Instant diagnosis of gastroscopic biopsy via deep-learned single-shot femtosecond stimulated Raman histology.
Ontology highlight
ABSTRACT:
SUBMITTER: Liu Z
PROVIDER: S-EPMC9279377 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Maximal resection of tumor while preserving the adjacent healthy tissue is particularly important for larynx surgery, hence precise and rapid intraoperative histology of laryngeal tissue is crucial for providing optimal surgical outcomes. We hypothesized that deep-learning based stimulated Raman scattering (SRS) microscopy could provide automated and accurate diagnosis of laryngeal squamous cell carcinoma on fresh, unprocessed surgical specimens without fixation, sectioning or staining. Methods: We first compared 80 pairs of adjacent frozen sections imaged with SRS and standard hematoxylin and eosin histology to evaluate their concordance. We then applied SRS imaging on fresh surgical tissues from 45 patients to reveal key diagnostic features, based on which we have constructed a deep learning based model to generate automated histologic results. 18,750 SRS fields of views were used to train and cross-validate our 34-layered residual convolutional neural network, which was used to classify 33 untrained fresh larynx surgical samples into normal and neoplasia. Furthermore, we simulated intraoperative evaluation of resection margins on totally removed larynxes. Results: We demonstrated near-perfect diagnostic concordance (Cohen's kappa, ? > 0.90) between SRS and standard histology as evaluated by three pathologists. And deep-learning based SRS correctly classified 33 independent surgical specimens with 100% accuracy. We also demonstrated that our method could identify tissue neoplasia at the simulated resection margins that appear grossly normal with naked eyes. Conclusion: Our results indicated that SRS histology integrated with deep learning algorithm provides potential for delivering rapid intraoperative diagnosis that could aid the surgical management of laryngeal cancer.
| S-EPMC6526002 | biostudies-literature
Project description:Intraoperative diagnosis is essential for providing safe and effective care during cancer surgery1. The existing workflow for intraoperative diagnosis based on hematoxylin and eosin staining of processed tissue is time, resource and labor intensive2,3. Moreover, interpretation of intraoperative histologic images is dependent on a contracting, unevenly distributed, pathology workforce4. In the present study, we report a parallel workflow that combines stimulated Raman histology (SRH)5-7, a label-free optical imaging method and deep convolutional neural networks (CNNs) to predict diagnosis at the bedside in near real-time in an automated fashion. Specifically, our CNNs, trained on over 2.5 million SRH images, predict brain tumor diagnosis in the operating room in under 150 s, an order of magnitude faster than conventional techniques (for example, 20-30 min)2. In a multicenter, prospective clinical trial (n = 278), we demonstrated that CNN-based diagnosis of SRH images was noninferior to pathologist-based interpretation of conventional histologic images (overall accuracy, 94.6% versus 93.9%). Our CNNs learned a hierarchy of recognizable histologic feature representations to classify the major histopathologic classes of brain tumors. In addition, we implemented a semantic segmentation method to identify tumor-infiltrated diagnostic regions within SRH images. These results demonstrate how intraoperative cancer diagnosis can be streamlined, creating a complementary pathway for tissue diagnosis that is independent of a traditional pathology laboratory.
| S-EPMC6960329 | biostudies-literature
Project description:Histological examination is crucial for cancer diagnosis, however, the labor-intensive sample preparation involved in the histology impedes the speed of diagnosis. Recently developed two-color stimulated Raman histology could bypass the complex tissue processing to generates result close to hematoxylin and eosin staining, which is one of the golden standards in cancer histology. Yet, the underlying chemical features are not revealed in two-color stimulated Raman histology, compromising the effectiveness of prognostic stratification. Here, we present a high-content stimulated Raman histology (HC-SRH) platform that provides both morphological and chemical information for cancer diagnosis based on un-stained breast tissues. Methods: By utilizing both hyperspectral SRS imaging in the C-H vibration window and sparsity-penalized unmixing of overlapped spectral profiles, HC-SRH enabled high-content chemical mapping of saturated lipids, unsaturated lipids, cellular protein, extracellular matrix (ECM), and water. Spectral selective sampling was further implemented to boost the speed of HC-SRH. To show the potential for clinical use, HC-SRH using a compact fiber laser-based stimulated Raman microscope was demonstrated. Harnessing the wide and rapid tuning capability of the fiber laser, both C-H and fingerprint vibration windows were accessed. Results: HC-SRH successfully mapped unsaturated lipids, cellular protein, extracellular matrix, saturated lipid, and water in breast tissue. With these five chemical maps, HC-SRH provided distinct contrast for tissue components including duct, stroma, fat cell, necrosis, and vessel. With selective spectral sampling, the speed of HC-SRH was improved by one order of magnitude. The fiber-laser-based HC-SRH produced the same image quality in the C-H window as the state-of-the-art solid laser. In the fingerprint window, nucleic acid and solid-state ester contrast was demonstrated. Conclusions: HC-SRH provides both morphological and chemical information of tissue in a label-free manner. The chemical information detected is beyond the reach of traditional hematoxylin and eosin staining and heralds the potential of HC-SRH for biomarker discovery.
| S-EPMC10879861 | biostudies-literature
Project description:BackgroundDetection of glioma recurrence remains a challenge in modern neuro-oncology. Noninvasive radiographic imaging is unable to definitively differentiate true recurrence versus pseudoprogression. Even in biopsied tissue, it can be challenging to differentiate recurrent tumor and treatment effect. We hypothesized that intraoperative stimulated Raman histology (SRH) and deep neural networks can be used to improve the intraoperative detection of glioma recurrence.MethodsWe used fiber laser-based SRH, a label-free, nonconsumptive, high-resolution microscopy method (<60 sec per 1 × 1 mm2) to image a cohort of patients (n = 35) with suspected recurrent gliomas who underwent biopsy or resection. The SRH images were then used to train a convolutional neural network (CNN) and develop an inference algorithm to detect viable recurrent glioma. Following network training, the performance of the CNN was tested for diagnostic accuracy in a retrospective cohort (n = 48).ResultsUsing patch-level CNN predictions, the inference algorithm returns a single Bernoulli distribution for the probability of tumor recurrence for each surgical specimen or patient. The external SRH validation dataset consisted of 48 patients (recurrent, 30; pseudoprogression, 18), and we achieved a diagnostic accuracy of 95.8%.ConclusionSRH with CNN-based diagnosis can be used to improve the intraoperative detection of glioma recurrence in near-real time. Our results provide insight into how optical imaging and computer vision can be combined to augment conventional diagnostic methods and improve the quality of specimen sampling at glioma recurrence.
| S-EPMC8631085 | biostudies-literature
Project description:Raman microscopy is a quantitative, label-free, and noninvasive optical imaging technique for studying inhomogeneous systems. However, the feebleness of Raman scattering significantly limits the use of Raman microscopy to low time resolutions and primarily static samples. Recent developments in narrowband stimulated Raman scattering (SRS) microscopy have significantly increased the acquisition speed of Raman based label-free imaging by a few orders of magnitude, at the expense of reduced spectroscopic information. On the basis of a spectral focusing approach, we present a fast SRS hyperspectral imaging system using chirped femtosecond lasers to achieve rapid Raman spectra acquisition while retaining the full speed and image quality of narrowband SRS imaging. We demonstrate that quantitative concentration determination of cholesterol in the presence of interfering chemical species can be achieved with sensitivity down to 4 mM. For imaging purposes, hyperspectral imaging data in the C-H stretching region is obtained within a minute. We show that mammalian cell SRS hyperspectral imaging reveals the spatially inhomogeneous distribution of saturated lipids, unsaturated lipids, cholesterol, and protein. The combination of fast spectroscopy and label-free chemical imaging will enable new applications in studying biological systems and material systems.
| S-EPMC3637845 | biostudies-literature
Project description:Nonlinear vibrational imaging of live cells and organisms is demonstrated by detecting femtosecond pulse stimulated Raman loss. Femtosecond pulse excitation produced a 12 times larger stimulated Raman loss signal than picosecond pulse excitation. The large signal allowed real-time imaging of the conversion of deuterated palmitic acid into lipid droplets inside live cells, and three-dimensional sectioning of fat storage in live C. elegans. With the majority of the excitation power contributed by the Stokes beam in the 1.0 to 1.2 ?m wavelength range, photodamage of biological samples was not observed.
| S-EPMC3124560 | biostudies-literature
Project description:Excited state vibrations are crucial for determining the photophysical and photochemical properties of molecular compounds. Stimulated Raman scattering can coherently stimulate and probe molecular vibrations with optical pulses, but it is generally restricted to ground state properties. Working under resonance conditions enables cross-section enhancement and selective excitation to a targeted electronic level but is hampered by an increased signal complexity due to the presence of overlapping spectral contributions. Here, we show how detailed information about ground and excited state vibrations can be disentangled by exploiting the relative time delay between Raman and probe pulses to control the excited state population, combined with a diagrammatic formalism to dissect the pathways concurring with the signal generation. The proposed method is then exploited to elucidate the vibrational properties of the ground and excited electronic states in the paradigmatic case of cresyl violet. We anticipate that the presented approach holds the potential for selective mapping of the reaction coordinates pertaining to transient electronic stages implied in photoactive compounds.
| S-EPMC7735730 | biostudies-literature
Project description:By augmentation of the collection optics utilized in transmission-based femtosecond stimulated Raman spectroscopy (FSRS), two novel diffuse reflectance-based femtosecond stimulated Raman spectroscopy (drFSRS) techniques were developed. These techniques were then used to collect the Raman spectra of opaque systems, those being cyclohexane-intercalated poly(tetrafluoroethylene) microbeads and ethanol in 1% intralipid solutions. The resulting drFSRS data from the cyclohexane:PTFE system show significant distortion of the depolarization ratio of the 803 cm-1 cyclohexane peak, indicating a loss of incident pump:probe polarization in a scattering environment. The drFSRS data from the ethanol in 1% intralipid solution demonstrate less signal strength but equal spectral resolution when compared to transmission-based FSRS of the same sample. The results presented in this Technical Note demonstrate the current capabilities of collecting stimulated Raman spectra of opaque systems using drFSRS.
| S-EPMC10620773 | biostudies-literature
Project description:Accurate intraoperative diagnosis is essential for providing safe and effective care during brain tumor surgery. Our standard-of-care diagnostic methods are time, resource, and labor intensive, which restricts access to optimal surgical treatments. To address these limitations, we propose an alternative workflow that combines stimulated Raman histology (SRH), a rapid optical imaging method, with deep learning-based automated interpretation of SRH images for intraoperative brain tumor diagnosis and real-time surgical decision support. Here, we present OpenSRH, the first public dataset of clinical SRH images from 300+ brain tumors patients and 1300+ unique whole slide optical images. OpenSRH contains data from the most common brain tumors diagnoses, full pathologic annotations, whole slide tumor segmentations, raw and processed optical imaging data for end-to-end model development and validation. We provide a framework for patch-based whole slide SRH classification and inference using weak (i.e. patient-level) diagnostic labels. Finally, we benchmark two computer vision tasks: multiclass histologic brain tumor classification and patch-based contrastive representation learning. We hope OpenSRH will facilitate the clinical translation of rapid optical imaging and real-time ML-based surgical decision support in order to improve the access, safety, and efficacy of cancer surgery in the era of precision medicine. Dataset access, code, and benchmarks are available at https://opensrh.mlins.org.
| S-EPMC10114931 | biostudies-literature
Project description:Accurate histopathologic diagnosis is essential for providing optimal surgical management of pediatric brain tumors. Current methods for intraoperative histology are time- and labor-intensive and often introduce artifact that limit interpretation. Stimulated Raman histology (SRH) is a novel label-free imaging technique that provides intraoperative histologic images of fresh, unprocessed surgical specimens. Here we evaluate the capacity of SRH for use in the intraoperative diagnosis of pediatric type brain tumors. SRH revealed key diagnostic features in fresh tissue specimens collected from 33 prospectively enrolled pediatric type brain tumor patients, preserving tumor cytology and histoarchitecture in all specimens. We simulated an intraoperative consultation for 25 patients with specimens imaged using both SRH and standard hematoxylin and eosin histology. SRH-based diagnoses achieved near-perfect diagnostic concordance (Cohen's kappa, κ > 0.90) and an accuracy of 92% to 96%. We then developed a quantitative histologic method using SRH images based on rapid image feature extraction. Nuclear density, tumor-associated macrophage infiltration, and nuclear morphology parameters from 3337 SRH fields of view were used to develop and validate a decision-tree machine-learning model. Using SRH image features, our model correctly classified 25 fresh pediatric type surgical specimens into normal versus lesional tissue and low-grade versus high-grade tumors with 100% accuracy. Our results provide insight into how SRH can deliver rapid diagnostic histologic data that could inform the surgical management of pediatric brain tumors.Significance: A new imaging method simplifies diagnosis and informs decision making during pediatric brain tumor surgery. Cancer Res; 78(1); 278-89. ©2017 AACR.
| S-EPMC5844703 | biostudies-literature