A Deep Learning Approach for Missing Data Imputation of Rating Scales Assessing Attention-Deficit Hyperactivity Disorder

Background: There is an ongoing discussion whether thyroid hormones are involved in the development and course of attention deficit/hyperactivity disorder (ADHD). Since obesity is associated with both higher thyroid-stimulating hormone (TSH) and free triiodothyronine (fT3) concentrations and increased rates of ADHD, we hypothesized that overweight children with ADHD show higher TSH and fT3 concentrations compared to overweight children without ADHD. Methods: TSH, fT3, fT4, and leptin levels were analyzed in 230 children (60.9% boys, 9.3 ± 1.7 years old, 35.7% migration background). The children were divided into four groups (I = 26 overweight children with ADHD, II = 56 normal-weight children with ADHD, III = 66 overweight children without ADHD, and IV = 82 normal-weight children without ADHD). Severity of ADHD was determined by the parent version of the Connors 3® rating scales. Results: Overweight children with ADHD did not differ significantly from overweight children without ADHD with respect to TSH, fT3, or fT4 concentrations. Comparing the thyroid hormones between the four groups also demonstrated no significant differences for TSH and fT4 concentrations. fT3 concentrations were significantly higher in normal-weight children with ADHD compared to normal-weight children without ADHD. Inattention and hyperactivity/impulsivity scores were not significantly related to TSH or fT3 in multiple regression analyses adjusted for age, gender, and migration background. In these analyses, TSH was associated with BMI SDS (β coefficient 0.19 ± 0.12, p = 0.002) and leptin (exp[β coefficient] 1.87 ± 1.36, p < 0.001). fT3 (β coefficient 0.06 ± 0.05, p = 0.009) and leptin (exp[β coefficient] 1.17 ± 1.13, p = 0.009) were also associated with BMI SDS. Conclusions: Our findings confirm the relation between overweight and thyroid hormones but point against the hypothesis that thyroid hormones might link overweight and ADHD in children.

Download Full-text

A Novel Missing Data Imputation Algorithm for Deep Learning-Based Anomaly Detection System in IIoT Networks

10.1201/9781003156123-2 ◽

2021 ◽

pp. 27-46

Author(s):

Ancy Jose ◽

S.V. Annlin Jeba ◽

Beulah Joslyn Jose

Keyword(s):

Deep Learning ◽

Missing Data ◽

Anomaly Detection ◽

Detection System ◽

Data Imputation ◽

Missing Data Imputation ◽

Anomaly Detection System

Download Full-text

Complex Data Imputation by Auto-Encoders and Convolutional Neural Networks—A Case Study on Genome Gap-Filling

Computers ◽

10.3390/computers9020037 ◽

2020 ◽

Vol 9 (2) ◽

pp. 37 ◽

Cited By ~ 1

Author(s):

Luca Cappelletti ◽

Tommaso Fontana ◽

Guido Walter Di Donato ◽

Lorenzo Di Tucci ◽

Elena Casiraghi ◽

...

Keyword(s):

Deep Learning ◽

Missing Data ◽

State Of The Art ◽

The State ◽

Complex Data ◽

Data Imputation ◽

Genome Sequences ◽

Missing Data Imputation ◽

The Past ◽

Learning Techniques

Missing data imputation has been a hot topic in the past decade, and many state-of-the-art works have been presented to propose novel, interesting solutions that have been applied in a variety of fields. In the past decade, the successful results achieved by deep learning techniques have opened the way to their application for solving difficult problems where human skill is not able to provide a reliable solution. Not surprisingly, some deep learners, mainly exploiting encoder-decoder architectures, have also been designed and applied to the task of missing data imputation. However, most of the proposed imputation techniques have not been designed to tackle “complex data”, that is high dimensional data belonging to datasets with huge cardinality and describing complex problems. Precisely, they often need critical parameters to be manually set or exploit complex architecture and/or training phases that make their computational load impracticable. In this paper, after clustering the state-of-the-art imputation techniques into three broad categories, we briefly review the most representative methods and then describe our data imputation proposals, which exploit deep learning techniques specifically designed to handle complex data. Comparative tests on genome sequences show that our deep learning imputers outperform the state-of-the-art KNN-imputation method when filling gaps in human genome sequences.

Download Full-text

Comparison of a standard psychiatric evaluation to rating scales and EEG in the differential diagnosis of attention-deficit/hyperactivity disorder

Psychiatry Research ◽

10.1016/j.psychres.2006.04.015 ◽

2007 ◽

Vol 152 (2-3) ◽

pp. 211-222 ◽

Cited By ~ 25

Author(s):

Humberto Quintana ◽

Steven M. Snyder ◽

William Purnell ◽

Carolina Aponte ◽

Janis Sita

Keyword(s):

Attention Deficit Hyperactivity Disorder ◽

Differential Diagnosis ◽

Attention Deficit ◽

Rating Scales ◽

Psychiatric Evaluation ◽

Hyperactivity Disorder

Download Full-text

Review of Lisdexamfetamine Dimesylate in Adults With Attention-Deficit/Hyperactivity Disorder

Journal of Central Nervous System Disease ◽

10.1177/1179573517728090 ◽

2017 ◽

Vol 9 ◽

pp. 117957351772809 ◽

Cited By ~ 3

Author(s):

Jadwiga Najib ◽

Dexter Wimer ◽

Julie Zeng ◽

Kristina W Lam ◽

Natalya Romanyak ◽

...

Keyword(s):

Attention Deficit Hyperactivity Disorder ◽

Attention Deficit ◽

English Language ◽

Rating Scales ◽

Post Dose ◽

Lisdexamfetamine Dimesylate ◽

Global Improvement ◽

Hyperactivity Disorder ◽

Search Terms ◽

Long Acting

Lisdexamfetamine dimesylate (LDX) is the first prodrug stimulant used for the treatment of attention-deficit/hyperactivity disorder (ADHD) dosed once daily. Due to its long-acting properties, LDX remains pharmacologically inactive until an enzymatic process predominantly associated with red blood cells converts it to the active ingredient, d-amphetamine and the amino acid lysine. The efficacy of LDX over placebo has been demonstrated in several studies in adults with moderate to severe ADHD with significant improvements noted in ADHD rating scales, Clinical Global Improvement scores, and assessments of executive function, for all doses of LDX (30-70 mg daily). Lisdexamfetamine dimesylate has demonstrated efficacy at 14 hours post dose in adults and may be used as a long-acting stimulant for managing ADHD symptoms, which may extend late into the day. Lisdexamfetamine dimesylate has demonstrated a safety profile consistent with long-acting stimulants use. Relevant English language articles were identified through computerized searches of MEDLINE (PubMed and EMBASE) from 1995 to 2016 using the following search terms: lisdexamfetamine dimesylate, attention-deficit hyperactivity disorder, NRP104, and Vyvanse.

Download Full-text

Attention-deficit hyperactivity disorder in adults

Advances in Psychiatric Treatment ◽

10.1192/apt.bp.107.003848 ◽

2010 ◽

Vol 16 (2) ◽

pp. 96-104 ◽

Cited By ~ 2

Author(s):

Raguraman Janakiraman ◽

Tony Benning

Keyword(s):

Attention Deficit Hyperactivity Disorder ◽

Attention Deficit ◽

Rating Scales ◽

Clinical Interview ◽

Diagnostic Evaluation ◽

Review Of The Literature ◽

Complete Understanding ◽

Distinct Entity ◽

Hyperactivity Disorder ◽

Clinical Presentations

SummaryAttention-deficit hyperactivity disorder (ADHD) is an established diagnosis in children but there is a lack of agreement about its validity as a distinct entity in adults. Literature suggests that between one-third and two-thirds of children diagnosed with ADHD continue to manifest symptoms into adulthood. An adult diagnosis should be done on the basis of a thorough assessment, structured and semi-structured clinical interview, and with a complete understanding of the symptoms that manifest in adults. This may be supplemented by the use of rating scales. We present a review of the literature covering aetiology, clinical presentations, diagnostic evaluation and management of ADHD in adults.

Download Full-text