Transient Motion Classification Through Turbid Volumes via Parallelized Single-Photon Detection and Deep Contrastive Embedding
Fast noninvasive probing of spatially varying decorrelating events, such as cerebral blood flow beneath the human skull, is an essential task in various scientific and clinical settings. One of the primary optical techniques used is diffuse correlation spectroscopy (DCS), whose classical implementat...
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-07-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2022.908770/full |
| _version_ | 1811325476376936448 |
|---|---|
| author | Shiqi Xu Wenhui Liu Wenhui Liu Xi Yang Joakim Jönsson Ruobing Qian Paul McKee Kanghyun Kim Pavan Chandra Konda Kevin C. Zhou Lucas Kreiß Lucas Kreiß Haoqian Wang Edouard Berrocal Scott A. Huettel Roarke Horstmeyer Roarke Horstmeyer |
| author_facet | Shiqi Xu Wenhui Liu Wenhui Liu Xi Yang Joakim Jönsson Ruobing Qian Paul McKee Kanghyun Kim Pavan Chandra Konda Kevin C. Zhou Lucas Kreiß Lucas Kreiß Haoqian Wang Edouard Berrocal Scott A. Huettel Roarke Horstmeyer Roarke Horstmeyer |
| author_sort | Shiqi Xu |
| collection | DOAJ |
| description | Fast noninvasive probing of spatially varying decorrelating events, such as cerebral blood flow beneath the human skull, is an essential task in various scientific and clinical settings. One of the primary optical techniques used is diffuse correlation spectroscopy (DCS), whose classical implementation uses a single or few single-photon detectors, resulting in poor spatial localization accuracy and relatively low temporal resolution. Here, we propose a technique termed ClassifyingRapid decorrelationEvents viaParallelized single photon dEtection (CREPE), a new form of DCS that can probe and classify different decorrelating movements hidden underneath turbid volume with high sensitivity using parallelized speckle detection from a 32 × 32 pixel SPAD array. We evaluate our setup by classifying different spatiotemporal-decorrelating patterns hidden beneath a 5 mm tissue-like phantom made with rapidly decorrelating dynamic scattering media. Twelve multi-mode fibers are used to collect scattered light from different positions on the surface of the tissue phantom. To validate our setup, we generate perturbed decorrelation patterns by both a digital micromirror device (DMD) modulated at multi-kilo-hertz rates, as well as a vessel phantom containing flowing fluid. Along with a deep contrastive learning algorithm that outperforms classic unsupervised learning methods, we demonstrate our approach can accurately detect and classify different transient decorrelation events (happening in 0.1–0.4 s) underneath turbid scattering media, without any data labeling. This has the potential to be applied to non-invasively monitor deep tissue motion patterns, for example identifying normal or abnormal cerebral blood flow events, at multi-Hertz rates within a compact and static detection probe. |
| first_indexed | 2024-04-13T14:34:56Z |
| format | Article |
| id | doaj.art-7bab300deff84ef7a093752ee5c382b3 |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-04-13T14:34:56Z |
| publishDate | 2022-07-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-7bab300deff84ef7a093752ee5c382b32022-12-22T02:43:05ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2022-07-011610.3389/fnins.2022.908770908770Transient Motion Classification Through Turbid Volumes via Parallelized Single-Photon Detection and Deep Contrastive EmbeddingShiqi Xu0Wenhui Liu1Wenhui Liu2Xi Yang3Joakim Jönsson4Ruobing Qian5Paul McKee6Kanghyun Kim7Pavan Chandra Konda8Kevin C. Zhou9Lucas Kreiß10Lucas Kreiß11Haoqian Wang12Edouard Berrocal13Scott A. Huettel14Roarke Horstmeyer15Roarke Horstmeyer16Department of Biomedical Engineering, Duke University, Durham, NC, United StatesDepartment of Biomedical Engineering, Duke University, Durham, NC, United StatesDepartment of Automation, Tsinghua University, Beijing, ChinaDepartment of Biomedical Engineering, Duke University, Durham, NC, United StatesDivision of Combustion Physics, Department of Physics, Lund University, Lund, SwedenDepartment of Biomedical Engineering, Duke University, Durham, NC, United StatesDepartment of Psychology and Neuroscience, Duke University, Durham, NC, United StatesDepartment of Biomedical Engineering, Duke University, Durham, NC, United StatesDepartment of Biomedical Engineering, Duke University, Durham, NC, United StatesDepartment of Biomedical Engineering, Duke University, Durham, NC, United StatesDepartment of Biomedical Engineering, Duke University, Durham, NC, United StatesInstitute of Medical Biotechnology, Friedrich-Alexander-University Erlangen-Nürnberg (FAU), Erlangen, GermanyTsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen, ChinaDivision of Combustion Physics, Department of Physics, Lund University, Lund, SwedenDepartment of Psychology and Neuroscience, Duke University, Durham, NC, United StatesDepartment of Biomedical Engineering, Duke University, Durham, NC, United StatesDepartment of Electrical Engineering, Duke University, Durham, NC, United StatesFast noninvasive probing of spatially varying decorrelating events, such as cerebral blood flow beneath the human skull, is an essential task in various scientific and clinical settings. One of the primary optical techniques used is diffuse correlation spectroscopy (DCS), whose classical implementation uses a single or few single-photon detectors, resulting in poor spatial localization accuracy and relatively low temporal resolution. Here, we propose a technique termed ClassifyingRapid decorrelationEvents viaParallelized single photon dEtection (CREPE), a new form of DCS that can probe and classify different decorrelating movements hidden underneath turbid volume with high sensitivity using parallelized speckle detection from a 32 × 32 pixel SPAD array. We evaluate our setup by classifying different spatiotemporal-decorrelating patterns hidden beneath a 5 mm tissue-like phantom made with rapidly decorrelating dynamic scattering media. Twelve multi-mode fibers are used to collect scattered light from different positions on the surface of the tissue phantom. To validate our setup, we generate perturbed decorrelation patterns by both a digital micromirror device (DMD) modulated at multi-kilo-hertz rates, as well as a vessel phantom containing flowing fluid. Along with a deep contrastive learning algorithm that outperforms classic unsupervised learning methods, we demonstrate our approach can accurately detect and classify different transient decorrelation events (happening in 0.1–0.4 s) underneath turbid scattering media, without any data labeling. This has the potential to be applied to non-invasively monitor deep tissue motion patterns, for example identifying normal or abnormal cerebral blood flow events, at multi-Hertz rates within a compact and static detection probe.https://www.frontiersin.org/articles/10.3389/fnins.2022.908770/fullSPAD arrayself-supervised learningzero-shot learningcontrastive learningmultimode fiberdiffuse correlation |
| spellingShingle | Shiqi Xu Wenhui Liu Wenhui Liu Xi Yang Joakim Jönsson Ruobing Qian Paul McKee Kanghyun Kim Pavan Chandra Konda Kevin C. Zhou Lucas Kreiß Lucas Kreiß Haoqian Wang Edouard Berrocal Scott A. Huettel Roarke Horstmeyer Roarke Horstmeyer Transient Motion Classification Through Turbid Volumes via Parallelized Single-Photon Detection and Deep Contrastive Embedding Frontiers in Neuroscience SPAD array self-supervised learning zero-shot learning contrastive learning multimode fiber diffuse correlation |
| title | Transient Motion Classification Through Turbid Volumes via Parallelized Single-Photon Detection and Deep Contrastive Embedding |
| title_full | Transient Motion Classification Through Turbid Volumes via Parallelized Single-Photon Detection and Deep Contrastive Embedding |
| title_fullStr | Transient Motion Classification Through Turbid Volumes via Parallelized Single-Photon Detection and Deep Contrastive Embedding |
| title_full_unstemmed | Transient Motion Classification Through Turbid Volumes via Parallelized Single-Photon Detection and Deep Contrastive Embedding |
| title_short | Transient Motion Classification Through Turbid Volumes via Parallelized Single-Photon Detection and Deep Contrastive Embedding |
| title_sort | transient motion classification through turbid volumes via parallelized single photon detection and deep contrastive embedding |
| topic | SPAD array self-supervised learning zero-shot learning contrastive learning multimode fiber diffuse correlation |
| url | https://www.frontiersin.org/articles/10.3389/fnins.2022.908770/full |
| work_keys_str_mv | AT shiqixu transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT wenhuiliu transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT wenhuiliu transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT xiyang transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT joakimjonsson transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT ruobingqian transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT paulmckee transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT kanghyunkim transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT pavanchandrakonda transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT kevinczhou transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT lucaskreiß transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT lucaskreiß transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT haoqianwang transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT edouardberrocal transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT scottahuettel transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT roarkehorstmeyer transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding AT roarkehorstmeyer transientmotionclassificationthroughturbidvolumesviaparallelizedsinglephotondetectionanddeepcontrastiveembedding |