#10: Junifer & Julearn:From Neuroimaging to MachineLearning Models Without Expert-level Coding Skill скачать в хорошем качестве

#10: Junifer & Julearn:From Neuroimaging to MachineLearning Models Without Expert-level Coding Skill

Speaker: Federico Raimondo (Team Leader at Institute of Neuroscience and Medicine [INM-7: Brain and Behaviour], Research Centre Jülich, Jülich, Germany) Thanks to big data and computational power, the study of brain-cognition relationships using neuroimaging and machine learning (ML) has gained significant popularity [1]. Importantly, decisions in data processing [2] and predictive modelling [3] strongly impact the results. Also, misconceptions about ML procedures can distort or even invalidate findings [4], hence escalating the neuroimaging reproducibility crisis [5]. These decisions and implementations can become increasingly complex, posing challenges for early-career researchers: they require proficiency in diverse skills to deal with large-scale datasets, algorithms, and complex ML set-ups, while they also require domain-specific knowledge for experiment design and interpretation. And while this might not seem a particular challenge for experimented programmers, mistakes or omissions in the experimental design are usually the reason why findings are found to be overestimated if not invalid [5]. Most importantly, due to the predictive nature of ML-based research, in most cases the solution involves acquiring an additional validation dataset, which is prohibitive in most of the cases. With the objective of lowering the programming skills required and avoiding coding errors, we introduce two complementary instruments: Junifer, a tool to process large-scale neuroimaging datasets and extract tabular features for ML applications, which does not require coding. And Julearn, a python-based library that enables users to build and compare ML models from any tabular dataset minimising the code complexity, lowering the coding skills required to perform advanced ML-based analysis. This way, researchers can focus on the experimental design and interpretation of the results, without doubting about the correctness of their codes. Aimed at both early career researchers and senior scientists, this webinar will address the development of large-scale ML-based neuroimaging research projects from beginning to end. First, we will look at the overall analysis, including how to design and plan its implementation. In a second step, we will focus on processing neuroimaging data to extract the relevant information that will be used for the analysis. Finally, we will showcase how to build and evaluate machine learning models using neuroimaging and behavioural information. References: [1] J. Wu, J. Li, S. B. Eickhoff, D. Scheinost, and S. Genon, ‘The challenges and prospects of brain-based prediction of behaviour’, Nat Hum Behav, vol. 7, no. 8, Art. no. 8, Aug. 2023, doi: 10.1038/s41562-023-01670-1. [2] G. Antonopoulos, S. More, F. Raimondo, S. B. Eickhoff, F. Hoffstaedter, and K. R. Patil, ‘A systematic comparison of VBM pipelines and their application to age prediction’, NeuroImage, vol. 279, p. 120292, Oct. 2023, doi: 10.1016/j.neuroimage.2023.120292. [3] S. More et al., ‘Brain-age prediction: A systematic comparison of machine learning workflows’, Neuroimage, vol. 270, p. 119947, Apr. 2023, doi: 10.1016/j.neuroimage.2023.119947. [4] L. Sasse et al., ‘On Leakage in Machine Learning Pipelines’. arXiv, Nov. 07, 2023. doi: 10.48550/arXiv.2311.04179. [5] K. J. Gorgolewski and R. A. Poldrack, ‘A Practical Guide for Improving Transparency and Reproducibility in Neuroimaging Research’, PLOS Biology, vol. 14, no. 7, p. e1002506, Jul. 2016, doi: 10.1371/journal.pbio.1002506.