Search

Pathological examination is essential for cancer diagnosis. Frozen sectioning has been the gold standard for intraoperative tissue assessment, which, however, is hampered by its laborious processing steps and often provides inadequate tissue slide quality. To address these limitations, we developed a deep-learning-assisted, ultraviolet light-emitting diode (UV-LED) microscope for label-free and slide-free tissue imaging. Using UV-based light-sheet (UV-LS) imaging mode as the learning target, UV-LED images with high contrast are generated by employing a weakly supervised network for contrast enhancement. With our approach, the image acquisition speed for providing contrast-enhanced UV-LED (CE-LED) images is 47s/cm2 , ∼25 times faster than that of the UV-LS system. The results show that this approach significantly enhances the image quality of UV-LED, revealing essential tissue structures in cancerous samples. The resulting CE-LED offers a low-cost, nondestructive, and high-throughput alternative histological imaging technique for intraoperative cancer detection.

A comprehensive understanding of a tumor is required for accurate diagnosis and effective treatment. However, currently, there is no single imaging modality that can provide sufficient information. Photoacoustic (PA) imaging is a hybrid imaging technique with high spatial resolution and detection sensitivity, which can be combined with ultrasound (US) imaging to provide both optical and acoustic contrast. Elastography can noninvasively map the elasticity distribution of biological tissue, which reflects pathological conditions. In this study, we incorporated PA elastography into a commercial US/PA imaging system to develop a tri-modality imaging system, which has been tested for tumor detection using four mice with different physiological conditions. The results show that this tri-modality imaging system can provide complementary information on acoustic, optical, and mechanical properties. The enabled visualization and dimension estimation of tumors can lead to a more comprehensive tissue characterization for diagnosis and treatment.

Life science has progressed with applications of fluorescent probes—fluorophores linked to functional units responding to biological events. To meet the varied demands across experiments, simple organic reactions to connect fluorophores and functional units have been developed, enabling the on-demand selection of fluorophore-functional unit combinations. However, organic synthesis requires professional equipment and skills, standing as a daunting task for life scientists. In this study, we present a simple, fast, and convenient strategy for probe preparation: co-aggregation of hydrophobic molecules. We focused on tetrazine—a difficult-to-prepare yet useful functional unit that provides effective bioorthogonal reactivity and strong fluorogenicity. Simply mixing the tetrazine molecules and aggregation-induced emission (AIE) luminogens in water, co-aggregation is induced, and the emission of AIE luminogens is quenched. Subsequent click reaction bioorthogonally turns on the emission, identifying these coaggregates as fluorogenic probes. Thanks to this bioorthogonal fluorogenicity, we established a new time-gated fluorescence bioimaging technique to distinguish overlapping emission signals, enabling multi-organelle imaging with two same-color fluorophores. Our study showcases the potential of this co-aggregation method for the on-demand preparation of fluorescent probes as well as protocols and molecular design principles in this approach, offering an effective solution to evolving needs in life science research.

Hematologists evaluate alterations in blood cell enumeration and morphology to confirm peripheral blood smear findings through manual microscopic examination. However, routine peripheral blood smear analysis is both time-consuming and labor-intensive. Here, we propose using smartphone-based autofluorescence microscopy (Smart-AM) for imaging label-free blood smears at subcellular resolution with automatic hematological analysis. Smart-AM enables rapid and label-free visualization of morphological features of normal and abnormal blood cells (including leukocytes, erythrocytes, and thrombocytes). Moreover, assisted with deep-learning algorithms, this technique can automatically detect and classify different leukocytes with high accuracy, and transform the autofluorescence images into virtual Giemsa-stained images which show clear cellular features. The proposed technique is portable, cost-effective, and user-friendly, making it significant for broad point-of-care applications.

Rapid multicolor three-dimensional (3D) imaging for entire organs with high resolution remains a captivating scientific pursuit. A significant challenge associated with whole-organ imaging pertains to the laborious and time-consuming sample preparation and image acquisition. To this end, we first proposed a fast, cost-effective, and robust multicolor whole-organ 3D imaging method assisted with ultraviolet (UV) surface excitation and vibratomy-assisted sectioning, termed translational rapid ultraviolet-excited sectioning tomography (TRUST). With an inexpensive UV light-emitting diode (UV-LED), TRUST achieves widefield fluorescence and autofluorescence imaging simultaneously while preserving low system complexity and system cost. By combining serial sectioning and real-time staining techniques, TRUST has achieved automated 3D imaging with excellent staining uniformity and time efficiency. Its fast, robust, and high-content multicolor 3D imaging capability has been demonstrated by imaging all vital organs and two embryos. TRUST also demonstrates potential in the field of 3D histopathology, specifically for acquiring high-quality histological images of whole organs. Traditional methods in 3D histopathology are known for their time-consuming and laborious nature, often taking several weeks to complete. However, by integrating TRUST with deep learning, we have developed HistoTRUST, a pipeline that can rapidly and automatically generate virtual hematoxylin and eosin (H&E) stained histological images of whole organs with subcellular resolution. With HistoTRUST, we have successfully produced 3D histology images of all vital organs within 1–3 days. An inherent drawback of TRUST or HistoTRUST is its limited axial resolution, determined solely by the penetration depth of UV light. To this end, we have introduced a new 3D imaging system known as HiLoTRUST by integrating the TRUST system with the structure illumination method known as HiLo microscopy. HiLoTRUST presents improved

In this work, a multiwavelength autofluorescence virtual instant stain (MAVIS) workflow is proposed to provide a multiple virtual staining solution to facilitate clinical diagnosis. Multiple ultraviolet excitation and visible emission wavelengths are used to highlight different biomolecules while a weakly supervised algorithm provides a robust and accurate virtual staining with adjacent tissue slices. The result of MAVIS with three histochemical stains on human tissue slices achieves a multi-scale structural similarity index measure > 0.6, demonstrating the potential of multiple virtual staining as a rapid and low-cost alternative to the current histological workflow.

High-resolution and label-free histological imaging modalities provide cell nuclear contrast that is analogous to standard hematoxylin and eosin (H&E) histological staining. Recently, a rapid and slide-free imaging technique, termed computational high-throughput autofluorescence microscopy by pattern illumination (CHAMP), has been developed to image thick and unprocessed tissues with subcellular resolution. Here, we propose a fast and low-cost light-emitting diode (LED) based CHAMP imaging system assisted by deep learning. The low-resolution widefield LED images can be translated into high-resolution LED-CHAMP images that highly resemble Laser-CHAMP images by enhanced super-resolution generative adversarial networks (ESRGAN). Moreover, LED-CHAMP images can be further translated into virtual H&E-stained images comparable to standard H&E histology by virtual staining models. The versatility of LEDCHAMP is experimentally demonstrated using mouse brain thin slices and thick sections, which takes only five minutes for imaging tissue surface area with 10 × 10 mm2. The promising LED-CHAMP workflow enables fast, low-cost, and comparable image quality for intraoperative assessment.

Lung cancer is one of the leading causes of cancer mortality worldwide, with an estimated 2.2 million new cancer cases and 1.8 million deaths in 2020. Adenocarcinoma is the most common type of non-small cell lung cancer (NSCLC), which is usually developed with a mixture of histologic subtypes. Surgery to remove the affected tissue or tumor is the most curative treatment option for the early-stage NSCLC currently. The clinical diagnosis of NSCLC based on pathological analysis of formalin-fixed and paraffin-embedded (FFPE) tissues is laborious and time-consuming, failing to guide surgeons intraoperatively. Although frozen section can serve as a rapid alternative to FFPE histology, it still requires a turnaround time of 20–30 minutes during surgery. Besides, the diagnostic accuracy of the frozen section could be affected due to the tissue freezing artifacts and inadequate sampling of resection margins. Here, we propose a rapid histological imaging method, termed microscopy with ultraviolet single-plane illumination (MUSI), which enables label-free and non-destructive imaging of freshly excised and unprocessed tissues. The MUSI system allows the surgical specimens with large irregular surfaces to be scanned in a label-free manner at a speed of 0.65 mm2/s with a subcellular resolution, showing great potential as an assistive imaging platform that can provide immediate feedback to surgeons and pathologists for intraoperative decision-making. We demonstrate that MUSI can differentiate between different subtypes of human lung adenocarcinomas, revealing diagnostically important features that are comparable to the gold standard FFPE histology, holding great promise to revolutionize the current practice of surgical pathology.

Ultraviolet photoacoustic microscopy (UV-PAM) has been investigated to provide label-free and registration-free volumetric histological images for whole organs, offering new insights into complex biological organs. However, because of the high UV absorption of lipids and pigments in tissue, UV-PAM suffers from low image contrast and shallow image depth, hindering its capability for revealing various microstructures in organs. To improve the UV-PAM imaging contrast and imaging depth, here we propose to implement a state-of-the-art optical clearing technique, CUBIC (clear, unobstructed brain/body imaging cocktails and computational analysis), to wash out the lipids and pigments from tissues. Our results show that the UV-PAM imaging contrast and quality can be significantly improved after tissue clearing. With the cleared tissue, multilayers of cell nuclei can also be extracted from time-resolved PA signals. Tissue clearing-enhanced UV-PAM can provide fine details for organ imaging. © 2021 The Authors

Rapid and high-resolution histological imaging with minimal tissue preparation has long been a challenging and yet captivating medical pursuit. Here, the authors propose a promising and transformative histological imaging method, termed computational high-throughput autofluorescence microscopy by pattern illumination (CHAMP). With the assistance of computational microscopy, CHAMP enables high-throughput and label-free imaging of thick and unprocessed tissues with large surface irregularity at an acquisition speed of 10 mm2/10 s with 1.1-µm lateral resolution. Moreover, the CHAMP image can be transformed into a virtually stained histological image (Deep-CHAMP) through unsupervised learning within 15 s, where significant cellular features are quantitatively extracted with high accuracy. The versatility of CHAMP is experimentally demonstrated using mouse brain/kidney and human lung tissues prepared with various clinical protocols, which enables a rapid and accurate intraoperative/postoperative pathological examination without tissue processing or staining, demonstrating its great potential as an assistive imaging platform for surgeons and pathologists to provide optimal adjuvant treatment.

Rapid and non-destructive histological imaging with minimal tissue preparation is challenging and yet appealing medical pursuit. Recent advances in optical imaging technologies enable rapid and high-resolution imaging of fresh resection tissues, showing great potential to streamline the current clinical practice in surgical margin assessment. In addition, deep-learning-assisted virtual staining holds great promise to facilitate clinical translations of these imaging methods with training-free image interpretation by pathologists. Here, we review different imaging platforms which can provide immediate on-site feedback to surgeons and pathologists for intraoperative decision-making.

We propose a rapid and label-free three-dimensional imaging technique to analyze paraffin-embedded whole organs without tissue staining or clearing. Various anatomical structures are revealed at subcellular resolution, facilitating comprehensive and volumetric cellular histopathological analysis.

Histopathology based on formalin-fixed and paraffin-embedded tissues has long been the gold standard for surgical margin assessment (SMA). However, routine pathological practice is lengthy and laborious, failing to guide surgeons intraoperatively. In this report, we propose a practical and low-cost histological imaging method with wide-field optical-sectioning microscopy (i.e., High-and-Low-frequency (HiLo) microscopy). HiLo can achieve rapid and non-destructive imaging of freshly-excised tissues at an extremely high acquisition speed of 5 cm2/min with a spatial resolution of 1.3 µm (lateral) and 5.8 µm (axial), showing great potential as an SMA tool that can provide immediate feedback to surgeons and pathologists for intraoperative decision-making. We demonstrate that HiLo enables rapid extraction of diagnostic features for different subtypes of human lung adenocarcinoma and hepatocellular carcinoma, producing surface images of rough specimens with large field-of-views and cellular features that are comparable to the clinical standard. Our results show promising clinical translations of HiLo microscopy to improve the current standard of care.

Core needle biopsy, in which a surgeon uses a biopsy needle to extract a small fragment of a suspected lesion, is a vital procedure for brain cancer diagnosis and staging tumors. However, the current practice of brain biopsy surgery is inaccurate and risk-prone; it is associated with a negative biopsy rate which leads to repeating the procedure, as well as the risks of rupturing blood vessels during the surgery and causing post-surgical complications. The inaccuracy and safety concerns of the biopsy procedure necessitate a system that can differentiate various types of brain tissue during the surgery so that surgeons can identify the types of tissue in the vicinity of the needle tip before sampling the biopsy. In this project, we explored the possibility of integrating Raman spectroscopy (RS) and diffuse reflectance spectroscopy (DRS) fiber optics system into a clinical biopsy instrument to collect the signals that pertain to light-tissue interaction. The backscattered optical spectra are acquired, processed, analyzed, and classified by a supervised machine learning algorithm in real-time. A guided user interface is written to control the hardware system and output the result of tissue classification on screen simultaneously as the biopsy probe collects data. To this end, we aim to provide intraoperative guidance to surgeons through verifying if the sampling tool is properly located within the lesion, while keeping minimal alteration to the current biopsy practice.

Objective: A sub-lobar resection is an attractive option for selected early-stage lung cancer patients. The frozen section is the current intraoperative technique to check if the lesion is lung cancer and whether the resection margin is adequate. However, the associated long waiting time unnecessarily prolongs the time for patients under anesthesia. This study introduces a novel and rapid intraoperative imaging technique (< 5 minutes, with a sample size ~1 × 1 cm2) for screening the histologic tumor type and checking the resection margin distance for conserving surgery. Methods: An autofluorescence (AF) based microscopy is introduced for imaging the surface of thick human lung cancer specimens without the need for physical thin-slice sectioning through a frozen section. Complementary with a color transformation assisted by deep learning, virtually stained hematoxylin and eosin (H&E) images can be generated for pathological evaluation. A translational study is conducted utilizing 10 lung cancer samples, including adenocarcinoma and large-cell carcinoma, from 10 patients. After acquiring the AF images of the thick specimen, the sample was sent for paraffin embedding, and its adjacent thin H&E slices were obtained for comparison. Results: To demonstrate the histologic information provided by the AF and virtual H&E image is highly similar, a set of thin tissue images of large cell carcinoma with similar histologic patterns is presented below. An average multi-scale structural similarity index measure of 0.645 was obtained between the virtually stained H&E images and the H&E images from the 10 thin tissue samples. Similar histologic information can also be identified in the images generated on a thick sample of large cell carcinoma, demonstrating the potential of this method as a slide-free intraoperative histologic tool. Conclusions: This study proposes an alternative to the frozen section for intraoperative tumor type evaluation. Without the need for phy

Histological image with hematoxylin and eosin (H&E) staining is essential for histopathological diagnosis and has been the gold standard of tumor diagnosis. However, the acquiring of H&E stained images takes around one week in the hospital due to the involved lengthy and laborious steps such as tissue preprocessing, sectioning, and staining. In this paper, I presented two novel deep learning assisted imaging methods that can shorten the acquisition time of H&E stained images from one week to several minutes, greatly increasing the efficacy of histopathological diagnosis. The first imaging method combines deep learning with ultraviolet photoacoustic microscopy (UV-PAM), which can rapidly get the H&E digital staining histological image of unprocessed fresh mouse brain within 15 minutes. The second method integrates deep learning with computational autofluorescence microscopy (CHAMP), which further reduces the time to 1 minute. In addition to the 2D histological imaging, by combing deep learning with a vibratome-assisted 3D ultraviolet imaging system, I further developed a pipeline for the rapid and fully automatic acquiring of H&E stained histopathological images of a whole organ. With the designed method, we rapidly get the 3D H&E histological images of versatile organs, such as brain, liver, kidney, heart, lung, heart, spleen with little human involvement. The proposed method might promote the routine use of 3D histology in research or clinic, as it greatly shortens the acquisition time of 3D H&E histological images from one month to ~3 days. Lastly, I summarize the advances that our three deep learning assisted histological imaging methods have brought to conventional 2D and 3D histology. Some improvements and new perspectives are also discussed.

Virtual histological staining technique with a label-free auto-fluorescence image as an input is a challenging scientific pursuit to visualize complicated biological structures with distinct features. Recently, most of the related methods follow the two-side image translation architecture to get rid of paired data restriction, which is necessary for unprocessed and thick tissue virtual histological staining style transformation. However, the associated cycle consistency loss will inevitably lead to huge calculation consumption and cannot address the problem of incorrect translation among intracellular and extracellular components, which we termed as incorrect staining. In this paper, we propose a novel and computational-efficient one-side image translation framework to transfer unstained auto-fluorescence images into virtual hematoxylin- and eosin-stained counterparts for both thin and thick human samples. To address the incorrect nuclear staining issue, we design a region-classification loss to incorporate partial supervision information. Experimental data on both thin and thick human lung samples are used to demonstrate that our method is computationally efficient while achieving a comparable transformation performance over the two-side framework. © 2022 IEEE.

Optical-resolution photoacoustic microscopy (OR-PAM) has been widely used for imaging blood vessel and oxygen saturation of hemoglobin (sO2), providing high-resolution functional images of living animals in vivo. However, most of them require one or multiple bulky and costly pulsed lasers, hindering their applicability in preclinical and clinical settings. In this paper, we demonstrate a reflection-mode low-cost high-resolution OR-PAM system by using two cost-effective and compact laser diodes (LDs), achieving microvasculature and sO2 imaging with a high lateral resolution of ∼6 µm. The cost of the excitation sources has dramatically reduced by ∼20–40 times compared to that of the pulsed lasers used in state-of-the-art OR-PAM systems. A blood phantom study was performed to show a determination coefficient R2 of 0.96 in linear regression analysis. Experimental results of in vivo mouse ear imaging show that the proposed dual-wavelength LD-based PAM system can provide high-resolution functional images at a low cost.

Surgical margin analysis (SMA), an essential procedure to confirm the complete excision of cancerous tissue in tumor resection surgery, requires intraoperative diagnostic tools to avoid repeated surgeries due to a positive surgical margin. Recently, by taking the advantage of the high intrinsic optical absorption of DNA/ RNA at 266 nm wavelength, ultraviolet photoacoustic microscopy (UV-PAM) has been developed to provide high-resolution histological images without labeling, showing great promise as an intraoperative tool for SMA. To enable the development of UV-PAM for SMA, here, a high-speed and open-top UV-PAM system is presented, which can be operated similarly to conventional optical microscopies. The UV-PAM system provides a high lateral resolution of 1.2 µm, and a high imaging speed of 55 kHz A-line rate with one-axis galvanometer mirror scanning. Moreover, to ensure UV-PAM images can be easily interpreted by pathologists without additional training, the original grayscale UV-PAM images are virtually stained by a deep-learning algorithm to mimic the standard hematoxylin-and eosin-stained images, enabling training-free histological analysis. Mouse brain slice imaging is performed to demonstrate the high performance of the open-top UV-PAM system, illustrating its great potential for SMA applications.

As a key component in a histological examination, histological images play a pivotal role in both tissue investigation and disease identification. To meet the gradually expanding demands of histological images utilization, technology and approaches from other disciplines have been widely applied to optimize the histological technique, especially in the aspects of microscope examination and staining. In this thesis, two projects on this topic are presented. The first project investigated and determined a tissue optical clearing method named CUBIC as the optimized approach to enhancing the performance of a novel imaging technique, ultraviolet photoacoustic microscopy. The outstanding clearing capacity provided by CUBIC was able to improve the imaging modality in terms of image contrast and image depth. Apart from the investigation on microscope examination, the second part of this manuscript focus on optimizing the staining protocol. Considering the current protocol involves a series of labour-intensive operations when a single slice is required to present in multiple stain colours, the second project introduced a novel unsupervised deep learning model for multiplexed virtual staining realization. Coined as StarcleGAN, the approach can synthesize high-quality staining images of single tissue slices under different staining methods.

Photoacoustic tomography (PAT) is a promising imaging technique which utilizes ultrasonic waves induced by pulse light energy absorbed by intrinsic biomolecules, such as DNA/RNA and hemoglobin, to reconstruct images. Taking the advantage of endogenous optical absorption contrast, PAT can provide label-free structural, histological, functional, and metabolic images for clinical applications and basic studies. In this thesis, we propose several techniques to improve the performance of PAT and promote its applications in biomedical and point-of-care fields. With the high optical absorption of DNA/RNA at 266 nm, ultraviolet photoacoustic microscopy (UV-PAM) has been developed for cellular imaging, providing histological information for revealing organ structures and disease diagnosis. However, the cellular contrast might be low because of the high UV absorption of lipids and pigments into tissues, affecting the microstructure analysis and degrading the diagnostic accuracy. To improve the cellular contrast of UV-PAM, we first develop a tissue clearing-enhanced UV-PAM to remove lipids and pigments, reducing the background signals. The image contrast has been significantly improved and multilayers of cell nuclei can be obtained after tissue clearing. Then, we further develop a dual-modality imaging system with UV-PA and auto-fluorescence microscopy (uvPA-AFM) to achieve cellular contrast improvement without any tissue processing, promoting the clinical applications of UV-PAM in histological examination of various tissues. High-quality histological images can even be obtained by our system under low excitation energy and high pulse-to-pulse energy fluctuation.

Eukaryotic translation is important, complex and tightly regulated process. In eukaryotes, the place for transcription to take place is nucleus, and the place for translation to take place is cytoplasm. The eukaryotic transcription and eukaryotic translation are not synchronized, gene is first transcribed into mRNA in nucleus then mRNA translated into amino acid chains in cytoplasm. In the transcription process, RNA is transcribed from DNA template and maturated by addition of 5’cap and polyadenylation tailing at 3’ terminus. In the normal maturation step, canonical polyadenylate polymerase (PAP) generates polyA tail at 3’terminus of mRNA. Non canonical PAP can decorate mRNA tailing to increase the diversity. RNA tailing takes a lead in regulating translation process of mRNA and prolonging mRNA half-life. Previous studies have been discovered that guanylation at mRNA tailling can shield mRNA from rapid decay. The reason why we start this project is to discover the mechanism of guanylation on mixed tail of mRNA. In this project, guanylation at mixed tailing decrease the translation efficiency and half-life of mRNA. Guanylation mechanism can be hijacked when virus infected. Therefore, we proposed that cellular self-defense might be the explanation. For the reason of protection, RNA protein kinase (PKR) is activated for apoptosis. Before the synthetic mRNAs enter cell to affect translation process, they are already discriminated by the cell.

As one of the most ordered clinical tests, hematological analysis has a pivotal role in screening and diagnosing blood conditions and diseases. The evaluation of the blood cell enumeration and morphology through manual microscopic examination is required for pathologists to confirm the peripheral blood smear findings. As a highly informative hematological tool, peripheral blood smear analysis involves highly trained personnel, and expensive equipment, with time-consuming and labor-intensive chemical staining procedures. In this thesis, a smartphone-based autofluorescence microscopy (Smart-AM) system is demonstrated for imaging blood smears without labels at sub-cellular resolution. The system is created by adding a reversed camera lens module which is well-designed to impede various aberrations and is applicable to work with a cell phone, providing high-quality and label-free blood smear images with a maximum field of view (FOV) of 2.58 mm × 1.94 mm. Apart from the imaging system, this thesis also integrates deep-learning algorithms to Smart-AM (DeepSmart-AM) for automatic detection of different blood cell types and virtual Giemsa staining realization.

In PhoMedics Limited, the team is commercializing the technology of CHAMP Microscope to revolutionize the way cancer cells are being detected during cancer surgery. Over 20% of all cancer patients taking cancer surgery as a treatment would fail to have a perfect cut and need to undergo a second surgery within a short time. The reason behind this is surgeons lack a tool to know whether all cancer cells have been successfully removed from the patient’s body or not during the cancer surgery. Here comes the CHAMP microscope. It is a fast, reliable cellular imaging tool that can provide medical professionals with rapid cellular imaging to guide tumour excision during the operation and potentially eliminates second cancer surgery. This thesis introduces the background on current cancer epidemiology and the pain point of cancer surgery during Chapter 1 to Chapter 2. Then, in Chapter 3, a literature review on the current cellular imaging technologies is introduced. In Chapter 4, details of the technology –– CHAMP microscope are described, including the technology principle of the hardware and software. Also, preliminary data from the CHAMP microscope is shown, and a competitive analysis is conducted among our potential competitors. In Chapter 5, the CHAMP microscope is transformed from a concept of technology to a product to better serve the users’ human value. Under the guidance of Luisa Mok, Oscar and I applied the design methodology to gradually develop the CHAMP microscope’s design concepts. Chapter 6 is the business plan of PhoMedics, which reveals the company development history and the future business plan. Also, in Chapter 6, the potential market size of the CHAMP microscope is researched and projected.

Histopathology based on formalin-fixed and paraffin-embedded tissues has long been the gold standard for surgical margin assessment (SMA). However, routine pathological practice is lengthy and laborious, failing to guide surgeons intraoperatively. In this report, we propose a practical and low-cost histological imaging method with wide-field optical-sectioning microscopy (i.e., High-and-Low-frequency (HiLo) microscopy). HiLo can achieve rapid and non-destructive imaging of freshly-excised tissues at an extremely high acquisition speed of 5 cm2/min with a spatial resolution of 1.3 µm (lateral) and 5.8 µm (axial), showing great potential as an SMA tool that can provide immediate feedback to surgeons and pathologists for intraoperative decision-making. We demonstrate that HiLo enables rapid extraction of diagnostic features for different subtypes of human lung adenocarcinoma and hepatocellular carcinoma, producing surface images of rough specimens with large field-of-views and cellular features that are comparable to the clinical standard. Our results show promising clinical translations of HiLo microscopy to improve the current standard of care.

Histopathological examination of tissue sections is the gold standard for disease diagnosis. However, the conventional histopathology workflow requires lengthy and laborious sample preparation steps including tissue fixation, paraffin embedding and microtome sectioning to obtain thin tissue slices. The slices with hematoxylin and eosin (H&E) staining will be imaged under brightfield microscope to obtain histological images. The whole procedure will cause about a one-week delay to generate an accurate diagnostic report. Microscopy with ultraviolet surface excitation (MUSE), a rapid and slide-free imaging technique, has been developed to image unprocessed fresh tissues with specific molecular contrast. The MUSE image of a typical brain biopsy (5 mm × 5 mm) with subcellular resolution can be acquired in 5 minutes under simple procedures including hand cutting, Hoechst staining and MUSE imaging. To further assist pathologists for easy adaptation on histological image, we apply the deep learning model to translate a 5 mm × 5 mm MUSE image into a 5 mm × 5 mm Deep-MUSE image that highly resemble H&E staining in 40 seconds. The unsupervised generative adversarial network has been utilized to eliminate the need for rigidly paired data training. The workflow takes ~ 6 minutes in total to obtain a 5 mm × 5 mm Deep-MUSE image. To demonstrate the proposed Deep-MUSE workflow, we imaged different mouse brain samples and fixed mouse liver sections. The imaging results show the quality of Deep MUSE images is comparable to the standard H&E-stained images. The novel Deep-MUSE workflow is promising to simplify the standard histopathology workflow and provide high quality histological images intraoperatively.

Cancer is one of the biggest threats to human health in our lifetime. The disease affects millions of people worldwide, leaving no one unaffected. In recent years, significant progress has been made in developing new screening and treatment methods, leading to the early discovery of cancer. Detection at an early stage creates a possibility to remove the cancerous tumor via surgery. However, a major problem occurring during cancer surgery is the massive number of patients having more than one surgery due to positive margins after tumor removal. The underlying reason is the unsatisfactory gold standard method that is currently used to conduct intraoperative margin assessment. It is time-consuming due to lengthy preparation procedures and generates inaccurate results. To tackle this issue, PhoMedics are in the process of commercializing a cutting-edge imaging technology coined CHAMP - Computational High-throughput Autofluorescence Microscopy with Pattern Illumination, which has the potential to revolutionize the standard of care. PhoMedics solution is a combination of hardware and software. The hardware generates high-resolution subcellular images of the tissue cells which are then virtually stained using our deep-learning software, imitating the golden standard images that medical professionals are used to interpreting. CHAMP can complete the whole procedure within a few minutes and the images are seamlessly sent to a pathologist for instant diagnosis. Such technology can help eliminate the need of second surgeries, thus saving tons of clinical resources for the healthcare system and easing the physical and mental burden among cancer patients. This thesis explains how such a revolutionizing technology can be taken to the market and commercialized.

Biomedical imaging plays a pivotal role in oncological prognosis and guided treatments accounted by the ability to visualize anatomical structures such as lesions, vasculatures, enabling non-invasive real-time monitoring and provides necessary information for clinical assessments without any destruction and perturbation of patients’ tissue. To further classify the stage and aggressiveness of cancers, techniques such as immunohistochemistry (IHC) and target-specific contrast agents are developed to target and amplify the imaging signals for cancer biomarkers. Breast cancer is one of the most common cancer worldwide, with nearly 12% of women in the United States developing invasive breast cancer in their lifetime, and as better diagnostic tools are available, clinicians can better stratify patients and provide precision treatments for a noticeably better outcome. Therefore, it is essential to have ways to probe and gather both anatomical and molecular information for accurate diagnosis. Herein, a novel protein-based exogenous contrast agent for photoacoustic imaging is proposed. The contrast agent capitalizes on the improved photostability and solubility of the far-red absorbing chlorophylls with the use of water-soluble chlorophyll-binding protein (WSCP). The contrast agent has shown promising properties for deep tissue in vivo photoacoustic imaging (PAI) and sheds light on the possibility of being used as a photosensitizer for photodynamic therapy (PDT).

Optical-based sensing approaches have long been an indispensable way to detect molecules in biological tissues for various biomedical research and applications. The advancement in optical microscopy is one of the main drivers for discoveries and innovations in both life science and biomedical imaging. However, the shallow imaging depth due to the use of ballistic photons fundamentally limits optical imaging approaches' translational potential to a clinical setting. Photoacoustic (PA) tomography (PAT) is a rapidly growing hybrid imaging modality that is capable of acoustically detecting optical contrast. PAT uniquely enjoys high-resolution deep-tissue imaging owing to the utilization of diffused photons. The exploration of endogenous contrast agents and the development of exogenous contrast agents further improve the molecular specificity for PAT. PAT's versatile design and non-invasive nature have proven its great potential as a biomedical imaging tool for a multitude of biomedical applications. In this review, representative endogenous and exogenous PA contrast agents will be introduced alongside common PAT system configurations, including the latest advances of all-optical acoustic sensing techniques.

We proposed a high-throughput label-free and slide-free imaging technique by ultraviolet-excited autofluorescence, assisted by speckle illumination and a deep learning algorithm, to generate subcellular-resolution histology-like images of unprocessed tissues, simplifying the workflow of standard-of-care histopathology. © OSA 2020 © 2020 The Author(s)

By implementing microscopy with ultraviolet surface excitation using speckle illumination, we have improved the spatial resolution by 1.5 times and preserved the long depth of focus, achieving sharp imaging of thick and rough specimens. © 2020 The Author(s)

Mid-infrared (MIR) microscopy provides rich chemical and structural information about biological samples, without staining. Conventionally, the long MIR wavelength severely limits the lateral resolution owing to optical diffraction; moreover, the strong MIR absorption of water ubiquitous in fresh biological samples results in high background and low contrast. To overcome these limitations, we propose a method that employs photoacoustic detection highly localized with a pulsed ultraviolet laser on the basis of the Grüneisen relaxation effect. For cultured cells, our method achieves water-background suppressed MIR imaging of lipids and proteins at ultraviolet resolution, at least an order of magnitude finer than the MIR diffraction limits. Label-free histology using this method is also demonstrated in thick brain slices. Our approach provides convenient high-resolution and high-contrast MIR imaging, which can benefit the diagnosis of fresh biological samples. © 2019, The Author(s), under exclusive licence to Springer Nature Limited.

Acquiring ultrafast and high spectral resolution optical images is key to measure transient physical or chemical processes, such as photon propagation, plasma dynamics, and femtosecond chemical reactions. At a trillion Hz frame rate, most ultrafast imaging modalities can acquire only a limited number of frames. Here, we present a compressed ultrafast spectral-temporal (CUST) photographic technique, enabling both an ultrahigh frame rate of 3.85 trillion Hz and a large frame number. We demonstrate that CUST photography records 60 frames, enabling precisely recording light propagation, reflection, and self-focusing in nonlinear media over 30 ps. CUST photography has the potential to further increase the frame number beyond hundreds of frames. Using spectral-temporal coupling, CUST photography can record multiple frames with a subnanometer spectral resolution with a single laser exposure, enabling ultrafast spectral imaging. CUST photography with high frame rate, high spectral resolution, and high frame number in a single modality offer a new tool for observing many transient phenomena with high temporal complexity and high spectral precision. © 2019 American Physical Society.

Photoacoustic computed tomography (PACT), a fast-developing modality for deep tissue imaging, images the spatial distribution of optical absorption. PACT usually treats the absorption coefficient as a scalar. However, the absorption coefficients of many biological tissues exhibit an anisotropic property, known as dichroism or diattenuation, which depends on molecular conformation and structural alignment. Here, we present a novel imaging method called dichroism- sensitive PACT (DS-PACT), which measures both the amplitude of tissue’s dichroism and the orientation of the optic axis of uniaxial dichroic tissue. By modulating the polarization of linearly polarized light and measuring the alternating signals through lock-in detection, DS-PACT can boost dichroic signals from biological tissues. To validate the proposed approach, we experimentally demonstrated the performance of DS-PACT by imaging plastic polarizers and ex vivo bovine tendons deep inside scattering media. We successfully detected the orientation of the optic axis of uniaxial dichroic materials, even at a depth of 4.5 transport mean free paths. We anticipate that the proposed method will extend the capability of PACT to imaging tissue absorption anisotropy. © 2018 Optical Society of America.

Conventional photoacoustic computed tomography (PACT) images the spatial distribution of optical absorption, which is approximated as an isotropic optical property. The optical absorption of many biological tissues, however, is anisotropic. This anisotropy, known as dichroism or diattenuation, encodes rich information about molecular conformation and structural alignment. Here we report a novel imaging method called dichroism-sensitive PACT (DS-PACT). Using a lock-detection strategy, our method can measure the amplitude of tissue’s dichroism and the orientation of the optic axis of uniaxial dichroic tissue, even at a depth of 3.25 transport mean free paths. We experimentally demonstrated DS-PACT by imaging plastic polarizers and ex vivo bovine tendons deep inside scattering media. Our method extends the functionality of PACT to include a new capability, imaging tissue absorption anisotropy.