Long-term stability of cortical population dynamics underlying consistent behavior

Nature Neuroscience, Год журнала: 2020, Номер 23(2), С. 260 - 270

Опубликована: Янв. 6, 2020

Язык: Английский

---

The basal ganglia control the detailed kinematics of learned motor skills

Nature Neuroscience, Год журнала: 2021, Номер 24(9), С. 1256 - 1269

Опубликована: Июль 15, 2021

Язык: Английский

---

Ultraflexible electrode arrays for months-long high-density electrophysiological mapping of thousands of neurons in rodents

Nature Biomedical Engineering, Год журнала: 2022, Номер 7(4), С. 520 - 532

Опубликована: Окт. 3, 2022

Язык: Английский

---

Class imbalance should not throw you off balance: Choosing the right classifiers and performance metrics for brain decoding with imbalanced data

NeuroImage, Год журнала: 2023, Номер 277, С. 120253 - 120253

Опубликована: Июнь 28, 2023

Machine learning (ML) is increasingly used in cognitive, computational and clinical neuroscience. The reliable efficient application of ML requires a sound understanding its subtleties limitations. Training models on datasets with imbalanced classes particularly common problem, it can have severe consequences if not adequately addressed. With the neuroscience user mind, this paper provides didactic assessment class imbalance problem illustrates impact through systematic manipulation data ratios (i) simulated (ii) brain recorded electroencephalography (EEG), magnetoencephalography (MEG) functional magnetic resonance imaging (fMRI). Our results illustrate how widely-used Accuracy (Acc) metric, which measures overall proportion successful predictions, yields misleadingly high performances, as increases. Because Acc weights per-class correct predictions proportionally to size, largely disregards performance minority class. A binary classification model that learns systematically vote for majority will yield an artificially decoding accuracy directly reflects between two classes, rather than any genuine generalizable ability discriminate them. We show other evaluation metrics such Area Under Curve (AUC) Receiver Operating Characteristic (ROC), less Balanced (BAcc) metric - defined arithmetic mean sensitivity specificity, provide more evaluations data. findings also highlight robustness Random Forest (RF), benefits using stratified cross-validation hyperprameter optimization tackle imbalance. Critically, applications seek minimize error, we recommend routine use BAcc, specific case balanced equivalent standard Acc, readily extends multi-class settings. Importantly, present list recommendations dealing data, well open-source code allow community replicate extend our observations explore alternative approaches coping

Язык: Английский

---

Preserved neural dynamics across animals performing similar behaviour

Nature, Год журнала: 2023, Номер 623(7988), С. 765 - 771

Опубликована: Ноя. 8, 2023

Abstract Animals of the same species exhibit similar behaviours that are advantageously adapted to their body and environment. These shaped at level by selection pressures over evolutionary timescales. Yet, it remains unclear how these common behavioural adaptations emerge from idiosyncratic neural circuitry each individual. The overall organization circuits is preserved across individuals 1 because evolutionarily specified developmental programme 2–4 . Such circuit may constrain activity 5–8 , leading low-dimensional latent dynamics population 9–11 Accordingly, here we suggested shared circuit-level constraints within a would lead suitably individuals. We analysed recordings populations monkey mouse motor cortex demonstrate in surprisingly when they perform behaviour. Neural were also animals consciously planned future movements without overt behaviour 12 enabled decoding ongoing movement different Furthermore, found extend beyond cortical regions dorsal striatum, an older structure 13,14 Finally, used network models similarity necessary but not sufficient for this preservation. posit emergent result on brain development thus reflect fundamental properties basis

Язык: Английский

---

Brain control of bimanual movement enabled by recurrent neural networks

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Янв. 18, 2024

Brain-computer interfaces have so far focused largely on enabling the control of a single effector, for example computer cursor or robotic arm. Restoring multi-effector motion could unlock greater functionality people with paralysis (e.g., bimanual movement). However, it may prove challenging to decode simultaneous multiple effectors, as we recently found that compositional neural code links movements across all limbs and tuning changes nonlinearly during dual-effector motion. Here, demonstrate feasibility high-quality two cursors via network (NN) decoders. Through simulations, show NNs leverage 'laterality' dimension distinguish between left right-hand both hands become increasingly correlated. In training recurrent networks (RNNs) two-cursor control, developed method alters temporal structure data by dilating/compressing in time re-ordering it, which helps RNNs successfully generalize online setting. With this method, person can simultaneously. Our results suggest decoders be advantageous decoding, provided they are designed transfer

Язык: Английский

---

Motor cortex retains and reorients neural dynamics during motor imagery

Nature Human Behaviour, Год журнала: 2024, Номер 8(4), С. 729 - 742

Опубликована: Янв. 29, 2024

Язык: Английский

---

The roles of supervised machine learning in systems neuroscience

Progress in Neurobiology, Год журнала: 2019, Номер 175, С. 126 - 137

Опубликована: Фев. 9, 2019

Язык: Английский

---

Altered brain-wide auditory networks in a zebrafish model of fragile X syndrome

BMC Biology, Год журнала: 2020, Номер 18(1)

Опубликована: Сен. 16, 2020

Abstract Background Loss or disrupted expression of the FMR1 gene causes fragile X syndrome (FXS), most common monogenetic form autism in humans. Although disruptions sensory processing are core traits FXS and autism, neural underpinnings these phenotypes poorly understood. Using calcium imaging to record from entire brain at cellular resolution, we investigated neuronal responses visual auditory stimuli larval zebrafish, using fmr1 mutants model FXS. The purpose this study was alterations networks, brain-wide that underlie aspects autism. Results Combining functional analyses with neurons’ anatomical positions, found −/− animals have normal motion. However, there were several animals. Auditory more plentiful hindbrain structures thalamus. thalamus, torus semicircularis, tegmentum had clusters neurons responded strongly Functional connectivity networks showed inter-regional lower sound intensities (a − 3 6 dB shift) larvae compared wild type. Finally, decoding capacities specific components ascending pathway altered: octavolateralis nucleus within significantly stronger amplitude while telencephalon weaker mutants. Conclusions We demonstrated hypersensitive sound, a 3–6 shift sensitivity, identified four sub-cortical regions and/or greater response strengths stimuli. also constructed an experimentally supported how information may be processed larvae. Our suggests early transmits information, less filtering modulation,

Язык: Английский

---

Motor imagery classification in brain-machine interface with machine learning algorithms: Classical approach to multi-layer perceptron model

Biomedical Signal Processing and Control, Год журнала: 2021, Номер 71, С. 103101 - 103101

Опубликована: Сен. 2, 2021

Язык: Английский

---