Multiscale Semi-Markov Dynamics for Intracortical Brain-Computer Interfaces
Authors: Daniel Milstein, Jason Pacheco, Leigh Hochberg, John D. Simeral, Beata Jarosiewicz, Erik Sudderth
NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In offline experiments with recorded neural data, we demonstrate significantly improved prediction of motion directions compared to the Kalman filter. We derive an efficient online inference algorithm, enabling a clinical trial participant with tetraplegia to control a computer cursor with neural activity in real time. Our findings show that the offline decoding performance of the MSSM is superior in all respects to baseline models. We also evaluate the MSSM decoder in two online clinical research sessions, and compare head-to-head performance with the Kalman filter. |
| Researcher Affiliation | Academia | Daniel J. Milstein Department of Computer Science, Brown University, Providence, RI, USA. Jason L. Pacheco Computer Science and Artificial Intelligence Laboratory, MIT, Cambridge, MA, USA. Leigh R. Hochberg School of Engineering, Brown University, Providence, RI, USA; and Department of Neurology, Massachusetts General Hospital, Boston, MA, USA. John D. Simeral Rehabilitation R&D Service, Department of Veterans Affairs Medical Center, Providence, RI, USA; and Brown Institute for Brain Science, Brown University, Providence, RI, USA. Beata Jarosiewicz Present affiliation: Dept. of Neurosurgery, Stanford University, Stanford, CA, USA. Erik B. Sudderth Department of Computer Science, University of California, Irvine, CA, USA. |
| Pseudocode | No | The paper describes algorithms but does not provide pseudocode or a clearly labeled algorithm block. |
| Open Source Code | No | No explicit statement or link providing concrete access to the source code for the methodology described in the paper was found. A video in supplementary material is mentioned, but not code. |
| Open Datasets | No | The paper mentions using "previously recorded data from two historical sessions of i BCI use with a single participant (T9)" and data from "clinical trial participant (T10)". There is no concrete access information (link, DOI, repository, or standard dataset citation) indicating that this data is publicly available or open. |
| Dataset Splits | Yes | We analyze decoder accuracy within each session using a leave-one-out approach. Specifically, we test the decoder on each held-out block using the remaining blocks in the same session for training. We use one block for testing and the remainder for training, and average errors across the choice of test block. |
| Hardware Specification | No | The paper mentions a "constrained embedded system" but does not provide specific hardware details such as CPU/GPU models, memory, or cloud instance types used for experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details with version numbers (e.g., library or solver names with versions). |
| Experiment Setup | Yes | At each time t we represent discrete cursor aim θt as 72 values in [0, 2π) and goal position gt as a regular grid of 40 40 = 1600 locations. The concentration parameter κ encodes the expected accuracy of user aim. The outlier weight 0 < α < 1 adds robustness to these noise bursts. In experiments our duration distributions were uniform, with parameters informed by knowledge about typical trajectory durations and reaction times. We used the feature selection method proposed by Malik et al. [2015] to select D = 60 channels of neural data. For MSSM we tried two values of η, which controls the sampling of goal states (6), and chose the remaining parameters offline. |