Comparison of the performance of machine learning algorithms for the task-switching functional magnetic resonance imaging data for distinguishing attention deficit hyperactivity disorder from bipolar disorder

Amac: Dikkat eksikligi hiperaktivite bozuklugu (DEHB) cocukluk doneminde baslayan ve dikkatsizlik, asiri hareketlilik ve durtusellik gibi temel belirtilerle kendini gosteren bir bozukluktur. Bipola...

FIGURE 1. Changes in cortical thickness provide one measure of brain maturation. A large longitudinal study found that for most areas of cortex, children with attention deficit hyperactivity disorder (ADHD) reach peak cortical thickness several years later than typically developing children, supporting presence of developmental delay. The rate of cortical thinning also differed between the group who continued to meet diagnostic criteria into adulthood (persistent ADHD) and those who did not (remitted ADHD). Areas of cortex in which the rate of thinning correlated with adult symptom level (green, more symptoms associated with more thinning) are approximated on medial and lateral simplified representations of cortex. An earlier study also identified multiple areas in which cortex was thinner in adults with persistent ADHD compared with controls (orange). In addition, this study noted some areas of thicker cortex in remitted ADHD when compared with persistent ADHD (blue).

Neuroimaging-based diagnosis could help clinicians in making accurate diagnosis, accessing accurate prognosis, and deciding faster, more effective, and personalized treatment on an individual person basis. In this research work, we aim to develop a neuro-imaging, i.e. functional magnetic resonance imaging (fMRI), based method to detect attention deficit hyper-activity disorder (ADHD), which is a psychiatric disorder categorized by the impulsive nature, lack of attention, and hyper activeness. We utilized fMRI scans as well as personal characteristic features (PCF) data provided as part of ADHD-200 challenge. We aim to train a machine learning classifier by using fMRI and PCF data to classify each participant into one of the following three classes: healthy control (HC), combined-type ADHD (ADHD-C), or inattentive-type ADHD (ADHD-I). We used participants’ PCF and fMRI data separately, and then evaluated the combined use of both the datasets in detecting different classes. Support vector machine classifier with linear kernel was used for the training. The experiments were conducted under two different configurations: (i) 2-way configuration where classification was conducted between HC and ADHD (ADHD-C+ADHD-I) patients, and between ADHD-C and ADHD-I, and (ii) 3-way configuration where data of all the categories (HC, ADHD-C and ADHD-I) was combined together for classification. The 2-way classification approach achieved the diagnostic accuracy of 86.52% and 82.43% in distinguishing HC from ADHD patients, and ADHD-C and ADHD-I, respectively. The 3-way classification revealed classification success rate of 78.59% when both fMRI and PCF data were used together. These results demonstrate the importance of utilizing fMRI data and PCF for the detection of psychiatric disorders.

Purpose: Attention Deficit Hyperactivity Disorder (ADHD) is now recognized as the most common childhood behavioral disorder. This disorder causes school problems and social incompatibility. Thus an accurate diagnosis can help diminish such problems. In this paper, we propose a brain connectomics approach based on eyes-open resting state Magnetoencephalography (rs-MEG) to diagnose subjects with ADHD from Healthy Controls (HC). Materials and Methods: We used the eyes-open rs-MEG signals recorded from 25 subjects with ADHD and 25 HC. We calculated Coherence (COH) between the MEG sensors in the conventional frequency bands (i.e., delta, theta, alpha, beta, and gamma), selected the most discriminative COH measures by the Neighborhood Component Analysis (NCA), and fed them to three classifiers, including Support Vector Machine (SVM) with Radial Basis Function (RBF) kernel, K-Nearest Neighbors (KNN), and Decision Tree to classify ADHD and HC. Results: We achieved the best average accuracy of 91.1% for a single-band classifier based on the COH in the delta-band as an input feature of the SVM. However, when we integrated the COH values of all frequency bands as input features, the average accuracy was slightly improved to 92.7% using the SVM classifier. Conclusion: Our results demonstrate the capability of a functional connectomics approach based on rs-MEG for the diagnosis of ADHD. It is noteworthy that, to the best of our knowledge, COH has not yet been used to diagnose ADHD using rs-MEG data. Furthermore, there is no study on diagnosing ADHD using eyes-open rs-MEG. Thus, a novelty of our proposed method is to use COH and eyes-open rs-MEG data to diagnose ADHD. Moreover, our proposed method showed promising results compared with previous rs-MEG studies for the diagnosis of ADHD.

Despite a marked increase in treatment for bipolar disorder among youths, little is known about their pattern of service use. This article describes mental health service use in the year before and after a new clinical diagnosis of bipolar disorder. Claims were reviewed between April 1, 2004, and March 31, 2005, for 1,274,726 privately insured youths (17 years and younger) who were eligible for services at least one year before and after a service claim; 2,907 youths had new diagnosis of bipolar disorder during this period. Diagnoses of other mental disorders and prescriptions filled for psychotropic drugs were assessed in the year before and after the initial diagnosis of bipolar disorder. The one-year rate of a new diagnosis of bipolar disorder was .23%. During the year before the new diagnosis of bipolar disorder, youths were commonly diagnosed as having depressive disorder (46.5%) or disruptive behavior disorder (36.7%) and had often filled a prescription for an antidepressant (48.5%), stimulant (33.0%), mood stabilizer (31.8%), or antipsychotic (29.1%). Most youths with a new diagnosis of bipolar disorder had only one (28.8%) or two to four (28.7%) insurance claims for bipolar disorder in the year starting with the index diagnosis. The proportion starting mood stabilizers after the index diagnosis was highest for youths with five or more insurance claims for bipolar disorder (42.1%), intermediate for those with two to four claims (24.2%), and lowest for those with one claim (13.8%). Most youths with a new diagnosis of bipolar disorder had recently received treatment for depressive or disruptive behavior disorders, and many had no claims listing a diagnosis of bipolar disorder after the initial diagnosis. The service pattern suggests that a diagnosis of bipolar disorder is often given tentatively to youths treated for mental disorders with overlapping symptom profiles and is subsequently reconsidered.

Treatment Controversies in Adult ADHD.

Poster Abstracts from the AAAP 20th Annual Meeting and Symposium

The purpose of this study was to examine the rate of attention deficit hyperactivity disorder in adolescents with bipolar disorder and to explore the potential effects of comorbid attention deficit hyperactivity disorder on the phenomenology of adolescent bipolar disorder. The authors assessed the rate of attention deficit hyperactivity disorder for adolescents with bipolar disorder who were hospitalized for treatment of acute mania or hypomania. Eight (57%) of 14 adolescent bipolar patients also met DSM-III-R criteria for attention deficit hyperactivity disorder. Patients with both disorders were more likely to be male and Caucasian and to have mixed rather than manic bipolar disorder. Patients with attention deficit hyperactivity disorder had a higher mean total score on the Young Mania Rating Scale than patients with bipolar disorder alone. Although preliminary, these findings may have important implications regarding the potential relationship between bipolar disorder and attention deficit hyperactivity disorder.

An overactive mind: Investigating racing thoughts in ADHD, hypomania and comorbid ADHD and bipolar disorder via verbal fluency tasks

Who Are the Children With Severe Mood Dysregulation, a.k.a. “Rages”?

This study explored various feature extraction methods for use in automated diagnosis of Attention-Deficit Hyperactivity Disorder (ADHD) from functional Magnetic Resonance Image (fMRI) data. Each participant's data consisted of a resting state fMRI scan as well as phenotypic data (age, gender, handedness, IQ, and site of scanning) from the ADHD-200 dataset. We used machine learning techniques to produce support vector machine (SVM) classifiers that attempted to differentiate between (1) all ADHD patients vs. healthy controls and (2) ADHD combined (ADHD-c) type vs. ADHD inattentive (ADHD-i) type vs. controls. In different tests, we used only the phenotypic data, only the imaging data, or else both the phenotypic and imaging data. For feature extraction on fMRI data, we tested the Fast Fourier Transform (FFT), different variants of Principal Component Analysis (PCA), and combinations of FFT and PCA. PCA variants included PCA over time (PCA-t), PCA over space and time (PCA-st), and kernelized PCA (kPCA-st). Baseline chance accuracy was 64.2% produced by guessing healthy control (the majority class) for all participants. Using only phenotypic data produced 72.9% accuracy on two class diagnosis and 66.8% on three class diagnosis. Diagnosis using only imaging data did not perform as well as phenotypic-only approaches. Using both phenotypic and imaging data with combined FFT and kPCA-st feature extraction yielded accuracies of 76.0% on two class diagnosis and 68.6% on three class diagnosis—better than phenotypic-only approaches. Our results demonstrate the potential of using FFT and kPCA-st with resting-state fMRI data as well as phenotypic data for automated diagnosis of ADHD. These results are encouraging given known challenges of learning ADHD diagnostic classifiers using the ADHD-200 dataset (see Brown et al., 2012).

Pediatric bipolar disorder (BPD) and attention-deficit hyperactivity disorder (ADHD) co-occur more frequently than expected by chance. In this review, we examine 4 potential explanations for the high rate of this common co-occurrence: (i) BPD symptom expression leads to overdiagnosis of ADHD in BPD youth; (ii) ADHD is a prodromal or early manifestation of pediatric-onset BPD; (iii) ADHD and associated factors (e.g., psychostimulants) lead to the onset of pediatric BPD; and (iv) ADHD and BPD share an underlying biological etiology (i.e., a common familial or genetic risk or underlying neurophysiology). Peer-reviewed publications of studies of children and adolescents with comorbid BPD and ADHD were reviewed. There is a bidirectional overlap between BPD and ADHD in youth, with high rates of ADHD present in children with BPD (up to 85%), and elevated rates of BPD in children with ADHD (up to 22%). Phenomenologic, genetic, family, neuroimaging, and treatment studies revealed that BPD and ADHD have both common and distinct characteristics. While there are data to support all 4 explanations postulated in this paper, the literature most strongly suggests that ADHD symptoms represent a prodromal or early manifestation of pediatric-onset BPD in certain at-risk individuals. Bipolar disorder with comorbid ADHD may thus represent a developmentally specific phenotype of early-onset BPD. The etiology of comorbid BPD and ADHD is likely multifactorial. Additional longitudinal and biological studies are warranted to clarify the relationships between BPD and ADHD since they may have important diagnostic and treatment implications.

Executive deficits are reported in both early onset bipolar disorder (BD) and attention-deficit hyperactivity disorder (ADHD), and controversies regarding comorbidity and symptom overlap have complicated the research on executive function in BD. Reports of the negative impact of executive difficulties on academic functioning indicate a need for a greater focus on executive difficulties in early onset psychiatric disorders. Executive function and processing speed in youths with BD (n = 4), ADHD (n = 26) and BD + ADHD (n = 13) were compared with controls (n = 69). All clinical groups demonstrated executive impairment. The combined group was most impaired. There were no significant differences between the groups. Executive deficit in the BD group was associated with a history of psychotic symptoms. The BD-nonpsychotic group was impaired only with regard to processing speed. Processing speed adjustment improved working memory and normalized interference control in both BD and ADHD. executive deficits in BD may be determined by a history of psychotic symptoms rather than by comorbid ADHD. Some aspects of executive problems appear speed-related.

To determine the prevalence of bipolar disorder (BD) and sub-threshold symptoms in children with attention deficit hyperactivity disorder (ADHD) through 14 years' follow-up, when participants were between 21-24 years old. First, we examined rates of BD type I and II diagnoses in youth participating in the NIMH-funded Multimodal Treatment Study of ADHD (MTA). We used the diagnostic interview schedule for children (DISC), administered to both parents (DISC-P) and youth (DISCY). We compared the MTA study subjects with ADHD (n = 579) to a local normative comparison group (LNCG, n = 289) at 4 different assessment points: 6, 8, 12, and 14 years of follow-ups. To evaluate the bipolar variants, we compared total symptom counts (TSC) of DSM manic and hypomanic symptoms that were generated by DISC in ADHD and LNCG subjects. Then we sub-divided the TSC into pathognomonic manic (PM) and non-specific manic (NSM) symptoms. We compared the PM and NSM in ADHD and LNCG at each assessment point and over time. We also evaluated the irritability as category A2 manic symptom in both groups and over time. Finally, we studied the irritability symptom in correlation with PM and NSM in ADHD and LNCG subjects. DISC-generated BD diagnosis did not differ significantly in rates between ADHD (1.89%) and LNCG 1.38%). Interestingly, no participant met BD diagnosis more than once in the 4 assessment points in 14 years. However, on the symptom level, ADHD subjects reported significantly higher mean TSC scores: ADHD 3.0; LNCG 1.7; P < 0.001. ADHD status was associated with higher mean NSM: ADHD 2.0 vs LNCG 1.1; P < 0.0001. Also, ADHD subjects had higher PM symptoms than LNCG, with PM means over all time points of 1.3 ADHD; 0.9 LNCG; P = 0.0001. Examining both NSM and PM, ADHD status associated with greater NSM than PM. However, Over 14 years, the NSM symptoms declined and changed to PM over time (df 3, 2523; F = 20.1; P < 0.0001). Finally, Irritability (BD DSM criterion-A2) rates were significantly higher in ADHD than LNCG (χ(2) = 122.2, P < 0.0001), but irritability was associated more strongly with NSM than PM (df 3, 2538; F = 43.2; P < 0.0001). Individuals with ADHD do not appear to be at significantly greater risk for developing BD, but do show higher rates of BD symptoms, especially NSM. The greater linkage of irritability to NSM than to PM suggests caution when making BD diagnoses based on irritability alone as one of 2 (A-level) symptoms for BD diagnosis, particularly in view of its frequent presentation with other psychopathologies.

Comorbidity between attention deficit hyperactivity disorder (ADHD) and bipolar disorder in a specialized mood disorders outpatient clinic