No abstract

Calcium imaging has become a powerful and popular technique to monitor the activity of large populations of neurons in vivo. However, for ethical considerations and despite recent technical developments, recordings are still constrained to a limited number of trials and animals. This limits the amount of data available from individual experiments and hinders the development of analysis techniques and models for more realistic size of neuronal populations. The ability to artificially synthesize realistic neuronal calcium signals could greatly alleviate this problem by scaling up the number of trials. Here we propose a Generative Adversarial Network (GAN) model to generate realistic calcium signals as seen in neuronal somata with calcium imaging. To this end, we adapt the WaveGAN architecture and train it with the Wasserstein distance. We test the model on artificial data with known ground-truth and show that the distribution of the generated signals closely resembles the underlying data distribution. Then, we train the model on real calcium signals recorded from the primary visual cortex of behaving mice and confirm that the deconvolved spike trains match the statistics of the recorded data. Together, these results demonstrate that our model can successfully generate realistic calcium imaging data, thereby providing the means to augment existing datasets of neuronal activity for enhanced data exploration and modeling. 11 Code available at github.com/bryanlimy/calciumgan Preprint. Under review.

Vast quantities of Magnetic Resonance Images (MRI) are routinely acquired in clinical practice but, to speed up acquisition, these scans are typically of a quality that is sufficient for clinical diagnosis but sub-optimal for large-scale precision medicine, computational diagnostics, and large-scale neuroimaging collaborative research. Here, we present a critic-guided framework to upsample low-resolution (often 2D) MRI full scans to help overcome these limitations. We incorporate feature-importance and self-attention methods into our model to improve the interpretability of this study. We evaluate our framework on paired low- and high-resolution brain MRI structural full scans (i.e., T1-, T2-weighted, and FLAIR sequences are simultaneously input) obtained in clinical and research settings from scanners manufactured by Siemens, Phillips, and GE. We show that the upsampled MRIs are qualitatively faithful to the ground-truth high-quality scans (PSNR = 35.39; MAE = 3.78E−3; NMSE = 4.32E−10; SSIM = 0.9852; mean normal-appearing gray/white matter ratio intensity differences ranging from 0.0363 to 0.0784 for FLAIR, from 0.0010 to 0.0138 for T1-weighted and from 0.0156 to 0.074 for T2-weighted sequences). The automatic raw segmentation of tissues and lesions using the super-resolved images has fewer false positives and higher accuracy than those obtained from interpolated images in protocols represented with more than three sets in the training sample, making our approach a strong candidate for practical application in clinical and collaborative research.

Mood disorders are among the leading causes of disease burden worldwide. They manifest with changes in mood, sleep, and motor-activity, observable with physiological data. Despite effective treatments being available, limited specialized care availability is a major bottleneck, hindering preemptive interventions. Near-continuous and passive collection of physiological data from wearables in daily life, analyzable with machine learning, could mitigate this problem, bringing mood disorders monitoring outside the doctor's office. Previous works attempted predicting a single label, e.g. disease state or a psychometric scale total score. However, clinical practice suggests that the same label can underlie different symptom profiles, requiring personalized treatment. In this work we address this limitation by proposing a new task: inferring all items from the Hamilton Depression Rating Scale (HDRS) and the Young Mania Rating Scale (YMRS), the most-widely used standardized questionnaires for assessing depression and mania symptoms respectively, the two polarities of mood disorders. Using a naturalistic, single-center cohort of patients with a mood disorder (N=75), we develop an artificial neural network (ANN) that inputs physiological data from a wearable device and scores patients on HDRS and YMRS in moderate agreement (quadratic Cohen's κ = 0.609) with assessments by a clinician. We also show that, when using as input physiological data recorded further away from when HDRS and YMRS were collected by the clinician, the ANN performance deteriorates, pointing to a distribution shift, likely across both psychometric scales and physiological data. This suggests the task is challenging and research into domain adaptation should be prioritized towards real-world implementations.

Understanding how activity in neural circuits reshapes following task learning could reveal fundamental mechanisms of learning. Thanks to the recent advances in neural imaging technologies, high-quality recordings can be obtained from hundreds of neurons over multiple days or even weeks. However, the complexity and dimensionality of population responses pose significant challenges for analysis. Existing methods of studying neuronal adaptation and learning often impose strong assumptions on the data or model, resulting in biased descriptions that do not generalize. In this work, we use a variant of deep generative models called -cycle-consistent adversarial networks, to learn the unknown mapping between pre-and post-learning neuronal activities recorded in vivo. To do so, we develop an end-to-end pipeline to preprocess, train and evaluate calcium fluorescence signals, and a procedure to interpret the resulting deep learning models. To assess the validity of our method, we first test our framework on a synthetic dataset with known ground-truth transformation. Subsequently, we applied our method to neuronal activities recorded from the primary visual cortex of behaving mice, where the mice transition from novice to expert-level performance in a visual-based virtual reality experiment. We evaluate model performance on generated calcium imaging signals and their inferred spike trains. To maximize performance, we derive a novel approach to pre-sort neurons such that convolutional-based networks can take advantage of the spatial information that exists in neuronal activities. In addition, we incorporate visual explanation methods to improve the interpretability of our work and gain insights into the learning process as manifested in the cellular activities. Together, our results demonstrate that analyzing neuronal learning processes with data-driven deep unsupervised methods holds the potential to unravel changes in an unbiased way.