1
|
Morales AW, Du J, Warren DJ, Fernández-Jover E, Martinez-Navarrete G, Bouteiller JMC, McCreery DC, Lazzi G. Machine learning enables non-Gaussian investigation of changes to peripheral nerves related to electrical stimulation. Sci Rep 2024; 14:2795. [PMID: 38307915 PMCID: PMC10837107 DOI: 10.1038/s41598-024-53284-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/12/2023] [Accepted: 01/30/2024] [Indexed: 02/04/2024] Open
Abstract
Electrical stimulation of the peripheral nervous system (PNS) is becoming increasingly important for the therapeutic treatment of numerous disorders. Thus, as peripheral nerves are increasingly the target of electrical stimulation, it is critical to determine how, and when, electrical stimulation results in anatomical changes in neural tissue. We introduce here a convolutional neural network and support vector machines for cell segmentation and analysis of histological samples of the sciatic nerve of rats stimulated with varying current intensities. We describe the methodologies and present results that highlight the validity of the approach: machine learning enabled highly efficient nerve measurement collection, while multivariate analysis revealed notable changes to nerves' anatomy, even when subjected to levels of stimulation thought to be safe according to the Shannon current limits.
Collapse
Affiliation(s)
- Andres W Morales
- Department of Biomedical Engineering, University of Southern California, Los Angeles, CA, 90089, USA.
| | - Jinze Du
- Department of Electrical Engineering, University of Southern California, Los Angeles, CA, 90089, USA
| | - David J Warren
- Department of Biomedical Engineering, University of Utah, Salt Lake City, UT, 84112, USA
| | | | | | - Jean-Marie C Bouteiller
- Department of Biomedical Engineering, University of Southern California, Los Angeles, CA, 90089, USA
- Institute for Technology and Medical Systems (ITEMS), Keck School of Medicine, University of Southern California, Los Angeles, CA, 90089, USA
| | | | - Gianluca Lazzi
- Department of Biomedical Engineering, University of Southern California, Los Angeles, CA, 90089, USA
- Department of Electrical Engineering, University of Southern California, Los Angeles, CA, 90089, USA
- Department of Ophthalmology, University of Southern California, Los Angeles, CA, 90089, USA
- Institute for Technology and Medical Systems (ITEMS), Keck School of Medicine, University of Southern California, Los Angeles, CA, 90089, USA
| |
Collapse
|
2
|
Computational optimization of delivery parameters to guide the development of targeted Nasal spray. Sci Rep 2023; 13:4099. [PMID: 36907909 PMCID: PMC10008197 DOI: 10.1038/s41598-023-30252-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/25/2022] [Accepted: 02/20/2023] [Indexed: 03/13/2023] Open
Abstract
Airborne transmission by droplets and aerosols is known to play a critical role in the spread of many viruses amongst which are the common flu and the more recent SARS-CoV-2 viruses. In the case of SARS-CoV-2, the nasal cavity not only constitutes an important viral entry point, but also a primary site of infection (Sungnak W. et al. Nat. Med. 26:681-687. https://doi.org/10.1038/s41591-020-0868-6 , 2020).. Although face masks are a well-established preventive measure, development of novel and easy-to-use prophylactic measures would be highly beneficial in fighting viral spread and the subsequent emergence of variants of concern (Tao K. et al. Nat Rev Genet 22:757-773. https://doi.org/10.1038/s41576-021-00408-x , 2021). Our group has been working on optimizing a nasal spray delivery system that deposits particles inside the susceptible regions of the nasal cavity to act as a mechanical barrier to impede viral entry. Here, we identify computationally the delivery parameters that maximize the protection offered by this barrier. We introduce the computational approach and quantify the protection rate obtained as a function of a broad range of delivery parameters. We also introduce a modified design and demonstrate that it significantly improves deposition, thus constituting a viable approach to protect against nasal infection of airborne viruses. We then discuss our findings and the implications of this novel system on the prevention of respiratory diseases and targeted drug delivery.
Collapse
|
3
|
Du J, Morales A, Kosta P, Bouteiller JMC, Martinez G, Warren D, Fernandez E, Lazzi G. Electrical Stimulation Induced Current Distribution in Peripheral Nerves Varies Significantly with the Extent of Nerve Damage: A Computational Study Utilizing Convolutional Neural Network and Realistic Nerve Models. INTERNATIONAL WORK-CONFERENCE ON THE INTERPLAY BETWEEN NATURAL AND ARTIFICIAL COMPUTATION 2022; 13258:526-535. [PMID: 37846407 PMCID: PMC10578432 DOI: 10.1007/978-3-031-06242-1_52] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 10/18/2023]
Abstract
Although electrical stimulation is an established treatment option for multiple central nervous and peripheral nervous system diseases, its effects on the tissue and subsequent safety of the stimulation are not well understood. Therefore, it is crucial to design stimulation protocols that maximize therapeutic efficacy while avoiding any potential tissue damage. Further, the stimulation levels need to be adjusted regularly to ensure that they are safe even with the changes to the nerve due to long-term stimulation. Using the latest advances in computing capabilities and machine learning approaches, we developed computational models of peripheral nerve stimulation based on very high-resolution cross-sectional images of the nerves. We generated nerve models constructed from non-stimulated (healthy) and over-stimulated (damaged) rat sciatic nerves to examine how the current density distribution is affected by nerve damage. Using our in-house numerical solver, the Admittance Method (AM), we computed the induced current distribution inside the nerves and compared the current penetration for healthy and damaged nerves. Our computational results indicate that when the nerve is damaged, primarily evidenced by the decreased nerve fiber packing, the current penetrates deeper inside the nerve than in the healthy case. As safety limits for electrical stimulation of biological tissue are still debated, we ultimately aim to utilize our computational models to determine refined safety criteria and help design safer and more efficacious electrical stimulation protocols.
Collapse
Affiliation(s)
- Jinze Du
- Department of Electrical Engineering, University of Southern California, Los Angeles, CA 90089, USA
- Institute for Technology and Medical Systems Innovation (ITEMS), Keck School of Medicine, University of Southern California, Los Angeles, CA 90089, USA
| | - Andres Morales
- Department of Biomedical Engineering, University of Southern California, Los Angeles, CA 90089, USA
- Institute for Technology and Medical Systems Innovation (ITEMS), Keck School of Medicine, University of Southern California, Los Angeles, CA 90089, USA
| | - Pragya Kosta
- Institute for Technology and Medical Systems Innovation (ITEMS), Keck School of Medicine, University of Southern California, Los Angeles, CA 90089, USA
| | - Jean-Marie C Bouteiller
- Department of Biomedical Engineering, University of Southern California, Los Angeles, CA 90089, USA
- Institute for Technology and Medical Systems Innovation (ITEMS), Keck School of Medicine, University of Southern California, Los Angeles, CA 90089, USA
| | - Gema Martinez
- Institute of Bioengineering, University Miguel Hernandez, Elche and CIBER-BBN, Madrid, Spain
| | - David Warren
- Department of Biomedical Engineering, University of Utah, Salt Lake City, UT 84112, USA
| | - Eduardo Fernandez
- Institute of Bioengineering, University Miguel Hernandez, Elche and CIBER-BBN, Madrid, Spain
| | - Gianluca Lazzi
- Department of Electrical Engineering, University of Southern California, Los Angeles, CA 90089, USA
- Department of Biomedical Engineering, University of Southern California, Los Angeles, CA 90089, USA
- Institute for Technology and Medical Systems Innovation (ITEMS), Keck School of Medicine, University of Southern California, Los Angeles, CA 90089, USA
| |
Collapse
|