5 November 2018 in GigaScience
Background: Deep learning is gaining importance in the prediction of cognitive states and brain pathology based on neuroimaging data. Including multiple hidden layers in artificial neural networks enables unprecedented predictive power; however, the proper training of deep neural networks requires thousands of exemplars. Collecting this amount of data is not feasible in typical neuroimaging experiments. A handy solution to this problem, which has largely fallen outside the scope of deep learning applications in neuroimaging, is to repurpose deep networks that have already been trained on large datasets by fine-tuning them to target datasets/tasks with fewer exemplars. Here, we investigated how this method, called transfer learning, can aid age category classification and regression based on brain functional connectivity patterns derived from resting-state functional magnetic resonance imaging. We trained a connectome-convolutional neural network on a larger public dataset and then examined how the knowledge learned can be used effectively to perform these tasks on smaller target datasets collected with a different type of scanner and/or imaging protocol and pre-processing pipeline. Results: Age classification on the target datasets benefitted from transfer learning. Significant improvement (~9%–13% increase in accuracy) was observed when the convolutional layers' weights were initialized based on the values learned on the public dataset and then fine-tuned to the target datasets. Transfer learning also appeared promising in improving the otherwise poor prediction of chronological age. Conclusions: Transfer learning is a plausible solution to adapt convolutional neural networks to neuroimaging data with few exemplars and different data acquisition and pre-processing protocols.
4 January 2018 in Neuroradiology
15 November 2023 in medRxiv
Background: Neurocognitive aging and the associated brain diseases impose a major social and economic burden. Therefore, substantial efforts have been put into revealing the lifestyle, neurobiological and genetic underpinnings of healthy neurocognitive aging. However, these studies take place almost exclusively in a limited number of highly-developed countries. Thus, it is an important open question to what extent their findings may generalize to neurocognitive aging in other, not yet investigated regions. Purpose: The purpose of the Hungarian Longitudinal Study of Healthy Brain Aging (HuBA) is to collect multi-modal longitudinal data on healthy neurocognitive aging to address the data gap in this field in Central and Eastern Europe. Methods: We adapted the Australian Imaging, Biomarkers and Lifestyle (AIBL) study of aging 1 study protocol to local circumstances and will collect demographic, lifestyle, mental and physical health, medication and medical history related information as well as record a series of magnetic resonance imaging (MRI) data. In addition, participants will also be offered to participate in the collection of blood samples to assess circulating inflammatory biomarkers as well as a sleep study aimed at evaluating the general sleep quality based on multi-day collection of subjective sleep questionnaires and whole-night electroencephalographic (EEG) data. Results & Discussion: Data collection will be longitudinal with 18 months between measurements and at least three sessions are intended. The collected data might reveal specific local trends or could also indicate the generalizability of previous findings. Moreover, as the HuBA protocol also offers a sleep study designed for thorough characterization of participants’ sleep quality and related factors, our extended multi-modal dataset might provide a base for incorporating these measures into healthy and clinical aging research. Conclusion: Besides its straightforward national benefits in terms of health expenditure, we hope that this Hungarian initiative could provide results valid for the whole Central and Eastern European region and could also promote aging and Alzheimer’s disease research in these countries.
1 January 2026 in Ideggyogyaszati Szemle
Background and purpose Mastering the operatic voice requires many years of intensive training and leads to a widespread, global impact on the structural and functional organization of the human brain. However, whether training and experience in producing operatic style singing develops differential neural representation for operatic and natural voices in opera singers remains unexplored. Methods We addressed this question by investigating the representation of short melodies sung in two distinct styles – a full operatic voice with vibrato (operatic style) and a straight voice without vibrato (natural style) – in working memory (WM) of professionally trained opera singers and amateur choir singers, using a distractor interference paradigm. Results The results revealed that distractor effects on WM performance for operatic, but not for natural voices significantly differed depending on the type of delay distractors (P=0.0011). WM performance for operatic voices was significantly higher in the case of congruent, operatic voice distractor stimuli as compared to incongruent, natural voice distractor stimuli (P=0.0043). Conclusion Our findings that operatic and natural voice distractors have differential effects on WM performance of both opera and choir singers for operatic voices imply that partially separate neural resources might be involved in maintaining natural and operatic voices in singers’ WM, independent of any prior opera singing training.