Deep Learning in Neuroimaging: Overcoming Challenges With Emerging Approaches

Predicting Clinical Improvement in Early Psychosis Using Circuit-Based Resting-State Functional Magnetic Resonance Imaging

BACKGROUND AND HYPOTHESIS

Identifying biomarkers that predict treatment response in early psychosis (EP) is a priority for psychiatry research. Previous work suggests that resting-state connectivity biomarkers may have promise as predictive measures, although prior results vary considerably in direction and magnitude. Here, we evaluated the relationship between intrinsic functional connectivity of the attention, default mode, and salience resting-state networks and 12-month clinical improvement in EP.

STUDY DESIGN

Fifty-eight individuals with EP (less than 2 years from illness onset, 35 males, average age 20 years) had baseline and follow-up clinical data and were included in the final sample. Of these, 30 EPs showed greater than 20% improvement in Brief Psychiatric Rating Scale (BPRS) total score at follow-up and were classified as "Improvers."

STUDY RESULTS

The overall logistic regression predicting Improver status was significant (χ2 = 23.66, Nagelkerke's R2 = 0.45, P < .001, with 85% concordance). Significant individual predictors of Improver status included higher default mode within-network connectivity, higher attention-default mode between-network connectivity, and higher attention-salience between-network connectivity. Including baseline BPRS as a predictor increased model significance and concordance to 92%, and the model was not significantly influenced by the dose of antipsychotic medication (chlorpromazine equivalents). Linear regression models predicting percent change in BPRS were also significant.

CONCLUSIONS

Overall, these results suggest that resting-state functional magnetic resonance imaging connectivity may serve as a useful biomarker of clinical outcomes in recent-onset psychosis.

The enlightening role of explainable artificial intelligence in medical & healthcare domains: A systematic literature review

In domains such as medical and healthcare, the interpretability and explainability of machine learning and artificial intelligence systems are crucial for building trust in their results. Errors caused by these systems, such as incorrect diagnoses or treatments, can have severe and even life-threatening consequences for patients. To address this issue, Explainable Artificial Intelligence (XAI) has emerged as a popular area of research, focused on understanding the black-box nature of complex and hard-to-interpret machine learning models. While humans can increase the accuracy of these models through technical expertise, understanding how these models actually function during training can be difficult or even impossible. XAI algorithms such as Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) can provide explanations for these models, improving trust in their predictions by providing feature importance and increasing confidence in the systems. Many articles have been published that propose solutions to medical problems by using machine learning models alongside XAI algorithms to provide interpretability and explainability. In our study, we identified 454 articles published from 2018-2022 and analyzed 93 of them to explore the use of these techniques in the medical domain.

Causal inference in cognitive neuroscience

Causal inference is a key step in many research endeavors in cognitive science and neuroscience, and particularly cognitive neuroscience. Statistical knowledge is sufficient for prediction and diagnosis, but causal knowledge is required for action and intervention. Most statistics courses and textbooks emphasize the difficulty of causal inference, focusing on the maxim that "correlation does not mean causation": there can be multiple causal possibilities, often many of them, consistent with given observed statistics. This paper focuses instead on the conceptual issues and assumptions that confront causal and other kinds of inference, primarily focusing on cognitive neuroscience. We connect inference methods with goals and challenges, and provide concrete guidance about how to select appropriate tools for the scientific task. This article is categorized under: Psychology > Theory and Methods Philosophy > Foundations of Cognitive Science.

Data augmentation with Mixup: Enhancing performance of a functional neuroimaging-based prognostic deep learning classifier in recent onset psychosis

Although deep learning holds great promise as a prognostic tool in psychiatry, a limitation of the method is that it requires large training sample sizes to achieve replicable accuracy. This is problematic for fMRI datasets as they are typically small due to the considerable time, cost, and resources necessary to obtain them. A recently developed self-supervised learning method called Mixup may help overcome this challenge. In Mixup, the learner combines pairs of training instances to produce a virtual third instance that is a linear combination of the two instances and their labels. This procedure is also well-suited to the coregistered images typically found in fMRI datasets. Here we compared performance of a task fMRI-based deep learner with Mixup vs without Mixup on predicting response to treatment in recent onset psychosis. Whole brain fMRI time series data were extracted from a cognitive control task in 82 patients with recent onset psychosis and used to predict "Improver" (n = 47) vs "Non-Improver" (n = 35) status, with Improver defined as showing a 20 % reduction in total Brief Psychiatric Rating Scale score after 1 year of treatment. Mixup significantly improved performance (accuracy without Mixup: 76.5 % [95 % CI: 75.9-77.1 %]; accuracy with Mixup: 80.1 % [95 % CI: 79.4-80.8 %]). Ablation showed the improvement was due to improvement in both Improvers and Non-Improvers. These results suggest that using Mixup may significantly improve performance and reduce overfitting of fMRI-based prognostic deep learners and may also help overcome the small sample size challenge inherent to many neuroimaging datasets.