Sex classification from functional brain connectivity: Generalization to multiple datasets.

Machine learning (ML) approaches are increasingly being applied to neuroimaging data. Studies in neuroscience typically have to rely on a limited set of training data which may impair the generalizability of ML models. However, it is still unclear which kind of training sample is best suited to optimize generalization performance. In the present study, we systematically investigated the generalization performance of sex classification models trained on the parcelwise connectivity profile of either single samples or compound samples of two different sizes. Generalization performance was quantified in terms of mean across-sample classification accuracy and spatial consistency of accurately classifying parcels. Our results indicate that the generalization performance of parcelwise classifiers (pwCs) trained on single dataset samples is dependent on the specific test samples. Certain datasets seem to "match" in the sense that classifiers trained on a sample from one dataset achieved a high accuracy when tested on the respected other one and vice versa. The pwCs trained on the compound samples demonstrated overall highest generalization performance for all test samples, including one derived from a dataset not included in building the training samples. Thus, our results indicate that both a large sample size and a heterogeneous data composition of a training sample have a central role in achieving generalizable results.

Julearn: an easy-to-use library for leakage-free evaluation and inspection of ML models.

The fast-paced development of machine learning (ML) and its increasing adoption in research challenge researchers without extensive training in ML. In neuroscience, ML can help understand brain-behavior relationships, diagnose diseases and develop biomarkers using data from sources like magnetic resonance imaging and electroencephalography. Primarily, ML builds models to make accurate predictions on unseen data. Researchers evaluate models' performance and generalizability using techniques such as cross-validation (CV). However, choosing a CV scheme and evaluating an ML pipeline is challenging and, if done improperly, can lead to overestimated results and incorrect interpretations. Here, we created julearn, an open-source Python library allowing researchers to design and evaluate complex ML pipelines without encountering common pitfalls. We present the rationale behind julearn's design, its core features, and showcase three examples of previously-published research projects. Julearn simplifies the access to ML providing an easy-to-use environment. With its design, unique features, simple interface, and practical documentation, it poses as a useful Python-based library for research projects.

Sex classification from functional brain connectivity: Generalization to multiple datasets Generalizability of sex classifiers.

Machine learning (ML) approaches are increasingly being applied to neuroimaging data. Studies in neuroscience typically have to rely on a limited set of training data which may impair the generalizability of ML models. However, it is still unclear which kind of training sample is best suited to optimize generalization performance. In the present study, we systematically investigated the generalization performance of sex classification models trained on the parcelwise connectivity profile of either single samples or a compound sample containing data from four different datasets. Generalization performance was quantified in terms of mean across-sample classification accuracy and spatial consistency of accurately classifying parcels. Our results indicate that generalization performance of pwCs trained on single dataset samples is dependent on the specific test samples. Certain datasets seem to "match" in the sense that classifiers trained on a sample from one dataset achieved a high accuracy when tested on the respected other one and vice versa. The pwC trained on the compound sample demonstrated overall highest generalization performance for all test samples, including one derived from a dataset not included in building the training samples. Thus, our results indicate that a big and heterogenous training sample comprising data of multiple datasets is best suited to achieve generalizable results.