Label‐free spectroscopic tissue characterization using fluorescence excitation‐scanning spectral imaging

Spectral imaging approaches provide new possibilities for measuring and discriminating fluorescent molecules in living cells and tissues. These approaches often employ tunable filters and robust image processing algorithms to identify many fluorescent labels in a single image set. Here, we present results from a novel spectral imaging technology that scans the fluorescence excitation spectrum, demonstrating that excitation‐scanning hyperspectral image data can discriminate among tissue types and estimate the molecular composition of tissues. This approach allows fast, accurate quantification of many fluorescent species from multivariate image data without the need of exogenous labels or dyes. We evaluated the ability of the excitation‐scanning approach to identify endogenous fluorescence signatures in multiple unlabeled tissue types. Signatures were screened using multi‐pass principal component analysis. Endmember extraction techniques revealed conserved autofluorescent signatures across multiple tissue types. We further examined the ability to detect known molecular signatures by constructing spectral libraries of common endogenous fluorophores and applying multiple spectral analysis techniques on test images from lung, liver and kidney. Spectral deconvolution revealed structure‐specific morphologic contrast generated from pure molecule signatures. These results demonstrate that excitation‐scanning spectral imaging, coupled with spectral imaging processing techniques, provides an approach for discriminating among tissue types and assessing the molecular composition of tissues. Additionally, excitation scanning offers the ability to rapidly screen molecular markers across a range of tissues without using fluorescent labels. This approach lays the groundwork for translation of excitation‐scanning technologies to clinical imaging platforms.     

Cd-free InP/ZnSeS quantum dots for ultrahigh-resolution imaging of stimulated emission depletion

Stimulated emission depletion (STED) nanoscopy is a promising super-resolution imaging technique for microstructure imaging; however, the performance of super-resolution techniques critically depends on the properties of the fluorophores (photostable fluorophores) used. In this study, a suitable probe for improving the resolution of STED nanoscopy was investigated. Quantum dots (QDs) typically exhibit good photobleaching resistance characteristics. In comparison with CdSe@ZnS QDs and CsPbBr₃ QDs, Cd-free InP/ZnSeS QDs have a smaller size and exhibit an improved photobleaching resistance. Through imaging using InP/ZnSeS QDs, we achieved an ultrahigh resolution of 26.1 nm. Furthermore, we achieved a 31 nm resolution in cell experiments involving InP/ZnSeS QDs. These results indicate that Cd-free InP/ZnSeS QDs have significant potential for application in fluorescent probes for STED nanoscopy.     

Improving the image quality in STED nanoscopy using frequency spectrum modulation

Since stimulated emission depletion (STED) nanoscopy was invented in 1994, this technique has been widely used in the fields of biomedicine and materials science. According to the imaging principle of STED technology, increasing the power of the depletion laser within a certain threshold can improve the resolution. However, it will cause not only severe photo-damage to the samples and photo-bleaching to the fluorophores but also serious background noise, leading to the degeneration of the quality of STED images. Here we propose a new processing method based on frequency spectrum modulation to improve the quality of STED images, abbreviated as FM-STED. We have demonstrated the performance of FM-STED in improving the signal-to-noise ratio and the resolution using fluorescent beads and biological cells as samples.     

Inside Cover

A high‐efficiency computer‐aided diagnostic model of ovarian cancer was developed, integrating SHG imaging technology for non‐invasive imaging of living tissue and machine learning method based on radiomics and TPOT. This model can rapidly, non‐destructively, and accurately perform ovarian cancer diagnosis and has great potential in improving diagnostic efficacy and efficiency of medical pathologists. Further details can be found in the article by Guangxing Wang, Yang Sun, Youting Chen, Qiqi Gao, Dongqing Peng, Hongxin Lin, Zhenlin Zhan, Zhiyi Liu, and Shuangmu Zhuo ( e202000050 ).

     

Intravascular molecular-structural imaging with a miniaturized integrated near-infrared fluorescence and ultrasound catheter

Coronary artery disease (CAD) remains a leading cause of mortality and warrants new imaging approaches to better guide clinical care. We report on a miniaturized, hybrid intravascular catheter and imaging system for comprehensive coronary artery imaging in vivo. Our catheter exhibits a total diameter of 1.0 mm (3.0 French), equivalent to standalone clinical intravascular ultrasound (IVUS) catheters but enables simultaneous near-infrared fluorescence (NIRF) and IVUS molecular-structural imaging. We demonstrate NIRF-IVUS imaging in vitro in coronary stents using NIR fluorophores, and compare NIRF signal strengths for prism and ball lens sensor designs in both low and high scattering media. Next, in vivo intravascular imaging in pig coronary arteries demonstrates simultaneous, co-registered molecular-structural imaging of experimental CAD inflammation on IVUS and distance-corrected NIRF images. The obtained results suggest substantial potential for the NIRF-IVUS catheter to advance standalone IVUS, and enable comprehensive phenotyping of vascular disease to better assess and treat patients with CAD.     

Low-power compact continuous-wave stimulated emission depletion microscopy

Stimulated emission depletion (STED) microscopy can break the optical diffraction barrier and provide subdiffraction resolution. According to the STED superresolution imaging principle, the resolution of STED is positively related to the power of the depletion laser. However, high-laser power largely limits the study of living cells or living bodies. Moreover, the high complexity and high cost of conventional pulsed STED microscopy limit the application of this technique. Therefore, this paper describes a simple continuous-wave STED (CW-STED) system constructed on a 45 × 60 cm breadboard and combined with digitally enhanced (DE) technology; low-power superresolution imaging is realized, which has the advantages of reducing system complexity and cost. The low-system complexity, low cost, and low-power superresolution imaging features of CW-STED have great potential to advance the application of STED microscopy in biological research.     

Toward absolute viability measurements for bacteria

We aim to develop a quantitative viability method that distinguishes individual quiescent from dead cells and is measured in time (ns) as a referenceable, comparable quantity. We demonstrate that fluorescence lifetime imaging of an anionic, fluorescent membrane voltage probe fulfills these requirements for Streptococcus mutans. A random forest machine-learning model assesses whether individual S. mutans can be correctly classified into their original populations: stationary phase (quiescent), heat killed and inactivated via chemical fixation. We compare the results to intensity using three models: lifetime variables (τ₁, τ₂ and p₁), phasor variables (G, S) or all five variables, with the five variable models having the most accurate classification. This initial work affirms the potential for using fluorescence lifetime of a membrane voltage probe as a viability marker for quiescent bacteria, and future efforts on other bacterial species and fluorophores will help refine this approach.     

Computed tomography for in vivo deep over‐1000 nm near‐infrared fluorescence imaging

This study aims to develop a novel cross‐sectional imaging of fluorescence in over‐1000 nm near‐infrared (OTN‐NIR), which allows in vivo deep imaging, using computed tomography (CT) system. Cylindrical specimens of composite of OTN‐NIR fluorophore, NaGdF₄ co‐doped with Yb³⁺ and Ho³⁺ (ex: 980 nm, em: 1150 nm), were embedded in cubic agar (10.5–12 mm) or in the peritoneal cavity of mice and placed on a rotatable stage. When the fluorescence from inside of the samples was serially captured from multiple angles, the images were disrupted by the reflection and refraction of emitted light on the sample‐air interface. Immersing the sample into water filled in a rectangular bath suppressed the disruption at the interface and successfully reconstructed the position and concentration of OTN‐NIR fluorophores on the cross‐sectional images using a CT technique. This is promising as a novel three‐dimensional imaging technique for OTN‐NIR fluorescent image projections of small animals captured from multiple angles.     

Unique spectral signature of human cutaneous squamous cell carcinoma by photoacoustic imaging

Cutaneous squamous cell carcinoma (cSCC) is a common skin cancer with metastatic potential. To reduce reoperations due to nonradical excision, there is a need to develop a technique for identification of tumor margins preoperatively. Photoacoustic (PA) imaging is a novel imaging technology that combines the strengths of laser optics and ultrasound. Our aim was to determine the spectral signature of cSCC using PA imaging and to use this signature to visualize tumor architecture and borders. Two‐dimensional PA images of 33 cSCCs and surrounding healthy skin were acquired ex vivo, using 59 excitation wavelengths from 680 to 970 nm. The spectral response of the cSCCs was compared to healthy tissue, and the difference was found to be greatest at wavelengths in the range 765 to 960 nm (P < .05). Three‐dimensional PA images were constructed from spectra obtained in the y‐z plane using a linear stepper motor moving along the x‐plane. Spectral unmixing was then performed which provided a clear three‐dimensional view of the distribution of tumor masses and their borders.     

Exploration research on the fusion of multimodal spectrum technology to improve performance of rapid diagnosis scheme for thyroid dysfunction

The spectral fusion by Raman spectroscopy and Fourier infrared spectroscopy combined with pattern recognition algorithms is utilized to diagnose thyroid dysfunction serum, and finds the spectral segment with the highest sensitivity to further advance diagnosis speed. Compared with the single infrared spectroscopy or Raman spectroscopy, the proposal can improve the detection accuracy, and can obtain more spectral features, indicating greater differences between thyroid dysfunction and normal serum samples. For discriminating different samples, principal component analysis (PCA) was first used for feature extraction to reduce the dimension of high‐dimension spectral data and spectral fusion. Then, support vector machine (SVM), back propagation neural network, extreme learning machine and learning vector quantization algorithms were employed to establish the discriminant diagnostic models. The accuracy of spectral fusion of the best analytical model PCA‐SVM, single Raman spectral accuracy and single infrared spectral accuracy is 83.48%, 78.26% and 80%, respectively. The accuracy of spectral fusion is higher than the accuracy of single spectrum in five classifiers. And the diagnostic accuracy of spectral fusion in the range of 2000 to 2500 cm^?1 is 81.74%, which greatly improves the sample measure speed and data analysis speed than analysis of full spectra. The results from our study demonstrate that the serum spectral fusion technique combined with multivariate statistical methods have great potential for the screening of thyroid dysfunction.     

Simultaneous multi-spectral,single-photon fluorescence imaging using a plasmonic colour filter array

We present the first realisation of simultaneous multi-spectral fluorescence imaging using a single-photon avalanche diode (SPAD) array, where the spectral unmixing is facilitated by a plasmonic metasurface mosaic colour filter array (CFA). A 64 × 64 pixel format silicon SPAD array is used to record widefield fluorescence and brightfield data from four biological samples. A plasmonic metasurface composed of an arrangement of circular and elliptical nanoholes etched into an aluminium thin film deposited on a glass substrate provides the high transmission efficiency CFA, enabling a bespoke spectral unmixing algorithm to reconstruct high fidelity, full colour images from as few as ∼3 photons per pixel. This approach points the way toward real-time, single-photon sensitive multi-spectral fluorescence imaging. Furthermore, this is possible without additional bulky components such as a filter wheel, prism or diffraction grating, nor the need for multiple sample exposures or multiple detectors.     

Multivariate analysis of Brillouin imaging data by supervised and unsupervised learning

Brillouin imaging relies on the reliable extraction of subtle spectral information from hyperspectral datasets. To date, the mainstream practice has been to use line fitting of spectral features to retrieve the average peak shift and linewidth parameters. Good results, however, depend heavily on sufficient signal-to-noise ratio and may not be applicable in complex samples that consist of spectral mixtures. In this work, we thus propose the use of various multivariate algorithms that can be used to perform supervised or unsupervised analysis of the hyperspectral data, with which we explore advanced image analysis applications, namely unmixing, classification and segmentation in a phantom and live cells. The resulting images are shown to provide more contrast and detail, and obtained on a timescale ∼10² faster than fitting. The estimated spectral parameters are consistent with those calculated from pure fitting.     

The use of optical coherence tomography and convolutional neural networks to distinguish normal and abnormal oral mucosa

Incomplete surgical resection of head and neck squamous cell carcinoma (HNSCC) is the most common cause of local HNSCC recurrence. Currently, surgeons rely on preoperative imaging, direct visualization, palpation and frozen section to determine the extent of tissue resection. It has been demonstrated that optical coherence tomography (OCT), a minimally invasive, nonionizing near infrared mesoscopic imaging modality can resolve subsurface differences between normal and abnormal head and neck mucosa. Previous work has utilized two‐dimensional OCT imaging which is limited to the evaluation of small regions of interest generated frame by frame. OCT technology is capable of performing rapid volumetric imaging, but the capacity and expertise to analyze this massive amount of image data is lacking. In this study, we evaluate the ability of a retrained convolutional neural network to classify three‐dimensional OCT images of head and neck mucosa to differentiate normal and abnormal tissues with sensitivity and specificity of 100% and 70%, respectively. This method has the potential to serve as a real‐time analytic tool in the assessment of surgical margins.     

Multispectral near infrared absorption imaging for histology of skin cancer

Multispectral imaging combines the spectral resolution of spectroscopy with the spatial resolution of imaging and is therefore very useful for biomedical applications. Currently, histological diagnostics use mainly stainings with standard dyes (eg, hematoxylin + eosin) to identify tumors. This method is not applicable in vivo and provides low amounts of chemical information. Biomolecules absorb near infrared light (NIR, 800‐1700 nm) at different wavelengths, which could be used to fingerprint tissue. Here, we built a NIR multispectral absorption imaging setup to study skin tissue samples. NIR light (900‐1500 nm) was used for homogenous wide‐field transmission illumination and detected by a cooled InGaAs camera. In this setup, images I(x, y, λ) from dermatological samples (melanoma, nodular basal‐cell carcinoma, squamous‐cell carcinoma) were acquired to distinguish healthy from diseased tissue regions. In summary, we show the potential of multispectral NIR imaging for cancer diagnostics.      

Multiphoton tomography in differentiation of morphological and molecular subtypes of breast cancer: A quantitative analysis

In this study multiphoton tomography, based on second harmonic generation (SHG), and two-photon-excited fluorescence (TPEF) was used to visualize both the extracellular matrix and tumor cells in different morphological and molecular subtypes of human breast cancer. It was shown, that quantified assessment of the SHG based imaging data has great potential to reveal differences of collagen quantity, organization and uniformity in both low- and highly- aggressive invasive breast cancers. The values of quantity and uniformity of the collagen fibers distribution were significantly higher in low-aggressive breast cancer compared to the highly-aggressive subtypes, while the value representing collagen organization was lower in the former type. Additionally, it was shown, that TPEF detection of elastin fibers and amyloid protein may be used as a biomarker of detection the low-aggressive breast cancer subtype. Thus, TPEF/SHG imaging offers the possibility of becoming a useful tool for the rapid diagnosis of various subtypes of breast cancer during biopsy as well as for the intraoperative determinination of tumor-positive resection margins.     

Deep learning a boon for biophotonics?

This review covers original articles using deep learning in the biophotonic field published in the last years. In these years deep learning, which is a subset of machine learning mostly based on artificial neural network geometries, was applied to a number of biophotonic tasks and has achieved state‐of‐the‐art performances. Therefore, deep learning in the biophotonic field is rapidly growing and it will be utilized in the next years to obtain real‐time biophotonic decision‐making systems and to analyze biophotonic data in general. In this contribution, we discuss the possibilities of deep learning in the biophotonic field including image classification, segmentation, registration, pseudostaining and resolution enhancement. Additionally, we discuss the potential use of deep learning for spectroscopic data including spectral data preprocessing and spectral classification. We conclude this review by addressing the potential applications and challenges of using deep learning for biophotonic data.     

Self-phase-modulated femtosecond laser source at 1603 nm and its application to deep-brain 3-photon microscopy in vivo

3-photon microscopy (3PM) excited at the 1700 nm window enables deep-tissue imaging in vivo, especially in brain. PC rod soliton source has previously been exclusively used as the excitation source, which is rather costly and difficult to align. Here we demonstrate a novel nonlinear optical technique to build femtosecond laser source at the 1700 nm window, based on self-phase modulation (SPM) in a short span of large-mode-area fiber. The spectral broadening experienced by the pump pulse leads to the generation of a red-shifted sidelobe at 1603 nm. After spectral filtering, this sidelobe corresponds to 170-fs, 167-nJ pulses at 1603 nm. Using this SPM source, we further demonstrate deep-brain 3 PM to a depth of 1500 μm below the mouse brain surface in vivo. Our SPM femtosecond laser source thus provides a cost effective and easy-to-align alternative excitation source to the PC rod soliton source.     

Neurosurgical brain tumor detection based on intraoperative optical intrinsic signal imaging technique: A case report of glioblastoma

The delineation of brain tumor margins has been a challenging objective in neurosurgery for decades. Despite the development of various preoperative imaging techniques, the current methodology is still insufficient for clinical practice. We present an intraoperative optical intrinsic signal imaging system for brain tumor surgery and establish a data processing procedure model to localize tumors. From the experimental result of a glioblastoma patient, we observe a relative small oscillation of ΔHbD in tumor region and speculate that vessels in tumor region have poor ability to provide oxygen. We applied the same data processing procedure on the second time data and proclaimed a successful surgery. Figure: Merged ΔHbD image captured prior and posterior to tumor removal.      

Investigation of human pancreatic cancer tissues by Fourier Transform Infrared Hyperspectral Imaging

Fourier‐transform infrared hyperspectral imaging (FTIR‐HSI) provides hyperspectral images containing both morphological and chemical information. It is widely applied in the biomedical field to detect tumor lesions, even at the early stage, by identifying specific spectral biomarkers. Pancreatic neoplasms present different prognoses and are not always easily classified by conventional analyses. In this study, tissue samples with diagnosis of pancreatic ductal adenocarcinoma and pancreatic neuroendocrine tumor were analyzed by FTIR‐HSI and the spectral data compared with those from healthy and dysplastic samples. Multivariate/univariate approaches were complemented to hyperspectral images, and definite spectral markers of the different lesions identified. The malignant lesions were recognizable both from healthy/dysplastic pancreatic tissues (high values of phospholipids and triglycerides with shorter, more branched and less unsaturated alkyl chains) and between each other (different amounts of total lipids, phosphates and carbohydrates). These findings highlight different metabolic pathways characterizing the different samples, well detectable by FTIR‐HSI.     

A new deep learning method for image deblurring in optical microscopic systems

Deconvolution is the most commonly used image processing method in optical imaging systems to remove the blur caused by the point‐spread function (PSF). While this method has been successful in deblurring, it suffers from several disadvantages, such as slow processing time due to multiple iterations required to deblur and suboptimal in cases where the experimental operator chosen to represent PSF is not optimal. In this paper, we present a deep‐learning‐based deblurring method that is fast and applicable to optical microscopic imaging systems. We tested the robustness of proposed deblurring method on the publicly available data, simulated data and experimental data (including 2D optical microscopic data and 3D photoacoustic microscopic data), which all showed much improved deblurred results compared to deconvolution. We compared our results against several existing deconvolution methods. Our results are better than conventional techniques and do not require multiple iterations or pre‐determined experimental operator. Our method has several advantages including simple operation, short time to compute, good deblur results and wide application in all types of optical microscopic imaging systems. The deep learning approach opens up a new path for deblurring and can be applied in various biomedical imaging fields.