A robust spike sorting method based on the joint optimization of linear discrimination analysis and density peaks

Zhang, Yiwei; Liu, Tengjun; Yang, Zhenjie; Chen, Weidong; Zhang, Shaomin

doi:10.1038/s41598-022-19771-8

Cited by 5 publications

(7 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SPDF+KM is a recent approach that considers a set of 24 phase, shape, and distribution features and KM to derive clusters. We have chosen SPDF+KM for three main reasons; first because we would like to validate the discriminative power of the set of 24 features in our datasets; second, because it has achieved very good results compared with other methods in [7] , and finally, because we have checked that it is also superior to those approaches presented in [14] for the simulated data sets common to the latter and the present paper, eliminating overlapping spikes, which is the setting in [14] . This latter comment also indicates that it is a very difficult competitor to beat and that we are presenting a fair evaluation of our method.…”

Section: Methodsmentioning

confidence: 99%

“…In this work, six different datasets have been analyzed for which the real class labels (ground truth) are known, encompassing a wide variety of real and simulated spike waveforms. The simulated datasets correspond to extracellular recordings provided by [41] , and have been used by many other authors as benchmarking datasets since then, such as [8] , [12] , [4] , [13] and [14] , among others. Specifically, we consider four sets of data E11, E21, D11, and D21 with the same noise level equal to 0.1 and three different waveforms each.…”

Section: Datasets For Evaluationmentioning

confidence: 99%

“…However, a complete list of alternative techniques used in stages 2 and 3 in SS is large. Some of the most recent proposals are: [5] , [6] , [2] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] .

Figure 1 Overview of SS: stages 0 and 1 (panels A and B), stages 2 and 3 in traditional SS procedures (panel C), and single model-based clustering stage in the MixFMM model (panel D).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Functional clustering of neuronal signals with FMM mixture models

Rueda,

Rodríguez-Collado

2023

Heliyon

View full text Add to dashboard Cite

Section: Methodsmentioning

confidence: 99%

Section: Datasets For Evaluationmentioning

confidence: 99%

Figure 1 Overview of SS: stages 0 and 1 (panels A and B), stages 2 and 3 in traditional SS procedures (panel C), and single model-based clustering stage in the MixFMM model (panel D).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Functional clustering of neuronal signals with FMM mixture models

Rueda,

Rodríguez-Collado

2023

Heliyon

View full text Add to dashboard Cite

“…We analyzed our MRI and questionnaire data through supervised machine learning to select the best parameters that could divide the patient and the healthy group and thus observe if any form of clustering was present [ 27 , 28 ]. Firstly, the missing data in the dataset were imputed through multiple imputations by predictive mean matching [ 29 ].…”

Section: Methodsmentioning

confidence: 99%

MRI-Based Quantification of Pan-Alimentary Function and Motility in Subjects with Diabetes and Gastrointestinal Symptoms

Bertoli,

Mark,

Liao

et al. 2023

JCM

View full text Add to dashboard Cite

Background: Diabetes-induced gastrointestinal (GI) symptoms are common but difficult to correctly diagnose and manage. We used multi-segmental magnetic resonance imaging (MRI) to evaluate structural and functional GI parameters in diabetic patients and to study the association with their symptomatic presentation. Methods: Eighty-six participants (46 with diabetes and GI symptoms, 40 healthy controls) underwent baseline and post-meal MRI scans at multiple timepoints. Questionnaires were collected at inclusion and following the scans. Data were collected from the stomach, small bowel, and colon. Associations between symptoms and collected data were explored. Utilizing machine learning, we determined which features differentiated the two groups the most. Key Results: The patient group reported more symptoms at inclusion and during MRI scans. They showed 34% higher stomach volume at baseline, 40% larger small bowel volume, 30% smaller colon volume, and less small bowel motility postprandially. They also showed positive associations between gastric volume and satiety scores, gastric emptying time and reflux scores, and small bowel motility and constipation scores. No differences in gastric emptying were observed. Small bowel volume and motility were used as inputs to a classification tool that separated patients and controls with 76% accuracy. Conclusions: In this work, we studied structural and functional differences between patients with diabetes and GI symptoms and healthy controls and observed differences in stomach, small bowel, and colon volumes, as well as an adynamic small bowel in patients with diabetes and GI symptoms. Associations between recorded parameters and perceived symptoms were also explored and discussed.

show abstract

“…The cost function as in equation ( 16) is optimized by updating matrix W and then class label i iteratively. LDA has recently emerged in the spike sorting field [60,61] due to its ability to ability to provide better class separability especially in high-noise recordings.…”

Section: Salient Featuresmentioning

confidence: 99%

Spike sorting algorithms and their efficient hardware implementation: a comprehensive survey

et al. 2023

View full text Add to dashboard Cite

Objective: Spike sorting is a set of techniques used to analyze extracellular neural recordings, attributing individual spikes to individual neurons. This field has gained significant interest in neuroscience due to advances in implantable microelectrode arrays, capable of recording thousands of neurons simultaneously. High-density electrodes, combined with efficient and accurate spike sorting systems, are essential for variousapplications, including Brain Machine Interfaces (BMI), experimental neural prosthetics, real-time neurological disorder monitoring, and neuroscience research. However, given the resource constraints of modern applications, relying solely on algorithmic innovation is not enough. Instead, a co-optimization approach that combines hardware and spike sorting algorithms must be taken to develop neural recording systems suitable for resource-constrained environments, such as wearable devices and BMIs. This co-design requires careful consideration when selecting appropriate spike-sorting algorithms that match specific hardware and use cases. Approach: We investigated the recent literature on spike sorting, both in terms of hardware advancements and algorithms innovations. Moreover, we dedicated special attention to identifying suitable algorithm-hardware combinations, and their respective real-world applicabilities. Main Results: In this review, we first examined the current progress in algorithms, and described the recent departure from the conventional ”3-step” algorithms in favor of more advanced template matching or machine-learning-based techniques. Next, we explored innovative hardware options, including Application-Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), and In-Memory Computing Devices (IMCs). Additionally, the challenges and future opportunities for spike sorting are discussed.Significance: This comprehensive review systematically summarizes the latest spike sorting techniques and demonstrates how they enable researchers to overcome traditional obstacles and unlock novel applications. Our goal is for this work to serve as a roadmap for future researchers seeking to identify the most appropriate spike sorting implementations for various experimental settings. By doing so, we aim to facilitate the advancement of this exciting field and promote the development of innovative solutions that drive progress in neural engineeringresearch.

show abstract

A robust spike sorting method based on the joint optimization of linear discrimination analysis and density peaks

Cited by 5 publications

References 52 publications

Functional clustering of neuronal signals with FMM mixture models

Functional clustering of neuronal signals with FMM mixture models

MRI-Based Quantification of Pan-Alimentary Function and Motility in Subjects with Diabetes and Gastrointestinal Symptoms

Spike sorting algorithms and their efficient hardware implementation: a comprehensive survey

Contact Info

Product

Resources

About