Diagnostic and Prognostic Classification of Brain Disorders Using Residual Learning on Structural MRI Data

A Demographic-Conditioned Variational Autoencoder for fMRI Distribution Sampling and Removal of Confounds

Objective

fMRI and derived measures such as functional connectivity (FC) have been used to predict brain age, general fluid intelligence, psychiatric disease status, and preclinical neurodegenerative disease. However, it is not always clear that all demographic confounds, such as age, sex, and race, have been removed from fMRI data. Additionally, many fMRI datasets are restricted to authorized researchers, making dissemination of these valuable data sources challenging.

Methods

We create a variational autoencoder (VAE)-based model, DemoVAE, to decorrelate fMRI features from demographics and generate high-quality synthetic fMRI data based on user-supplied demographics. We train and validate our model using two large, widely used datasets, the Philadelphia Neurodevel-opmental Cohort (PNC) and Bipolar and Schizophrenia Network for Intermediate Phenotypes (BSNIP).

Results

We find that DemoVAE recapitulates group differences in fMRI data while capturing the full breadth of individual variations. Significantly, we also find that most clinical and computerized battery fields that are correlated with fMRI data are not correlated with DemoVAE latents. An exception are several fields related to schizophrenia medication and symptom severity.

Conclusion

Our model generates fMRI data that captures the full distribution of FC better than traditional VAE or GAN models. We also find that most prediction using fMRI data is dependent on correlation with, and prediction of, demographics.

Significance

Our DemoVAE model allows for generation of high quality synthetic data conditioned on subject demographics as well as the removal of the confounding effects of demographics. We identify that FC-based prediction tasks are highly influenced by demographic confounds.

Evaluation of boundaries between mood and psychosis disorder using dynamic functional network connectivity (dFNC) via deep learning classification

The validity and reliability of diagnoses in psychiatry is a challenging topic in mental health. The current mental health categorization is based primarily on symptoms and clinical course and is not biologically validated. Among multiple ongoing efforts, neurological observations alongside clinical evaluations are considered to be potential solutions to address diagnostic problems. The Bipolar-Schizophrenia Network on Intermediate Phenotypes (B-SNIP) has published multiple papers attempting to reclassify psychotic illnesses based on biological rather than symptomatic measures. However, the effort to investigate the relationship between this new categorization approach and other neuroimaging techniques, including resting-state fMRI data, is still limited. This study focused on investigating the relationship between different psychotic disorders categorization methods and resting-state fMRI-based measures called dynamic functional network connectivity (dFNC) using state-of-the-art artificial intelligence (AI) approaches. We applied our method to 613 subjects, including individuals with psychosis and healthy controls, which were classified using both the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) and the B-SNIP biomarker-based (Biotype) approach. Statistical group differences and cross-validated classifiers were performed within each framework to assess how different categories. Results highlight interesting differences in occupancy in both DSM-IV and Biotype categorizations compared to healthy individuals, which are distributed across specific transient connectivity states. Biotypes tended to show less distinctiveness in occupancy level and included fewer cellwise differences. Classification accuracy obtained by DSM-IV and Biotype categories were both well above chance. Results provided new insights and highlighted the benefits of both DSM-IV and biology-based categories while also emphasizing the importance of future work in this direction, including employing further data types.

NeuroCrypt: Machine Learning Over Encrypted Distributed Neuroimaging Data

The field of neuroimaging can greatly benefit from building machine learning models to detect and predict diseases, and discover novel biomarkers, but much of the data collected at various organizations and research centers is unable to be shared due to privacy or regulatory concerns (especially for clinical data or rare disorders). In addition, aggregating data across multiple large studies results in a huge amount of duplicated technical debt and the resources required can be challenging or impossible for an individual site to build. Training on the data distributed across organizations can result in models that generalize much better than models trained on data from any of organizations alone. While there are approaches for decentralized sharing, these often do not provide the highest possible guarantees of sample privacy that only cryptography can provide. In addition, such approaches are often focused on probabilistic solutions. In this paper, we propose an approach that leverages the potential of datasets spread among a number of data collecting organizations by performing joint analyses in a secure and deterministic manner when only encrypted data is shared and manipulated. The approach is based on secure multiparty computation which refers to cryptographic protocols that enable distributed computation of a function over distributed inputs without revealing additional information about the inputs. It enables multiple organizations to train machine learning models on their joint data and apply the trained models to encrypted data without revealing their sensitive data to the other parties. In our proposed approach, organizations (or sites) securely collaborate to build a machine learning model as it would have been trained on the aggregated data of all the organizations combined. Importantly, the approach does not require a trusted party (i.e. aggregator), each contributing site plays an equal role in the process, and no site can learn individual data of any other site. We demonstrate effectiveness of the proposed approach, in a range of empirical evaluations using different machine learning algorithms including logistic regression and convolutional neural network models on human structural and functional magnetic resonance imaging datasets.

Deep learning encodes robust discriminative neuroimaging representations to outperform standard machine learning

Recent critical commentaries unfavorably compare deep learning (DL) with standard machine learning (SML) approaches for brain imaging data analysis. However, their conclusions are often based on pre-engineered features depriving DL of its main advantage — representation learning. We conduct a large-scale systematic comparison profiled in multiple classification and regression tasks on structural MRI images and show the importance of representation learning for DL. Results show that if trained following prevalent DL practices, DL methods have the potential to scale particularly well and substantially improve compared to SML methods, while also presenting a lower asymptotic complexity in relative computational time, despite being more complex. We also demonstrate that DL embeddings span comprehensible task-specific projection spectra and that DL consistently localizes task-discriminative brain biomarkers. Our findings highlight the presence of nonlinearities in neuroimaging data that DL can exploit to generate superior task-discriminative representations for characterizing the human brain.

Recent critical commentaries unfavorably compare deep learning (DL) with standard machine learning (SML) for brain imaging data analysis. Here, the authors show that if trained following prevalent DL practices, DL methods substantially improve compared to SML methods by encoding robust discriminative brain representations.

Addressing Inaccurate Nosology in Mental Health: A Multilabel Data Cleansing Approach for Detecting Label Noise From Structural Magnetic Resonance Imaging Data in Mood and Psychosis Disorders

BACKGROUND

Mental health diagnostic approaches are seeking to identify biological markers to work alongside advanced machine learning approaches. It is difficult to identify a biological marker of disease when the traditional diagnostic labels themselves are not necessarily valid.

METHODS

We worked with T1 structural magnetic resonance imaging data collected from 1493 individuals comprising healthy control subjects, patients with psychosis, and their unaffected first-degree relatives. Specifically, the dataset included 176 bipolar disorder probands, 134 schizoaffective disorder probands, 240 schizophrenia probands, 362 control subjects, and 581 patient relatives. We assumed that there might be noise in the diagnostic labeling process. We detected label noise by classifying the data multiple times using a support vector machine classifier, and then we flagged those individuals in which all classifiers unanimously mislabeled those subjects. Next, we assigned a new diagnostic label to these individuals, based on the biological data (magnetic resonance imaging), using an iterative data cleansing approach.

RESULTS

Simulation results showed that our method was highly accurate in identifying label noise. Both diagnostic and biotype categories showed about 65% and 63% of noisy labels, respectively, with the largest amount of relabeling occurring between the healthy control subjects and individuals with bipolar disorder and schizophrenia as well as in unaffected close relatives. The extraction of imaging features highlighted regional brain changes associated with each group.

CONCLUSIONS

This approach represents an initial step toward developing strategies that need not assume that existing mental health diagnostic categories are always valid but rather allows us to leverage this information while also acknowledging that there are misassignments.