Improved individualized prediction of schizophrenia in subjects at familial high risk, based on neuroanatomical data, schizotypal and neurocognitive features

Neurocognitive performance and cognitive biases in young adults with schizotypal traits

Recent research suggests that neurocognitive deficits in patients with schizophrenia may increase the risk of developing cognitive biases. As such, we set out to determine this predictive relationship as it pertains to the development of a first-episode psychosis. We hypothesized that poorer performance in processing speed would be associated with jumping to conclusions and an externalizing bias. Poorer performance in working memory would be associated with belief inflexibility and jumping to conclusions, and poorer performance in attention would be associated with attention to threat. We hypothesized that all cognitive biases would be associated with subsyndromal positive symptoms, and schizotypal traits would moderate these relationships. Undergraduate students (N = 130) completed the Schizotypal Personality Questionnaire, DAVOS Assessment of Cognitive Biases, Community Assessment of Psychic Experiences, and a computerized neuropsychological assessment battery. Processing speed had a small effect on externalizing bias, which in turn affected subsyndromal positive symptoms. There was no moderating effect of schizotypal traits on externalizing bias, but it was significantly associated with subsyndromal positive symptoms. Only the externalizing bias was associated with subsyndromal positive symptomatology, which might be explained by a restricted range and reduced variance in performance as a result of using a university student sample. This is one of few studies that sought to explain the mechanism responsible for the development of subsyndromal positive symptoms in a healthy sample using self-report measures.

Combining Clinical With Cognitive or Magnetic Resonance Imaging Data for Predicting Transition to Psychosis in Ultra High-Risk Patients: Data From the PACE 400 Cohort

BACKGROUND

Multimodal modeling that combines biological and clinical data shows promise in predicting transition to psychosis in individuals who are at ultra-high risk. Individuals who transition to psychosis are known to have deficits at baseline in cognitive function and reductions in gray matter volume in multiple brain regions identified by magnetic resonance imaging.

METHODS

In this study, we used Cox proportional hazards regression models to assess the additive predictive value of each modality-cognition, cortical structure information, and the neuroanatomical measure of brain age gap-to a previously developed clinical model using functioning and duration of symptoms prior to service entry as predictors in the Personal Assessment and Crisis Evaluation (PACE) 400 cohort. The PACE 400 study is a well-characterized cohort of Australian youths who were identified as ultra-high risk of transitioning to psychosis using the Comprehensive Assessment of At Risk Mental States (CAARMS) and followed for up to 18 years; it contains clinical data (from N = 416 participants), cognitive data (n = 213), and magnetic resonance imaging cortical parameters extracted using FreeSurfer (n = 231).

RESULTS

The results showed that neuroimaging, brain age gap, and cognition added marginal predictive information to the previously developed clinical model (fraction of new information: neuroimaging 0%-12%, brain age gap 7%, cognition 0%-16%).

CONCLUSIONS

In summary, adding a second modality to a clinical risk model predicting the onset of a psychotic disorder in the PACE 400 cohort showed little improvement in the fit of the model for long-term prediction of transition to psychosis.

Artificial intelligence in psychiatry research, diagnosis, and therapy

Psychiatric disorders are now responsible for the largest proportion of the global burden of disease, and even more challenges have been seen during the COVID-19 pandemic. Artificial intelligence (AI) is commonly used to facilitate the early detection of disease, understand disease progression, and discover new treatments in the fields of both physical and mental health. The present review provides a broad overview of AI methodology and its applications in data acquisition and processing, feature extraction and characterization, psychiatric disorder classification, potential biomarker detection, real-time monitoring, and interventions in psychiatric disorders. We also comprehensively summarize AI applications with regard to the early warning, diagnosis, prognosis, and treatment of specific psychiatric disorders, including depression, schizophrenia, autism spectrum disorder, attention-deficit/hyperactivity disorder, addiction, sleep disorders, and Alzheimer's disease. The advantages and disadvantages of AI in psychiatry are clarified. We foresee a new wave of research opportunities to facilitate and improve AI technology and its long-term implications in psychiatry during and after the COVID-19 era.

Deep Clinical Phenotyping of Schizophrenia Spectrum Disorders Using Data-Driven Methods: Marching towards Precision Psychiatry

Heterogeneity is the main challenge in the traditional classification of mental disorders, including schizophrenia spectrum disorders (SSD). This can be partly attributed to the absence of objective diagnostic criteria and the multidimensional nature of symptoms and their associated factors. This article provides an overview of findings from the Genetic Risk and Outcome of Psychosis (GROUP) cohort study on the deep clinical phenotyping of schizophrenia spectrum disorders targeting positive and negative symptoms, cognitive impairments and psychosocial functioning. Three to four latent subtypes of positive and negative symptoms were identified in patients, siblings and controls, whereas four to six latent cognitive subtypes were identified. Five latent subtypes of psychosocial function-multidimensional social inclusion and premorbid adjustment-were also identified in patients. We discovered that the identified subtypes had mixed profiles and exhibited stable, deteriorating, relapsing and ameliorating longitudinal courses over time. Baseline positive and negative symptoms, premorbid adjustment, psychotic-like experiences, health-related quality of life and PRS_SCZ were found to be the strong predictors of the identified subtypes. Our findings are comprehensive, novel and of clinical interest for precisely identifying high-risk population groups, patients with good or poor disease prognosis and the selection of optimal intervention, ultimately fostering precision psychiatry by tackling diagnostic and treatment selection challenges pertaining to heterogeneity.

Application of machine learning methods in predicting schizophrenia and bipolar disorders: A systematic review

Background and Aim

Schizophrenia and bipolar disorder (BD) are critical and high-risk inherited mental disorders with debilitating symptoms. Worldwide, 3% of the population suffers from these disorders. The mortality rate of these patients is higher compared to other people. Current procedures cannot effectively diagnose these disorders because it takes an average of 10 years from the onset of the first symptoms to the definitive diagnosis of the disease. Machine learning (ML) techniques are used to meet this need. This study aimed to summarize information on the use of ML techniques for predicting schizophrenia and BD to help early and timely diagnosis of the disease.

Methods

A systematic literature search included articles published until January 19, 2020 in 3 databases. Two reviewers independently assessed original papers to determine eligibility for inclusion in this review. PRISMA guidelines were followed to conduct the study, and the Prediction Model Risk of Bias Assessment Tool (PROBAST) to assess included papers.

Results

In this review, 1243 papers were retrieved through database searches, of which 15 papers were included based on full-text assessment. ML techniques were used to predict schizophrenia and BDs. The main algorithms applied were support vector machine (SVM) (10 studies), random forests (RF) (5 studies), and gradient boosting (GB) (3 studies). Input and output characteristics were very diverse and have been kept to enable future research. RFs algorithms demonstrated significantly higher accuracy and sensitivity than SVM and GB. GB demonstrated significantly higher specificity than SVM and RF. We found no significant difference between RF and SVM in terms of specificity.

Conclusion

ML can precisely predict results and assist in making clinical decisions-concerning schizophrenia and BD. RF often performed better than other algorithms in supervised learning tasks. This study identified gaps in the literature and opportunities for future psychological ML research.

Identification of texture MRI brain abnormalities on first-episode psychosis and clinical high-risk subjects using explainable artificial intelligence

Structural MRI studies in first-episode psychosis and the clinical high-risk state have consistently shown volumetric abnormalities. Aim of the present study was to introduce radiomics texture features in identification of psychosis. Radiomics texture features describe the interrelationship between voxel intensities across multiple spatial scales capturing the hidden information of underlying disease dynamics in addition to volumetric changes. Structural MR images were acquired from 77 first-episode psychosis (FEP) patients, 58 clinical high-risk subjects with no later transition to psychosis (CHR_NT), 15 clinical high-risk subjects with later transition (CHR_T), and 44 healthy controls (HC). Radiomics texture features were extracted from non-segmented images, and two-classification schemas were performed for the identification of FEP vs. HC and FEP vs. CHR_NT. The group of CHR_T was used as external validation in both schemas. The classification of a subject's clinical status was predicted by importing separately (a) the difference of entropy feature map and (b) the contrast feature map, resulting in classification balanced accuracy above 72% in both analyses. The proposed framework enhances the classification decision for FEP, CHR_NT, and HC subjects, verifies diagnosis-relevant features and may potentially contribute to identification of structural biomarkers for psychosis, beyond and above volumetric brain changes.

Combined HTR1A/1B methylation and human functional connectome to recognize patients with MDD

OBJECTIVES

This study aimed to use a machine-learning method to identify HTR1A/1B methylation and resting-state functional connectivity (rsFC) related to the diagnosis of MDD, then try to build classification models for MDD diagnosis based on the identified features.

METHODS

Peripheral blood samples were collected from all recruited participants, and part of the participants underwent the resting-state fMRI scan. Features including HTR1A/1B methylation and rsFC were calculated. Then, the initial feature sets of epigenetics and neuroimaging were separately input into an all-relevant feature selection to generate significant discriminative power for MDD diagnosis. Random forest classifiers were constructed and evaluated based on identified features. In addition, the SHapley Additive exPlanations (SHAP) method was adapted to interpret the diagnostic model.

RESULTS

A combination of selected HTR1A/1B methylation and rsFC feature sets achieved better performance than using either one alone - a distinction between MDD and healthy control groups was achieved at 81.78% classification accuracy and 0.8948 AUC.

CONCLUSION

A high classification accuracy can be achieved by combining multidimensional information from epigenetics and cerebral radiomic features in MDD. Our approach can be helpful for accurate clinical diagnosis of MDD and further exploring the pathogenesis of MDD.

Hybrid brain model accurately predict human procrastination behavior

Procrastination behavior is quite ubiquitous, and should warrant cautions to us owing to its significant influences in poor mental health, low subjective well-beings and bad academic performance. However, how to identify this behavioral problem have not yet to be fully elucidated. 1132 participants were recruited as distribution of benchmark. 81 high trait procrastinators (HP) and matched low trait procrastinators (LP) were screened. To address this issue, we have built upon the hybrid brain model by using hierarchical machine learning techniques to classify HP and LP with multi-modalities neuroimaging data (i.e., grey matter volume, fractional anisotropy, static/dynamic amplitude of low frequency fluctuation and static/dynamic degree centrality). Further, we capitalized on the multiple Canonical Correlation Analysis (mCCA) and joint Independent Component Analysis algorithm (mCCA + jICA) to clarify its fusion neural components as well. The hybrid brain model showed high accuracy to discriminate HP and LP (accuracy rate = 87.04%, sensitivity rate = 86.42%, specificity rate = 85.19%). Moreover, results of mCCA + jICA model revealed several joint-discriminative neural independent components (ICs) of this classification, showing wider co-variants of frontoparietal cortex and hippocampus networks. In addition, this study demonstrated three modal-specific discriminative ICs for classification, highlighting the temporal variants of brain local and global natures in ventromedial prefrontal cortex (vmPFC) and PHC in HP. To sum-up, this research developed a hybrid brain model to identify trait procrastination with high accuracy, and further revealed the neural hallmarks of this trait by integrating neuroimaging fusion data.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11571-021-09765-z.

Classification of Parkinson's disease using a region-of-interest- and resting-state functional magnetic resonance imaging-based radiomics approach

To investigate the value of combining amplitude of low-frequency fluctuations-based radiomics and the support vector machine classifier method in distinguishing patients with Parkinson's disease from healthy controls. A total of 123 patients with Parkinson's disease and 90 healthy controls from three centers with functional and structural MRI images were included in this study. We extracted radiomics features using the Brainnetome 246 atlas from the mean amplitude of low-frequency fluctuations maps. Two-sample t-tests and recursive feature elimination combined with support vector machine method were applied for feature selection and dimensionality reduction. We used support vector machine classifier to construct model and identify the discriminative features. The automated anatomical labeling 90 atlas and fivefold cross-validation were used to evaluate the robustness and generalization of the classifier. We found our model obtained a high classification performance with an accuracy of 78.07%, and AUC, sensitivity, and specificity of 0.8597, 78.80%, and 76.08%, respectively. We detected 7 discriminative brain subregions. The fivefold cross-validation and automated anatomical labeling 90 atlas also got high classification accuracy, and we found Brainnetome 246 atlas achieved a higher classification performance than the automated anatomical labeling 90 atlas both with tenfold and fivefold cross-validation. Our findings may help the early diagnosis of Parkinson's disease and provide support for research on Parkinson's disease mechanisms and clinical evaluation.

Classification of Low and High Schizotypy Levels via Evaluation of Brain Connectivity

Schizotypy is a latent cluster of personality traits that denote a vulnerability for schizophrenia or a type of spectrum disorder. The aim of the study is to investigate parametric effective brain connectivity features for classifying high versus low schizotypy (LS) status. Electroencephalography (EEG) signals are recorded from 13 high schizotypy (HS) and 11 LS participants during an emotional auditory odd-ball task. The brain connectivity signals for machine learning are taken after the settlement of event-related potentials. A multivariate autoregressive (MVAR)-based connectivity measure is estimated from the EEG signals using the directed transfer functions (DTFs) method. The values of DTF power in five standard frequency bands are used as features. The support vector machines (SVMs) revealed significant differences between HS and LS. The accuracy, specificity, and sensitivity of the results using SVM are as high as 89.21%, 90.3%, and 88.2%, respectively. Our results demonstrate that the effective brain connectivity in prefrontal/parietal and prefrontal/frontal brain regions considerably changes according to schizotypal status. These findings prove that the brain connectivity indices offer valuable biomarkers for detecting schizotypal personality. Further monitoring of the changes in DTF following the diagnosis of schizotypy may lead to the early identification of schizophrenia and other spectrum disorders.

The Electroretinogram as a Potential Biomarker of Psychosis in Children at Familial Risk

We previously proposed the electroretinogram (ERG) as a promising biomarker of major psychiatric disorders such as schizophrenia (SZ) and bipolar disorder (BP), given that we found anomalies in the ERG parameters of patients with these diagnoses as well as in their children who are at high risk (HR) of developing such disorders. The aim of the present study is to investigate the usefulness of the ERG for individual detection, among HR children, of an ERG profile resembling that of a SZ patient, as this may indicate a stronger likelihood of transition to psychosis. Using a logistic regression model previously derived from the ERG assessments of SZ patients and control (CT) subjects, individual risk scores were obtained for 61 HR and 80 CT youth. Those with a very high individual risk score were classified as "schizophrenia-like" (SZ-like). We found that the HR subjects were 3.5 times more likely to be classified as SZ-like than the CT subjects (95% CI [1.1-11.8]). Furthermore, among the HR subjects, we studied the relationship between the SZ-like classification and psychotic-like experiences and found that HR subjects classified as SZ-like were 2.7 times more likely than all remaining HR subjects to have experienced psychotic-like symptoms (95% CI [1.3-4.6]), and 6.8 times more likely than those with a very low individual risk score (95% CI [1.4-40.4]). Our results suggest that a model previously derived from ERG data on SZ patients could be a potential tool for early detection of the susceptibility to a psychotic-like disorder among familial HR children.

Prediction of transition to psychosis from an at-risk mental state using structural neuroimaging, genetic, and environmental data

INTRODUCTION

Psychosis is usually preceded by a prodromal phase in which patients are clinically identified as being at in an "At Risk Mental State" (ARMS). A few studies have demonstrated the feasibility of predicting psychosis transition from an ARMS using structural magnetic resonance imaging (sMRI) data and machine learning (ML) methods. However, the reliability of these findings is unclear due to possible sampling bias. Moreover, the value of genetic and environmental data in predicting transition to psychosis from an ARMS is yet to be explored.

METHODS

In this study we aimed to predict transition to psychosis from an ARMS using a combination of ML, sMRI, genome-wide genotypes, and environmental risk factors as predictors, in a sample drawn from a pool of 246 ARMS subjects (60 of whom later transitioned to psychosis). First, the modality-specific values in predicting transition to psychosis were evaluated using several: (a) feature types; (b) feature manipulation strategies; (c) ML algorithms; (d) cross-validation strategies, as well as sample balancing and bootstrapping. Subsequently, the modalities whose at least 60% of the classification models showed an balanced accuracy (BAC) statistically better than chance level were included in a multimodal classification model.

RESULTS AND DISCUSSION

Results showed that none of the modalities alone, i.e., neuroimaging, genetic or environmental data, could predict psychosis from an ARMS statistically better than chance and, as such, no multimodal classification model was trained/tested. These results suggest that the value of structural MRI data and genome-wide genotypes in predicting psychosis from an ARMS, which has been fostered by previous evidence, should be reconsidered.

Structural brain volumes of individuals at clinical high risk for psychosis: a meta-analysis

Background

Structural magnetic resonance imaging studies in individuals at clinical high risk (CHR) for psychosis have yielded conflicting results.

Methods

The aims of this study were to compare intracranial and structural brain volumes and variability of CHR individuals with those of healthy control (HC) subjects and to investigate brain volume differences and variability in CHR subjects with and without transition to psychosis. The PubMed and Embase databases were searched for relevant studies published before June 1, 2020.

Results

A total of 34 studies were deemed eligible, which included baseline data of 2111 CHR and 1472 HC participants. In addition, data were included for 401 CHR subjects who subsequently transitioned to psychosis and 1023 nontransitioned CHR participants. Whole-brain and left, right, and bilateral hippocampal volume were significantly smaller in CHR subjects than in HC subjects. Cerebrospinal fluid and lateral ventricle volumes were significantly larger in CHR subjects than in HC subjects. Variability was not significantly different in CHR subjects compared with HC subjects. CHR individuals with and without subsequent transition to psychosis did not show significant differences in any of the volumetric assessments or in variability.

Conclusions

This meta-analysis demonstrates reduced whole-brain and hippocampal volumes and increased cerebrospinal fluid and lateral ventricle volumes in CHR individuals. However, no significant differences were observed in any of the volumetric assessments between CHR individuals with and without subsequent transition to psychosis. These findings suggest that although structural brain alterations are present before the onset of the disorder, they may not significantly contribute to the identification of CHR individuals at the highest risk for the development of psychosis.

Predicting subclinical psychotic-like experiences on a continuum using machine learning

Previous studies applying machine learning methods to psychosis have primarily been concerned with the binary classification of chronic schizophrenia patients and healthy controls. The aim of this study was to use electroencephalographic (EEG) data and pattern recognition to predict subclinical psychotic-like experiences on a continuum between these two extremes in otherwise healthy people. We applied two different approaches to an auditory oddball regularity learning task obtained from N = 73 participants: A feature extraction and selection routine incorporating behavioural measures, event-related potential components and effective connectivity parameters; Regularisation of spatiotemporal maps of event-related potentials. Using the latter approach, optimal performance was achieved using the response to frequent, predictable sounds. Features within the P50 and P200 time windows had the greatest contribution toward lower Prodromal Questionnaire (PQ) scores and the N100 time window contributed most to higher PQ scores. As a proof-of-concept, these findings demonstrate that EEG data alone are predictive of individual psychotic-like experiences in healthy people. Our findings are in keeping with the mounting evidence for altered sensory responses in schizophrenia, as well as the notion that psychosis may exist on a continuum expanding into the non-clinical population.

An integrated machine learning framework for a discriminative analysis of schizophrenia using multi-biological data

Finding effective and objective biomarkers to inform the diagnosis of schizophrenia is of great importance yet remains challenging. Relatively little work has been conducted on multi-biological data for the diagnosis of schizophrenia. In this cross-sectional study, we extracted multiple features from three types of biological data, including gut microbiota data, blood data, and electroencephalogram data. Then, an integrated framework of machine learning consisting of five classifiers, three feature selection algorithms, and four cross validation methods was used to discriminate patients with schizophrenia from healthy controls. Our results show that the support vector machine classifier without feature selection using the input features of multi-biological data achieved the best performance, with an accuracy of 91.7% and an AUC of 96.5% (p < 0.05). These results indicate that multi-biological data showed better discriminative capacity for patients with schizophrenia than single biological data. The top 5% discriminative features selected from the optimal model include the gut microbiota features (Lactobacillus, Haemophilus, and Prevotella), the blood features (superoxide dismutase level, monocyte-lymphocyte ratio, and neutrophil count), and the electroencephalogram features (nodal local efficiency, nodal efficiency, and nodal shortest path length in the temporal and frontal-parietal brain areas). The proposed integrated framework may be helpful for understanding the pathophysiology of schizophrenia and developing biomarkers for schizophrenia using multi-biological data.

Longitudinal outcome of attenuated positive symptoms, negative symptoms, functioning and remission in people at clinical high risk for psychosis: a meta-analysis

BACKGROUND

Little is known about clinical outcomes other than transition to psychosis in people at Clinical High-Risk for psychosis (CHR-P). Our aim was to comprehensively meta-analytically evaluate for the first time a wide range of clinical and functional outcomes beyond transition to psychosis in CHR-P individuals.

METHODS

PubMed and Web of Science were searched until November 2020 in this PRISMA compliant meta-analysis (PROSPERO:CRD42020206271). Individual longitudinal studies conducted in individuals at CHR-P providing data on at least one of our outcomes of interest were included. We carried out random-effects pairwise meta-analyses, meta-regressions, and assessed publication bias and study quality. Analyses were two-tailed with α=0.05.

FINDINGS

75 prospective studies were included (n=5,288, age=20.0 years, females=44.5%). Attenuated positive symptoms improved at 12 (Hedges' g=0.753, 95%CI=0.495-1.012) and 24 (Hedges' g=0.836, 95%CI=0.463-1.209), but not ≥36 months (Hedges' g=0.315. 95%CI=-0.176-0.806). Negative symptoms improved at 12 (Hedges' g=0.496, 95%CI=0.315-0.678), but not 24 (Hedges' g=0.499, 95%CI=-0.137-1.134) or ≥36 months (Hedges' g=0.033, 95%CI=-0.439-0.505). Depressive symptoms improved at 12 (Hedges' g=0.611, 95%CI=0.441-0.782) and 24 (Hedges' g=0.583, 95%CI=0.364-0.803), but not ≥36 months (Hedges' g=0.512 95%CI=-0.337-1.361). Functioning improved at 12 (Hedges' g=0.711, 95%CI=0.488-0.934), 24 (Hedges' g=0.930, 95%CI=0.553-1.306) and ≥36 months (Hedges' g=0.392, 95%CI=0.117-0.667). Remission from CHR-P status occurred in 33.4% (95%CI=22.6-44.1%) at 12 months, 41.4% (95%CI=32.3-50.5%) at 24 months and 42.4% (95%CI=23.4-61.3%) at ≥36 months. Heterogeneity across the included studies was significant and ranged from I2=53.6% to I2=96.9%. The quality of the included studies (mean±SD) was 4.6±1.1 (range=2-8).

INTERPRETATION

CHR-P individuals improve on symptomatic and functional outcomes over time, but these improvements are not maintained in the longer term, and less than half fully remit. Prolonged duration of care may be needed for this patient population to optimize outcomes.

FUNDING

None.

Changes in Brain Structure, Function, and Network Properties in Patients With First-Episode Schizophrenia Treated With Antipsychotics

Purpose: Comprehensive and longitudinal brain analysis is of great significance for understanding the pathological changes of antipsychotic drug treatment in patients with schizophrenia. This study aimed to investigate the changes of structure, function, and network properties in patients with first-episode schizophrenia (FES) after antipsychotic therapy and their relationship with clinical symptoms. Materials and Methods: A total of 30 patients diagnosed with FES and 30 healthy subjects matched for sex and age were enrolled in our study. Patients at baseline were labeled as antipsychotic-naive first-episode schizophrenia (AN-FES), and patients after antipsychotic treatment were labeled as antipsychotic treatment first-episode schizophrenia (AT-FES). The severity of illness was measured by using the PANSS and CGI score. Structural and functional MRI data were also performed. Differences in GMV, ALFF, and ReHo between the FES group and healthy control group were tested using a voxel-wise two-sample t-test, and the comparison of AN-FES group and AT-FES group was evaluated by paired-sample t-test. Results: After the 1-year follow-up, the FES patients showed increased GMV in the right cerebellum, right inferior temporal gyrus, left middle frontal gyrus, parahippocampal gyrus, bilateral inferior parietal lobule, and reduced GMV in the left occipital lobe, gyrus rectus, right orbital frontal cortex. The patients also showed increased ALFF in the medial superior frontal gyrus and right precentral gyrus. For network properties, the patients showed reduced characteristic path length and increased global efficiency. The GMV of the right inferior parietal lobule was negatively correlated with the clinical symptoms. Conclusions: Our study showed that the antipsychotic treatment contributed to the structural alteration and functional improvement, and the GMV alteration may be associated with the improvement of clinical symptoms.

Cost-effectiveness of neuroimaging technologies in management of psychiatric and insomnia disorders: A meta-analysis and prospective cost analysis

BACKGROUND

The optimal diagnostic strategy for patients with psychiatric and insomnia disorders has not been established yet.

PURPOSE

The purpose of this study was to perform cost-effectiveness analysis of six neuroimaging technologies in diagnosis of patients with psychiatric and insomnia disorders.

METHODS

An economic evaluation study was conducted in three parts, including a systematic review for determining diagnostic accuracy, a descriptive cross-sectional study with Activity-Based Costing (ABC) technique for tracing resource consumption, and a cost-effectiveness analysis using a short-term decision-analytic model.

RESULTS

In the first phase, 93 diagnostic accuracy studies were included in the systematic review. The accumulated results (meta-analysis) showed that the highest diagnostic accuracy for psychiatric and insomnia disorders was attributed to PET (sensitivity of 90% and specificity of 80%) and MRI (sensitivity of 76% and specificity of 78%) respectively. In the second phase of the study, we calculated the cost of each technology. The results showed that MRI has the lowest cost. Based on the results in the model of cost-effectiveness sMRI ($ 50.08 per accurate diagnosis) and MRI ($ 58.54 per accurate diagnosis) were more cost-effective neuroimaging technologies.

CONCLUSION

In psychiatric disorders, no single strategy was characterized by both low cost and high accuracy. However, MRI and PET scan had lower cost and higher accuracy for psychiatric disorders, respectively. MRI was the least costly with the highest diagnostic accuracy in insomnia disorders. Based on our model, sMRI in psychiatric disorders and MRI in insomnia disorders were the most cost-effective technologies.

Support vector machine-based classification of schizophrenia patients and healthy controls using structural magnetic resonance imaging from two independent sites

Structural brain alterations have been repeatedly reported in schizophrenia; however, the pathophysiology of its alterations remains unclear. Multivariate pattern recognition analysis such as support vector machines can classify patients and healthy controls by detecting subtle and spatially distributed patterns of structural alterations. We aimed to use a support vector machine to distinguish patients with schizophrenia from control participants on the basis of structural magnetic resonance imaging data and delineate the patterns of structural alterations that significantly contributed to the classification performance. We used independent datasets from different sites with different magnetic resonance imaging scanners, protocols and clinical characteristics of the patient group to achieve a more accurate estimate of the classification performance of support vector machines. We developed a support vector machine classifier using the dataset from one site (101 participants) and evaluated the performance of the trained support vector machine using a dataset from the other site (97 participants) and vice versa. We assessed the performance of the trained support vector machines in each support vector machine classifier. Both support vector machine classifiers attained a classification accuracy of >70% with two independent datasets indicating a consistently high performance of support vector machines even when used to classify data from different sites, scanners and different acquisition protocols. The regions contributing to the classification accuracy included the bilateral medial frontal cortex, superior temporal cortex, insula, occipital cortex, cerebellum, and thalamus, which have been reported to be related to the pathogenesis of schizophrenia. These results indicated that the support vector machine could detect subtle structural brain alterations and might aid our understanding of the pathophysiology of these changes in schizophrenia, which could be one of the diagnostic findings of schizophrenia.

Progression from being at-risk to psychosis: next steps

Over the past 20 years there has been a great deal of research into those considered to be at risk for developing psychosis. Much has been learned and studies have been encouraging. The aim of this paper is to offer an update of the current status of research on risk for psychosis, and what the next steps might be in examining the progression from CHR to psychosis. Advances have been made in accurate prediction, yet there are some methodological issues in ascertainment, diagnosis, the use of data-driven selection methods and lack of external validation. Although there have been several high-quality treatment trials the heterogeneity of this clinical high-risk population has to be addressed so that their treatment needs can be properly met. Recommendations for the future include more collaborative research programmes, and ensuring they are accessible and harmonized with respect to criteria and outcomes so that the field can continue to move forward with the development of large collaborative consortiums as well as increased funding for multisite projects.

Role for Positive Schizotypy and Hallucination Proneness in Semantic Processing

Semantic processing underpins the organization of verbal information for both storage and retrieval. Deficits in semantic processing are associated with both the risk for and symptoms presented in schizophrenia. However, studies are mixed and could reflect the confounding effects of medication and symptom heterogeneity. Therefore, we considered whether two risk phenotypes, positive schizotypy and hallucinatory predisposition, present in the general population were associated with differential responding profiles for a semantic processing task. One hundred and eighty-three participants completed the Schizotypal Personality Questionnaire, Launay-Slade Hallucination Scale, National Adult Reading Test, a handedness measure, and a computerized semantic relatedness judgment task. Pairs of words were related through their dominant or subordinate meanings, or unrelated. Participants were divided into four groups using a mean split on cognitive-perceptual (positive) schizotypy and hallucination proneness. Significant differences between groups were found for reaction time on the semantic relatedness task, with the high cognitive-perceptual schizotypy groups responding significantly slower to all word pairs compared to their low scoring counterparts. There was some evidence that high hallucination proneness was associated with significantly faster reaction times which may reflect disinhibitive processes, however additional support is required. The results suggest that these two components of psychosis risk are associated with different patterns of responding to semantic processing. More diffuse activation of semantic information appeared to be associated with positive schizotypy, while those predisposed to hallucinations appeared to respond quicker. These results have significant implications in the re-conceptualization of hallucination proneness as distinct from positive schizotypy. Additional research is required to investigate the association between psychotic-like experiences separate from personality variables such as positive schizotypy and semantic processing.

Implementing Precision Psychiatry: A Systematic Review of Individualized Prediction Models for Clinical Practice

BACKGROUND

The impact of precision psychiatry for clinical practice has not been systematically appraised. This study aims to provide a comprehensive review of validated prediction models to estimate the individual risk of being affected with a condition (diagnostic), developing outcomes (prognostic), or responding to treatments (predictive) in mental disorders.

METHODS

PRISMA/RIGHT/CHARMS-compliant systematic review of the Web of Science, Cochrane Central Register of Reviews, and Ovid/PsycINFO databases from inception until July 21, 2019 (PROSPERO CRD42019155713) to identify diagnostic/prognostic/predictive prediction studies that reported individualized estimates in psychiatry and that were internally or externally validated or implemented. Random effect meta-regression analyses addressed the impact of several factors on the accuracy of prediction models.

FINDINGS

Literature search identified 584 prediction modeling studies, of which 89 were included. 10.4% of the total studies included prediction models internally validated (n = 61), 4.6% models externally validated (n = 27), and 0.2% (n = 1) models considered for implementation. Across validated prediction modeling studies (n = 88), 18.2% were diagnostic, 68.2% prognostic, and 13.6% predictive. The most frequently investigated condition was psychosis (36.4%), and the most frequently employed predictors clinical (69.5%). Unimodal compared to multimodal models (β = .29, P = .03) and diagnostic compared to prognostic (β = .84, p < .0001) and predictive (β = .87, P = .002) models were associated with increased accuracy.

INTERPRETATION

To date, several validated prediction models are available to support the diagnosis and prognosis of psychiatric conditions, in particular, psychosis, or to predict treatment response. Advancements of knowledge are limited by the lack of implementation research in real-world clinical practice. A new generation of implementation research is required to address this translational gap.

Individualized Diagnostic and Prognostic Models for Patients With Psychosis Risk Syndromes: A Meta-analytic View on the State of the Art

BACKGROUND

The clinical high risk (CHR) paradigm has facilitated research into the underpinnings of help-seeking individuals at risk for developing psychosis, aiming at predicting and possibly preventing transition to the overt disorder. Statistical methods such as machine learning and Cox regression have provided the methodological basis for this research by enabling the construction of diagnostic models (i.e., distinguishing CHR individuals from healthy individuals) and prognostic models (i.e., predicting a future outcome) based on different data modalities, including clinical, neurocognitive, and neurobiological data. However, their translation to clinical practice is still hindered by the high heterogeneity of both CHR populations and methodologies applied.

METHODS

We systematically reviewed the literature on diagnostic and prognostic models built on Cox regression and machine learning. Furthermore, we conducted a meta-analysis on prediction performances investigating heterogeneity of methodological approaches and data modality.

RESULTS

A total of 44 articles were included, covering 3707 individuals for prognostic studies and 1052 individuals for diagnostic studies (572 CHR patients and 480 healthy control subjects). CHR patients could be classified against healthy control subjects with 78% sensitivity and 77% specificity. Across prognostic models, sensitivity reached 67% and specificity reached 78%. Machine learning models outperformed those applying Cox regression by 10% sensitivity. There was a publication bias for prognostic studies yet no other moderator effects.

CONCLUSIONS

Our results may be driven by substantial clinical and methodological heterogeneity currently affecting several aspects of the CHR field and limiting the clinical implementability of the proposed models. We discuss conceptual and methodological harmonization strategies to facilitate more reliable and generalizable models for future clinical practice.

Multivariate patterns of gray matter volume in thalamic nuclei are associated with positive schizotypy in healthy individuals

BACKGROUND

Previous models suggest biological and behavioral continua among healthy individuals (HC), at-risk condition, and full-blown schizophrenia (SCZ). Part of these continua may be captured by schizotypy, which shares subclinical traits and biological phenotypes with SCZ, including thalamic structural abnormalities. In this regard, previous findings have suggested that multivariate volumetric patterns of individual thalamic nuclei discriminate HC from SCZ. These results were obtained using machine learning, which allows case-control classification at the single-subject level. However, machine learning accuracy is usually unsatisfactory possibly due to phenotype heterogeneity. Indeed, a source of misclassification may be related to thalamic structural characteristics of those HC with high schizotypy, which may resemble structural abnormalities of SCZ. We hypothesized that thalamic structural heterogeneity is related to schizotypy, such that high schizotypal burden would implicate misclassification of those HC whose thalamic patterns resemble SCZ abnormalities.

METHODS

Following a previous report, we used Random Forests to predict diagnosis in a case-control sample (SCZ = 131, HC = 255) based on thalamic nuclei gray matter volumes estimates. Then, we investigated whether the likelihood to be classified as SCZ (π-SCZ) was associated with schizotypy in 174 HC, evaluated with the Schizotypal Personality Questionnaire.

RESULTS

Prediction accuracy was 72.5%. Misclassified HC had higher positive schizotypy scores, which were correlated with π-SCZ. Results were specific to thalamic rather than whole-brain structural features.

CONCLUSIONS

These findings strengthen the relevance of thalamic structural abnormalities to SCZ and suggest that multivariate thalamic patterns are correlates of the continuum between schizotypy in HC and the full-blown disease.

The benefit of foresight? An ethical evaluation of predictive testing for psychosis in clinical practice

Risk prediction for psychosis has advanced to the stage at which it could feasibly become a clinical reality. Neuroimaging biomarkers play a central role in many risk prediction models. Using such models to predict the likelihood of transition to psychosis in individuals known to be at high risk has the potential to meaningfully improve outcomes, principally through facilitating early intervention. However, this compelling benefit must be evaluated in light of the broader ethical ramifications of this prospective development in clinical practice. This paper advances ethical discussion in the field in two ways: firstly, through in-depth consideration of the distinctive implications of the clinical application of predictive tools; and, secondly, by evaluating the manner in which newer predictive models incorporating neuroimaging alter the ethical landscape. We outline the current state of the science of predictive testing for psychosis, with a particular focus on emerging neuroimaging biomarkers. We then proceed to ethical analysis employing the four principles of biomedical ethics as a conceptual framework. We conclude with a call for scientific advancement to proceed in tandem with ethical consideration, informed by empirical study of the views of high risk individuals and their families. This collaborative approach will help ensure that predictive testing progresses in an ethically acceptable manner that minimizes potential adverse effects and maximizes meaningful benefits for those at high risk of psychosis.

Cognitive Functioning and Schizotypy: A Four-Years Study

Although there is ample evidence from cross-sectional studies indicating cognitive deficits in high schizotypal individuals that resemble the cognitive profile of schizophrenia-spectrum patients, there is still lack of evidence by longitudinal/follow-up studies. The present study included assessments of schizotypal traits and a wide range of cognitive functions at two time points (baseline and 4-years assessments) in order to examine (a) their stability over time, (b) the predictive value of baseline schizotypy on cognition at follow-up and (c) differences in cognition between the two time points in high negative schizotypal and control individuals. Only high negative schizotypal individuals were compared with controls due to the limited number of participants falling in the other schizotypal groups at follow-up. Seventy participants (mean age: 36.17; 70% females) were assessed at baseline and follow-up. Schizotypal traits were evaluated with the Schizotypal Personality Questionnaire. We found that schizotypal traits decreased over time, except in a sub-group of participants ("schizotypy congruent") that includes individuals who consistently meet normative criteria of inclusion in either a schizotypal or control group. In these individuals, negative schizotypy and aspects of cognitive-perceptual and disorganized schizotypy remained stable. The stability of cognitive functioning also varied over time: response inhibition, aspects of cued attention switching, set-shifting and phonemic/semantic verbal fluency improved at follow-up. High negative schizotypy at baseline predicted poorer response inhibition and semantic switching at follow-up while high disorganized schizotypy predicted poorer semantic processing and complex processing speed/set-shifting. The between-group analyses revealed that response inhibition, set-shifting and complex processing speed/set-shifting were poorer in negative schizotypals compared with controls at both time points, while maintaining set and semantic switching were poorer only at follow-up. Taken together, the findings show differential stability of the schizotypal traits over time and indicate that different aspects of schizotypy predict a different pattern of neuropsychological task performance during a 4-years time window. These results are of significant use in the formulation of targeted early-intervention strategies for high-risk populations.

Recursive Feature Elimination by Sensitivity Testing

There is great interest in methods to improve human insight into trained non-linear models. Leading approaches include producing a ranking of the most relevant features, a non-trivial task for non-linear models. We show theoretically and empirically the benefit of a novel version of recursive feature elimination (RFE) as often used with SVMs; the key idea is a simple twist on the kinds of sensitivity testing employed in computational learning theory with membership queries (e.g., [1]). With membership queries, one can check whether changing the value of a feature in an example changes the label. In the real-world, we usually cannot get answers to such queries, so our approach instead makes these queries to a trained (imperfect) non-linear model. Because SVMs are widely used in bioinformatics, our empirical results use a real-world cancer genomics problem; because ground truth is not known for this task, we discuss the potential insights provided. We also evaluate on synthetic data where ground truth is known.

Accuracy of diagnostic classification algorithms using cognitive-, electrophysiological-, and neuroanatomical data in antipsychotic-naïve schizophrenia patients

BACKGROUND

A wealth of clinical studies have identified objective biomarkers, which separate schizophrenia patients from healthy controls on a group level, but current diagnostic systems solely include clinical symptoms. In this study, we investigate if machine learning algorithms on multimodal data can serve as a framework for clinical translation.

METHODS

Forty-six antipsychotic-naïve, first-episode schizophrenia patients and 58 controls underwent neurocognitive tests, electrophysiology, and magnetic resonance imaging (MRI). Patients underwent clinical assessments before and after 6 weeks of antipsychotic monotherapy with amisulpride. Nine configurations of different supervised machine learning algorithms were applied to first estimate the unimodal diagnostic accuracy, and next to estimate the multimodal diagnostic accuracy. Finally, we explored the predictability of symptom remission.

RESULTS

Cognitive data significantly classified patients from controls (accuracies = 60-69%; p values = 0.0001-0.009). Accuracies of electrophysiology, structural MRI, and diffusion tensor imaging did not exceed chance level. Multimodal analyses with cognition plus any combination of one or more of the remaining three modalities did not outperform cognition alone. None of the modalities predicted symptom remission.

CONCLUSIONS

In this multivariate and multimodal study in antipsychotic-naïve patients, only cognition significantly discriminated patients from controls, and no modality appeared to predict short-term symptom remission. Overall, these findings add to the increasing call for cognition to be included in the definition of schizophrenia. To bring about the full potential of machine learning algorithms in first-episode, antipsychotic-naïve schizophrenia patients, careful a priori variable selection based on independent data as well as inclusion of other modalities may be required.

Using machine learning to explain the heterogeneity of schizophrenia. Realizing the promise and avoiding the hype

Despite extensive research and prodigious advances in neuroscience, our comprehension of the nature of schizophrenia remains rudimentary. Our failure to make progress is attributed to the extreme heterogeneity of this condition, enormous complexity of the human brain, limitations of extant research paradigms, and inadequacy of traditional statistical methods to integrate or interpret increasingly large amounts of multidimensional information relevant to unravelling brain function. Fortunately, the rapidly developing science of machine learning appears to provide tools capable of addressing each of these impediments. Enthusiasm about the potential of machine learning methods to break the current impasse is reflected in the steep increase in the number of scientific publication about the application of machine learning to the study of schizophrenia. Machine learning approaches are, however, poorly understood by schizophrenia researchers and clinicians alike. In this paper, we provide a simple description of the nature and techniques of machine learning and their application to the study of schizophrenia. We then summarize its potential and constraints with illustrations from six studies of machine learning in schizophrenia and address some common misconceptions about machine learning. We suggest some guidelines for researchers, readers, science editors and reviewers of the burgeoning machine learning literature in schizophrenia. In order to realize its enormous promise, we suggest the need for the disciplined application of machine learning methods to the study of schizophrenia with a clear recognition of its capability and challenges accompanied by a concurrent effort to improve machine learning literacy among neuroscientists and mental health professionals.

Predictors of Transition to Psychosis in Individuals at Clinical High Risk

PURPOSE OF REVIEW

Current research is examining predictors of the transition to psychosis in youth who are at clinical high risk based on attenuated psychotic symptoms (APS). Determining predictors of the development of psychosis is important for an improved understanding of mechanisms as well as the development of preventative strategies. The purpose is to review the most recent literature identifying predictors of the transition to psychosis in those who are already assessed as being at risk.

RECENT FINDINGS

Multidomain models, in particular, integrated models of symptoms, social functioning, and cognition variables, achieve better predictive performance than individual factors. There are many methodological issues; however, several solutions have now been described in the literature. For youth who already have APS, predicting who may go on to later develop psychosis is possible. Several studies are underway in large consortiums that may overcome some of the methodological concerns and develop improved means of prediction.

Predicting major mental illness: ethical and practical considerations

SummaryAn increasing body of genetic and imaging research shows that it is becoming possible to forecast the onset of major psychiatric disorders such as depression and schizophrenia before people become ill with ever improving accuracy. Practical issues such as the optimal combination of clinical and biological variables are being addressed, but the application of predictive algorithms to individuals or in routine clinical settings have yet to be tested. The development of predictive methods in mental health comes with substantial ethical questions, including whether people wish to know their level of risk, as well as individual and societal attitudes to the potential adverse effects of data sharing, early diagnosis and treatment, which so far have been largely ignored. Preliminary data suggests that at least some people think predictive research is valuable and would take part in such studies, and some would welcome knowing the results. Future initiatives should systematically assess opinions and attitudes in conjunction with scientific and technical advances.Declaration of interestIn the past 3 years, S.M.L. has received personal fees from Otsuaka, Sunovion and Janssen, and research grant support from Janssen and Lundbeck. A.M.M. has received research support from the Sackler Trust, Eli Lilly and Janssen. S.M.L. is part of the PSYSCAN consortium.

What accounts for poor functioning in people with schizophrenia: a re-evaluation of the contributions of neurocognitive v. attitudinal and motivational factors

BACKGROUND

Neurocognitive deficits are often seen as core features of schizophrenia, and as primary determinants of poor functioning. Yet, our clinical observations suggest that individuals who score within the impaired range on standardized tests can reliably perform better in complex real-world situations, especially when performance is embedded within a positive socio-affective context.

METHODS

We analyzed literature on the influence of non-neurocognitive factors on test performance in order to clarify their contributions.

RESULTS

We identified seven non-neurocognitive factors that significantly contribute to neurocognitive test performance: avolition, dysfunctional attitudes, effort, stress, negative emotions, asociality, and disorganized symptoms. We then proposed an alternative model based on dysfunctional (e.g. defeatist) attitudes and their consequences for motivation and sustained task engagement. We demonstrated that these factors account for substantial variance in negative symptoms, neurocognitive test performance, and functional outcomes. We then demonstrated that recovery-oriented cognitive therapy - which is derived from this alternative model and primarily targets dysfunctional beliefs - has been successful in the treatment of low functioning individuals with schizophrenia.

CONCLUSION

The contributions of neurocognitive impairments to poor real-world functioning in people with schizophrenia may be overstated in the literature, and may even be limited relative to non-neurocognitive factors. We offer suggestions for further research to more precisely quantify the contributions of attitudinal/motivation v. neurocognitive factors in schizophrenia.

A reassessment of the higher-order factor structure of the German Schizotypal Personality Questionnaire (SPQ-G) in German-speaking adults

The Schizotypal Personality Questionnaire (SPQ) is a widely-used self-report instrument for the assessment of schizotypal personality traits. However, the factor structure of scores on English and non-English translations of the SPQ has been a matter of debate. With little previous factorial evaluation of the German version of the SPQ (SPQ-G), we re-assessed the higher-order factor structure of the measure. A total of 2,428 German-speaking adults from Central Europe (CE) and the United Kingdom (UK) completed the SPQ-G. Confirmatory factor analysis - testing proposed 2-, 3-, and 4-factor models of SPQ-G scores - indicated that the 4-factor solution had best fit. Partial measurement invariance across cultural group (CE and UK) and sex was obtained for the 4-factor model. Further analyses showed CE participants had significantly higher scores than UK participants on one schizotypal facet. These results suggest that scores on the SPQ-G are best explained in terms of a higher-order, 4-factor solution in German migrant and non-migrant adults.

The Network Structure of Schizotypal Personality Traits

Elucidating schizotypal traits is important if we are to understand the various manifestations of psychosis spectrum liability and to reliably identify individuals at high risk for psychosis. The present study examined the network structures of (1) 9 schizotypal personality domains and (2) 74 individual schizotypal items, and (3) explored whether networks differed across gender and culture (North America vs China). The study was conducted in a sample of 27001 participants from 12 countries and 21 sites (M age = 22.12; SD = 6.28; 37.5% males). The Schizotypal Personality Questionnaire (SPQ) was used to assess 74 self-report items aggregated in 9 domains. We used network models to estimate conditional dependence relations among variables. In the domain-level network, schizotypal traits were strongly interconnected. Predictability (explained variance of each node) ranged from 31% (odd/magical beliefs) to 55% (constricted affect), with a mean of 43.7%. In the item-level network, variables showed relations both within and across domains, although within-domain associations were generally stronger. The average predictability of SPQ items was 27.8%. The network structures of men and women were similar (r = .74), node centrality was similar across networks (r = .90), as was connectivity (195.59 and 199.70, respectively). North American and Chinese participants networks showed lower similarity in terms of structure (r = 0.44), node centrality (r = 0.56), and connectivity (180.35 and 153.97, respectively). In sum, the present article points to the value of conceptualizing schizotypal personality as a complex system of interacting cognitive, emotional, and affective characteristics.

Perspectives on Machine Learning for Classification of Schizotypy Using fMRI Data

Functional magnetic resonance imaging is capable of estimating functional activation and connectivity in the human brain, and lately there has been increased interest in the use of these functional modalities combined with machine learning for identification of psychiatric traits. While these methods bear great potential for early diagnosis and better understanding of disease processes, there are wide ranges of processing choices and pitfalls that may severely hamper interpretation and generalization performance unless carefully considered. In this perspective article, we aim to motivate the use of machine learning schizotypy research. To this end, we describe common data processing steps while commenting on best practices and procedures. First, we introduce the important role of schizotypy to motivate the importance of reliable classification, and summarize existing machine learning literature on schizotypy. Then, we describe procedures for extraction of features based on fMRI data, including statistical parametric mapping, parcellation, complex network analysis, and decomposition methods, as well as classification with a special focus on support vector classification and deep learning. We provide more detailed descriptions and software as supplementary material. Finally, we present current challenges in machine learning for classification of schizotypy and comment on future trends and perspectives.

Imaging and Genetic Biomarkers Predicting Transition to Psychosis

The search for diagnostic and prognostic biomarkers in schizophrenia care and treatment is the focus of many within the research community. Longitudinal cohorts of patients presenting at elevated genetic and clinical risk have provided a wealth of data that has informed our understanding of the development of schizophrenia and related psychotic disorders.Imaging follow-up of high-risk cohorts has demonstrated changes in cerebral grey matter of those that eventually transition to schizophrenia that predate the onset of symptoms and evolve over the course of illness. Longitudinal follow-up studies demonstrate that observed grey matter changes can be employed to differentiate those who will transition to schizophrenia from those who will not prior to the onset of the disorder.In recent years our understanding of the genetic makeup of schizophrenia has advanced significantly. The development of modern analysis techniques offers researchers the ability to objectively quantify genetic risk; these have been successfully applied within a high-risk paradigm to assist in differentiating between high-risk individuals who will subsequently become unwell and those who will not.This chapter will discuss the application of imaging and genetic biomarkers within high-risk groups to predict future transition to schizophrenia and related psychotic disorders. We aim to provide an overview of current approaches focussing on grey matter changes that are predictive of future transition to illness, the developing field of genetic risk scores and other methods being developed to aid clinicians in diagnosis and prognosis.

Three-dimensional texture features from intensity and high-order derivative maps for the discrimination between bladder tumors and wall tissues via MRI

PURPOSE

This study aims to determine the three-dimensional (3D) texture features extracted from intensity and high-order derivative maps that could reflect textural differences between bladder tumors and wall tissues, and propose a noninvasive, image-based strategy for bladder tumor differentiation preoperatively.

METHODS

A total of 62 cancerous and 62 wall volumes of interest (VOI) were extracted from T2-weighted MRI datasets of 62 patients with pathologically confirmed bladder cancer. To better reflect heterogeneous distribution of tumor tissues, 3D high-order derivative maps (the gradient and curvature maps) were calculated from each VOI. Then 3D Haralick features based on intensity and high-order derivative maps and Tamura features based on intensity maps were extracted from each VOI. Statistical analysis and recursive feature elimination-based support vector machine classifier (RFE-SVM) was proposed to first select the features with significant differences and then obtain a more predictive and compact feature subset to verify its differentiation performance.

RESULTS

From each VOI, a total of 58 texture features were derived. Among them, 37 features showed significant inter-class differences ([Formula: see text]). With 29 optimal features selected by RFE-SVM, the classification results namely the sensitivity, specificity, accuracy and area under the curve (AUC) of the receiver operating characteristics were 0.9032, 0.8548, 0.8790 and 0.9045, respectively. By using synthetic minority oversampling technique to augment the sample number of each group to 200, the sensitivity, specificity, accuracy an AUC value of the feature selection-based classification were improved to 0.8967, 0.8780, 0.8874 and 0.9416, respectively.

CONCLUSIONS

Our results suggest that 3D texture features derived from intensity and high-order derivative maps can better reflect heterogeneous distribution of cancerous tissues. Texture features optimally selected together with sample augmentation could improve the performance on differentiating bladder carcinomas from wall tissues, suggesting a potential way for tumor noninvasive staging of bladder cancer preoperatively.