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.

Characteristics of amnestic patients with hypometabolism patterns suggestive of Lewy body pathology

A clinical diagnosis of Alzheimer's disease dementia (ADD) encompasses considerable pathological and clinical heterogeneity. While Alzheimer's disease patients typically show a characteristic temporo-parietal pattern of glucose hypometabolism on 18F-fluorodeoxyglucose (FDG)-PET imaging, previous studies have identified a subset of patients showing a distinct posterior-occipital hypometabolism pattern associated with Lewy body pathology. Here, we aimed to improve the understanding of the clinical relevance of these posterior-occipital FDG-PET patterns in patients with Alzheimer's disease-like amnestic presentations. Our study included 1214 patients with clinical diagnoses of ADD (n = 305) or amnestic mild cognitive impairment (aMCI, n = 909) from the Alzheimer's Disease Neuroimaging Initiative, who had FDG-PET scans available. Individual FDG-PET scans were classified as being suggestive of Alzheimer's (AD-like) or Lewy body (LB-like) pathology by using a logistic regression classifier trained on a separate set of patients with autopsy-confirmed Alzheimer's disease or Lewy body pathology. AD- and LB-like subgroups were compared on amyloid-β and tau-PET, domain-specific cognitive profiles (memory versus executive function performance), as well as the presence of hallucinations and their evolution over follow-up (≈6 years for aMCI, ≈3 years for ADD). Around 12% of the aMCI and ADD patients were classified as LB-like. For both aMCI and ADD patients, the LB-like group showed significantly lower regional tau-PET burden than the AD-like subgroup, but amyloid-β load was only significantly lower in the aMCI LB-like subgroup. LB- and AD-like subgroups did not significantly differ in global cognition (aMCI: d = 0.15, P = 0.16; ADD: d = 0.02, P = 0.90), but LB-like patients exhibited a more dysexecutive cognitive profile relative to the memory deficit (aMCI: d = 0.35, P = 0.01; ADD: d = 0.85 P < 0.001), and had a significantly higher risk of developing hallucinations over follow-up [aMCI: hazard ratio = 1.8, 95% confidence interval = (1.29, 3.04), P = 0.02; ADD: hazard ratio = 2.2, 95% confidence interval = (1.53, 4.06) P = 0.01]. In summary, a sizeable group of clinically diagnosed ADD and aMCI patients exhibit posterior-occipital FDG-PET patterns typically associated with Lewy body pathology, and these also show less abnormal Alzheimer's disease biomarkers as well as specific clinical features typically associated with dementia with Lewy bodies.

Longitudinal changes in hippocampal texture from healthy aging to Alzheimer's disease

Early detection of Alzheimer's disease is essential to develop preventive treatment strategies. Detectible change in brain volume emerges relatively late in the pathogenic progression of disease, but microstructural changes caused by early neuropathology may cause subtle changes in the MR signal, quantifiable using texture analysis. Texture analysis quantifies spatial patterns in an image, such as smoothness, randomness and heterogeneity. We investigated whether the MRI texture of the hippocampus, an early site of Alzheimer's disease pathology, is sensitive to changes in brain microstructure before the onset of cognitive impairment. We also explored the longitudinal trajectories of hippocampal texture across the Alzheimer's continuum in relation to hippocampal volume and other biomarkers. Finally, we assessed the ability of texture to predict future cognitive decline, over and above hippocampal volume. Data were acquired from the Alzheimer's Disease Neuroimaging Initiative. Texture was calculated for bilateral hippocampi on 3T T1-weighted MRI scans. Two hundred and ninety-three texture features were reduced to five principal components that described 88% of total variance within cognitively unimpaired participants. We assessed cross-sectional differences in these texture components and hippocampal volume between four diagnostic groups: cognitively unimpaired amyloid-β- (n = 406); cognitively unimpaired amyloid-β+ (n = 213); mild cognitive impairment amyloid-β+ (n = 347); and Alzheimer's disease dementia amyloid-β+ (n = 202). To assess longitudinal texture change across the Alzheimer's continuum, we used a multivariate mixed-effects spline model to calculate a 'disease time' for all timepoints based on amyloid PET and cognitive scores. This was used as a scale on which to compare the trajectories of biomarkers, including volume and texture of the hippocampus. The trajectories were modelled in a subset of the data: cognitively unimpaired amyloid-β- (n = 345); cognitively unimpaired amyloid-β+ (n = 173); mild cognitive impairment amyloid-β+ (n = 301); and Alzheimer's disease dementia amyloid-β+ (n = 161). We identified a difference in texture component 4 at the earliest stage of Alzheimer's disease, between cognitively unimpaired amyloid-β- and cognitively unimpaired amyloid-β+ older adults (Cohen's d = 0.23, Padj = 0.014). Differences in additional texture components and hippocampal volume emerged later in the disease continuum alongside the onset of cognitive impairment (d = 0.30-1.22, Padj < 0.002). Longitudinal modelling of the texture trajectories revealed that, while most elements of texture developed over the course of the disease, noise reduced sensitivity for tracking individual textural change over time. Critically, however, texture provided additional information than was provided by volume alone to more accurately predict future cognitive change (d = 0.32-0.63, Padj < 0.0001). Our results support the use of texture as a measure of brain health, sensitive to Alzheimer's disease pathology, at a time when therapeutic intervention may be most effective.