

“…Imagine building a diagnostic classifier for attention deficit hyperactivity disorder (ADHD) based on speech patterns. This will be a useful clinical tool aiding objective diagnosis [1]. However, like most disorders, ADHD has comorbidity, for instance with depression.…”

Section: Introductionmentioning

confidence: 99%

Confound-leakage: Confound Removal in Machine Learning Leads to Leakage

Hamdan¹,

Love²,

Polier³

et al. 2022

Preprint

View full text Add to dashboard Cite

Machine learning (ML) approaches to data analysis are now widely adopted in many fields including epidemiology and medicine. For meaningful application of these approaches, confounds must first be removed as is commonly done by featurewise removal of their variance by linear regression before applying ML. Here, we show this common approach to confound removal biases ML models, leading to misleading results. Specifically, null or moderate effects are erroneously turned in to near-perfect prediction due to leaked information after deconfounding when nonlinear ML approaches are subsequently applied. We identify and evaluate possible mechanisms for such confound-leakage and provide practical guidance to mitigate its negative impact. We demonstrate the danger of confound-leakage in a real-world clinical application where the accuracy of predicting attention deficit hyperactivity disorder (ADHD) is overestimated when using depression as a confound. Our results have wide-reaching implications for implementation and deployment of ML workflows and beg caution against naïve use of standard confound removal approaches.

show abstract

“…Prosodic variations in loudness and fundamental frequency [1] Amyotrophic Lateral Sclerosis (ALS) Voice tremor, flutter [2], incomplete vocal fold closure, dysarthria [3]; Dystonia, dysarthria [4]; Low-frequency (< 4 hz) tremor [5] Alzheimer's Disease (Dementia) Abnormal fundamental frequency, pause and voice-break patterns, reduction in vocal range [6,7]; Dysphonia [8] Arthritis: Fibromyalgia Changes in Jitter, shimmer, harmonic-to-noise ratio, and phonation time [9] Cerebral Palsy Dysphonia [10]; Breathiness, Asthenia, Roughness, Strain [11]…”

Section: How It Affects Voice Attention Deficit Hyperactivity Disorde...mentioning

confidence: 99%

An Overview of Techniques for Biomarker Discovery in Voice Signal

Singh,

Shah,

Dhamyal

2021

Preprint

View full text Add to dashboard Cite

This paper reflects on the effect of several categories of medical conditions on human voice, focusing on those that may be hypothesized to have effects on voice, but for which the changes themselves may be subtle enough to have eluded observation in standard analytical examinations of the voice signal. It presents three categories of techniques that can potentially uncover such elusive biomarkers and allow them to be measured and used for predictive and diagnostic purposes. These approaches include proxy techniques, model-based analytical techniques and data-driven AI techniques.

show abstract

Predicting adult Attention Deficit Hyperactivity Disorder (ADHD) using vocal acoustic features

Cited by 5 publications

References 90 publications

Confound-leakage: confound removal in machine learning leads to leakage

Confound-leakage: confound removal in machine learning leads to leakage

Confound-leakage: Confound Removal in Machine Learning Leads to Leakage

An Overview of Techniques for Biomarker Discovery in Voice Signal

Contact Info

Product

Resources

About