BACKGROUND:Both in clinical practice and translational research, cell differentiation of leukocytes provides important diagnostic information and insights into pathophysiological mechanisms. The current gold-standard method for bronchoalveolar lavage fluid (BALF) analysis involves histochemical staining of cytospins, followed by manual morphological quantification. This approach however is labor-intensive, time-consuming, and highly operator-dependent, limiting its efficiency and throughput. This study proposes a deep learning framework for rapid, automated 3D leukocyte differentiation using label-free higher harmonic generation microscopy (HHGM). METHODS:3D leukocyte imaging was performed with label-free HHGM in a few minutes. Two deep learning models, ResNet 3D-50 and Vision Transformer (ViT) 3D, were trained, validated and tested for leucocyte differentiation on both BALF and blood fraction samples from 16 interstitial lung disease (ILDs) and 19 acute respiratory distress syndrome (ARDS) patients. Deep-learning model-prediction and cytospin analysis were performed by separate investigators. The results were compared using Bland-Altman analysis. RESULTS:The proposed framework achieved accuracies above 86% for BALF and above 96% for blood samples under five-fold cross-validation. The approach shows close agreement with gold-standard cytological analyses, with mean differences of <5% across leukocyte subpopulations. CONCLUSIONS:By integrating the label-free imaging capabilities of HHGM with deep learning, this study established a fast, accurate and high-throughput leukocyte differentiation in fresh BALF and blood samples. By significantly improving efficiency and reproducibility, this technology has the potential to transform clinical workflows and advance precision medicine.

Standard hematoxylin and eosin (H&E) histology remains the cornerstone for tumor diagnosis but is unavailable intraoperatively. The frozen section analysis also requires ∼30 min. Here, we present a compact label-free third harmonic generation (THG) microscope combined with a large-scale model Cell-SAM for rapid intraoperative imaging of CNS tumors. We provide direct mechanistic evidence that condensed DNA aggregates generate the nuclear THG contrast in human gliomas. We further validate chromatin condensation as a unique label-free biomarker across CNS tumors and lymphomas, revealing diagnostic features comparable to H&E histology. To handle the feature complexity arising from myelin and chromatin-derived contrast, we introduce Cell-SAM for segmenting key elements in THG images of CNS tumors and normal brain, with human-level accuracy (>98%). Statistical analysis of cellular and vascular densities further demonstrates the potential of THG for tumor typing. Together, compact THG imaging and Cell-SAM establish a mechanism-based and data-driven framework for real-time intraoperative pathology.

Myelin damage is a hallmark of several neurological disorders, but how it occurs remains to be fully understood. In this study, we found that early damage in zebrafish and rodent demyelination models is characterized by myelin swelling. We show, through live imaging, that myelin swelling does not always lead to myelin loss and that swellings can sometimes resolve, allowing sheaths to remodel. Increased neuronal activity during early demyelination exacerbates myelin damage, whereas reducing neuronal activity mitigates myelin swelling in both zebrafish and mice. In human multiple sclerosis tissue, myelin swelling is also dynamic and is prominent around active lesions. Our data indicate that myelin swelling is a conserved feature of demyelination and that damage to myelin sheaths can resolve, opening opportunities for targeting human disease.

BackgroundCurrent methods to study patient-specific disease processes rely on a variety of ex vivo techniques, like static histological analysis or tissue cultures with static endpoints, that provide limited insight into how the tissue microenvironment behaves in real time. Yet, dynamic features, such as immune cell infiltration and extracellular matrix remodeling, are crucial predictors of therapy response, be it in cancer or fibrotic diseases. At present, no label-free and clinically viable assay exists that can capture these behaviors in live human tissue.MethodsWe propose an innovative platform that provides metrics on cell density, cell dynamics and microstructure of the extracellular matrix for the characterization of treatment response. The platform integrates fully label-free, time-lapse 3D higher harmonic generation microscopy to visualize both cells and the extracellular matrix, and a stage-top incubator in which tissue slices are cultured during time-lapse measurements.ResultsHere, we demonstrate the viability of the tissue slices during imaging and the ability to resolve endogenous immune cell and seeded monocyte dynamics, in the context of healthy alveolar structures. Time-lapse dynamics were recorded in 3D for up to 59 hours during a total culture period of 90 hours, and the migration of endogenous smaller cells and seeded monocytes was tracked and analyzed.ConclusionsThis platform is relatively simple to operate and has potential for studies on drug development and patient-specific treatment response prediction in the context of human lung disease.

Background/Objectives: Pediatric gonadal tumors are rare tumors, and germ cell tumors (GCTs) are the most common subgroup. GCTs are heterogeneous tumors and have different subtypes that can be either benign or malignant. Therefore, extensive sampling of the resected tumor is required to obtain an accurate diagnosis. Higher harmonic generation microscopy (HHGM) is an innovative imaging technique that enables rapid visualization of fresh tissue without the need for preparation or staining. This makes it particularly valuable for sample selection, as the tissue can be reused for further analysis. This study aims to evaluate the accuracy of pathologists detecting normal gonadal tissue, germ cell tumors, and other pediatric gonadal tumors in HHGM images. Methods: Twenty-eight samples of twenty-two patients with a germ cell tumor or other gonadal tumor were imaged with the HHG microscope and subsequently processed for histology. Overview images of the samples were made in 10 s per mm2, and high-quality images in 1 min per mm2. A multi-observer study was performed with five expert pathologists. Results: Pathologists were able to differentiate between tumor and non-tumor tissue with an accuracy of 75% (21/28) on the HHGM images versus an accuracy of 89% (25/28) on the corresponding histology. Discrepancies mainly concerned teratoma cases for HHGM as well as H&E, indicating that sampling errors of these heterogeneous tumors affected the outcomes of this study adversely. Conclusions: Although the sample size was limited by the rarity of these tumors, our data show that HHGM is a promising technique for the rapid assessment of pediatric gonadal tumor samples, particularly in evaluating their representativeness.

Lung cancer is one of the most prevalent and lethal cancers. To improve health outcomes while reducing health care burden, it becomes crucial to move toward early detection and cost-effective workflows. Currently, there is no method for the on-site rapid histologic feedback on biopsies taken in diagnostic, endoscopic, or surgical procedures. Higher harmonic generation (HHG) microscopy is a laser-based technique that provides images of unprocessed tissue. In this study, we report the feasibility of an HHG portable microscope in the clinical workflow in terms of acquisition time, image quality, and diagnostic accuracy in suspected pulmonary and pleural malignancy. One hundred nine biopsies of 47 patients were imaged and a biopsy overview image was provided within a median acquisition time of 6 minutes after excision. The assessment by pathologists and an artificial intelligence algorithm showed that image quality was sufficient for a malignancy or non- malignancy diagnosis in 97% of the biopsies, and 87% of the HHG images were correctly scored by the pathologists. HHG is therefore an excellent candidate to provide a rapid pathology outcome on biopsy samples, enabling immediate diagnosis and (local) treatment. (c) 2024 THE AUTHORS. Published by Elsevier Inc. on behalf of the United States & Canadian Academy of Pathology. This is an open access article under the CC BY license (http://creativecommons.org/ licenses/by/4.0/).

Photonics is transforming pathology by enabling fast, label-free tissue diagnostics. We developed Higher Harmonic Generation Microscopy (HHGM), a compact femtosecond-laser platform combining SHG, THG, and autofluorescence to image unprocessed tissue with submicron resolution. HHGM reproduces histological architecture within minutes, preserves tissue for further analysis, and enables AI-driven, real-time diagnostics across cancer types.

Osteogenesis imperfecta (OI) and Ehlers-Danlos syndrome (EDS) are inherited connective tissue disorders caused by diverse genetic defects, many of which affect collagen biosynthesis. However, the identified genetic variants do not always fully explain the clinical heterogeneity observed in patients, highlighting the need for advanced models and imaging techniques to assess collagen structure and fibroblast behavior at the microscopic level. In this study, we employed 5-week three-dimensional (3D) dermal fibroblast cultures derived from patients with haploinsufficient (HI) and dominant-negative (DN) OI, EDS, and healthy controls. Using label-free higher harmonic generation microscopy (HHGM), we visualized and quantified secreted collagen fibers and fibroblast morphology in situ. We analyzed fibroblast 3D orientation, collagen fiber diameter, collagen amount per cell, and the spatial alignment between fibroblasts and collagen fibers. HI OI fibroblasts secreted significantly less collagen than both control and EDS-derived cells, while EDS samples exhibited thinner collagen fibers compared to controls. Across all groups, collagen fiber orientation was strongly correlated with fibroblast alignment, in line with the role of fibroblasts in matrix organization. In healthy controls and HI OI samples, we observed a depth-dependent, counterclockwise rotation in fibroblast orientation from the culture bottom to the surface-a pattern that was less prominent in DN OI and EDS samples, potentially reflecting altered matrix guidance in diseased tissues. Overall, the quantity and quality of collagen, as well as fibroblast morphology and organization, were markedly altered in the OI and EDS model systems. These alterations may mirror tissue-level manifestations of the diseases, demonstrating the physiological relevance of patient-derived 3D fibroblast models for OI and EDS, as well as the power of harmonic generation microscopy in probing the cellular and extracellular consequences of disease-related gene defects in collagen or its biosynthetic pathways. Extensions of this methodological approach provide a way towards deeper understanding of tissue-level manifestations of collagen dysregulation in connective tissue disorders.

Background/Objectives: Wilms tumors, the most common pediatric renal tumors, are heterogeneous and consist of varying amounts of three components: blastema, epithelium, and stroma. Postoperative chemotherapy is tailored based on risk group classification and stage. Due to this heterogeneity, pathologists perform extensive tumor sampling to ensure accurate classification. Higher-harmonic-generation microscopy (HHGM) is an innovative imaging technique that enables rapid visualization of fresh tissue without preparation or staining. This makes it particularly valuable for sample selection, as the tissue can be reused for further analysis. This study aims to evaluate the accuracy of pathologists in distinguishing normal renal tissue, abnormal renal tissue, and three types of pediatric renal tumors, Wilms tumor, renal cell carcinoma, and congenital mesoblastic nephroma, in HHGM images. Methods: Twenty-nine samples from eighteen patients with a pediatric renal tumor were imaged using an HHG microscope and subsequently processed for histological analysis. Overview images of the samples were acquired at a rate of 10 s per mm2, while high-quality images took 1 min per mm2. A multi-observer study involving ten international expert pathologists of the SIOP-RTSG was conducted. Results: Pathologists were able to differentiate between normal and abnormal tissue with 100% (29/29) accuracy and correctly identified tumor versus non-tumor tissue with 97% (28/29) accuracy. Conclusions: These results show that HHGM is a highly promising technique for the rapid assessment of pediatric renal tumor samples, particularly for evaluating sample representativeness.

Collagen is critical to the structure and function of skin tissues, with the collagen I/III ratios influencing fibrillogenesis, fiber organization, and skin mechanics. Abnormal collagen organization, such as in fibrosis or scar tissue, compromises both skin functionality and aesthetics. In this study, we employed label-free polarization resolved second harmonic generation (PSHG) microscopy to investigate collagen structure in artificial collagen matrices with various Col I/III ratios at the fibril scale ( $$\sim$$ ∼ 1 to $$3\,\upmu \hbox {m}$$ 3 μ m ) and in ex vivo human healthy and scarred skin at the fiber scale ( $$\sim 10$$ ∼ 10 to $$20\,\upmu \hbox {m}$$ 20 μ m ). Complementary third harmonic generation (THG) microscopy provided additional structural information. Our results indicate that an increasing Col I/III ratio is associated with longer fibril length, higher PSHG intensity, and a reduced effective $$\alpha$$ α -helix pitch angle of fibrils. In pure Col I, the effective $$\alpha$$ α -helix pitch angle is determined to be $$47.72^{\circ }$$ 47 . 72 ∘ . These observations indicate alterations in fibril assembly. Furthermore, although the $$\alpha$$ α -helix pitch angle of fibers in both healthy and scarred skin was approximately $$46.7^{\circ }$$ 46 . 7 ∘ , healthy skin exhibited $$24\%$$ 24 % greater variability in fiber orientation, suggesting a more randomized organization compared to scar tissue. THG imaging further revealed a higher cellular density in scar tissue, consistent with the inflammatory activity associated with wound healing. Immunohistochemical (IHC) staining using dermatansulphate and Col III-specific antibodies confirmed that the Col I/III ratio is higher in healthy skin (2.2) than in scarred skin (1.6). These findings underscore the potential of PSHG microscopy for label-free, quantitative assessment of collagen structure across multiple scales, with THG offering complementary cellular insights. This integrated approach represents a promising strategy for real-time, in vivo monitoring and automated quantification of collagen organization in clinical applications, including dermatology, burn treatment, and fibrosis monitoring.

Myelin pathology is known to play a central role in disorders such as multiple sclerosis (MS) among others. Despite this, the pathological mechanisms underlying these conditions are often difficult to unravel. Conventional techniques like immunohistochemistry or dye-based approaches, do not provide a temporal characterization of the pathophysiological aberrations responsible for myelin changes in human specimens. Here, to circumvent this curb, we present a label-free, live-cell imaging approach of myelin using recent advancements in nonlinear harmonic generation microscopy applied to physiologically viable human brain tissue from post-mortem donors. Gray and white matter brain tissue from epilepsy surgery and post-mortem donors was excised. To sustain viability of the specimens for several hours, they were subjected to either acute or organotypic slice culture protocols in artificial cerebral spinal fluid. Imaging was performed using a femtosecond pulsed 1050 nm laser to generate second harmonic generation (SHG) and third harmonic generation (THG) signals directly from myelin and axon-like structures without the need to add any labels. Experiments on acute human brain slices and post-mortem human slice cultures reveal that myelin, along with lipid bodies, are the prime sources of THG signal. We show that tissue viability is maintained over extended periods during THG microscopy, and that prolonged THG imaging is able to detect experimentally induced subtle alterations in myelin morphology. Finally, we provide practical evidence that live-cell imaging of myelin with THG microscopy is a sensitive tool to investigate subtle changes in white matter of neurological donors. Overall, our findings support that nonlinear live-cell imaging is a suitable setup for researching myelin morphology in neurological conditions like MS.