A Siamese Network With Node Convolution for Individualized Predictions Based on Connectivity Maps Extracted From Resting-State fMRI Data.

Deep learning has demonstrated great potential for objective diagnosis of neuropsychiatric disorders based on neuroimaging data, which includes the promising resting-state functional magnetic resonance imaging (RS-fMRI). However, the insufficient sample size has long been a bottleneck for deep model training for the purpose. In this study, we proposed a Siamese network with node convolution (SNNC) for individualized predictions based on RS-fMRI data. With the involvement of Siamese network, which uses sample pair (rather than a single sample) as input, the problem of insufficient sample size can largely be alleviated. To adapt to connectivity maps extracted from RS-fMRI data, we applied node convolution to each of the two branches of the Siamese network. For regression purposes, we replaced the contrastive loss in classic Siamese network with the mean square error loss and thus enabled Siamese network to quantitatively predict label differences. The label of a test sample can be predicted based on any of the training samples, by adding the label of the training sample to the predicted label difference between them. The final prediction for a test sample in this study was made by averaging the predictions based on each of the training samples. The performance of the proposed SNNC was evaluated with age and IQ predictions based on a public dataset (Cam-CAN). The results indicated that SNNC can make effective predictions even with a sample size of as small as 40, and SNNC achieved state-of-the-art accuracy among a variety of deep models and standard machine learning approaches.

The abilities of movie-watching functional connectivity in individual identifications and individualized predictions.

Quite a few studies have been performed based on movie-watching functional connectivity (FC). As compared to its resting-state counterpart, however, there is still much to know about its abilities in individual identifications and individualized predictions. To pave the way for appropriate usage of movie-watching FC, we systemically evaluated the minimum number of time points, as well as the exact functional networks, supporting individual identifications and individualized predictions of apparent traits based on it. We performed the study based on the 7T movie-watching fMRI data included in the HCP S1200 Release, and took IQ as the test case for the prediction analyses. The results indicate that movie-watching FC based on only 15 time points can support successful individual identifications (99.47%), and the connectivity contributed more to identifications were much associated with higher-order cognitive processes (the secondary visual network, the frontoparietal network and the posterior multimodal network). For individualized predictions of IQ, it was found that successful predictions necessitated 60 time points (predicted vs. actual IQ correlation significant at P < 0.05, based on 5,000 permutations), and the prediction accuracy increased logarithmically with the number of time points used for connectivity calculation. Furthermore, the connectivity that contributed more to individual identifications exhibited the strongest prediction ability. Collectively, our findings demonstrate that movie-watching FC can capture rich information about human brain function, and its ability in individualized predictions depends heavily on the length of fMRI scans.

Are inter-subject functional correlations consistent across different movies?

Movie fMRI has been increasingly used in investigations of human brain function. Inter-subject functional correlation (ISFC), which evaluates stimulus-dependent inter-regional synchrony between brains exposed to the same stimulus, is emerging as an influencing measure for movie fMRI data analyses. Before the wide application of ISFC analyses, it will be useful to investigate the degree to which they are similar and different across different movies. Based on the four movie fMRI runs of 178 subjects included in the "human connectome project (HCP) S1200 Release", we evaluated ISFCs throughout the brain and analyzed their consistency across different movies using intra-class correlation (ICC). We also investigated the generalizability of ISFC-based predictive models, which is closely related to their consistency, with sex classification and grip strength prediction used as test cases. The results showed that the intensity of ISFCs was generally weak (0.047). Except a few within-network ones (e.g., ICC of ISFC in the PON was 0.402), ISFCs throughout the brain exhibited low consistency, as indicated by a mean ICC of 0.130. The accuracies for inter-run predictions (60.7-72.8% for sex classification, and R = 0.122-0.275 for grip strength prediction) were much lower than those for intra-run predictions (73.2-83.0% for sex classification, and R = 0.325-0.403 for grip strength prediction), and this indicates poor generalizability of predictive models based on ISFCs. According to these findings, ISFC analyses capture aspects of brain function that are specific to each individual movie, and this specificity should be taken into account (in some cases might be especially useful) in future naturalistic studies.