Dataset Information

Imaging sub-diffuse optical properties of cancerous and normal skin tissue using machine learning-aided spatial frequency domain imaging.

ABSTRACT: Sub-diffuse optical properties may serve as useful cancer biomarkers, and wide-field heatmaps of these properties could aid physicians in identifying cancerous tissue. Sub-diffuse spatial frequency domain imaging (sd-SFDI) can reveal such wide-field maps, but the current time cost of experimentally validated methods for rendering these heatmaps precludes this technology from potential real-time applications. Our study renders heatmaps of sub-diffuse optical properties from experimental sd-SFDI images in real time and reports these properties for cancerous and normal skin tissue subtypes. A phase function sampling method was used to simulate sd-SFDI spectra over a wide range of optical properties. A machine learning model trained on these simulations and tested on tissue phantoms was used to render sub-diffuse optical property heatmaps from sd-SFDI images of cancerous and normal skin tissue. The model accurately rendered heatmaps from experimental sd-SFDI images in real time. In addition, heatmaps of a small number of tissue samples are presented to inform hypotheses on sub-diffuse optical property differences across skin tissue subtypes. These results bring the overall process of sd-SFDI a fundamental step closer to real-time speeds and set a foundation for future real-time medical applications of sd-SFDI such as image guided surgery.

SUBMITTER: Stier AC

PROVIDER: S-EPMC8459901 | biostudies-literature | 2021 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Imaging sub-diffuse optical properties of cancerous and normal skin tissue using machine learning-aided spatial frequency domain imaging.

Stier Andrew C AC Goth Will W Hurley Aislinn A Brown Treshayla T Feng Xu X Zhang Yao Y Lopes Fabiana C P S FCPS Sebastian Katherine R KR Ren Pengyu P Fox Matthew C MC Reichenberg Jason S JS Markey Mia K MK Tunnell James W JW

Journal of biomedical optics 20210901 9

<h4>Significance</h4>Sub-diffuse optical properties may serve as useful cancer biomarkers, and wide-field heatmaps of these properties could aid physicians in identifying cancerous tissue. Sub-diffuse spatial frequency domain imaging (sd-SFDI) can reveal such wide-field maps, but the current time cost of experimentally validated methods for rendering these heatmaps precludes this technology from potential real-time applications.<h4>Aim</h4>Our study renders heatmaps of sub-diffuse optical proper ...[more]

PMID: 34558235

Similar Datasets

Project description:Spatial frequency domain imaging (SFDI), a noncontact wide-field imaging technique using patterned illumination with multiple wavelengths, has been used to quantitatively measure structural and functional parameters of in vivo tissue. Using SFDI in a porcine model, we previously found that scattering changes in skin could potentially be used to noninvasively assess burn severity and monitor wound healing. Translating these findings to human subjects necessitates a better understanding of the variation in "baseline" human skin scattering properties across skin types and anatomical locations. Using SFDI, we aim to characterize the variation in the reduced scattering coefficient (μs') for skin across a range of pigmentation and anatomic sites (including common burn locations) for normal human subjects. These measurements are expected to characterize baseline human skin properties to inform our use of SFDI for clinical burn severity and wound healing assessments. SFDI was used to measure μs' in the visible- and near-infrared regime (471 to 851 nm) in 15 subjects at 10 anatomical locations. Subjects varied in age, gender, and Fitzpatrick skin type. For all anatomical locations, the coefficient of variation in measured μs' decreased with increasing wavelength. High intersubject variation in μs' at visible wavelengths coincided with large values of the melanin extinction coefficient at those wavelengths. At 851 nm, where intersubject variation in μs' was smallest for all anatomical locations and absorption from melanin is minimal, significant intrasubject differences in μs' were observed at the different anatomical locations. Our study is the first report of wide-field mapping of human skin scattering properties across multiple skin types and anatomical locations using SFDI. Measured μs' values varied notably between skin types at wavelengths where absorption from melanin was prominent. Additionally, μs' varied considerably across different anatomical locations at 851 nm, where the confounding effects from melanin absorption are minimized.

Project description:SignificanceAccurate early diagnosis of malignant skin lesions is critical in providing adequate and timely treatment; unfortunately, initial clinical evaluation of similar-looking benign and malignant skin lesions can result in missed diagnosis of malignant lesions and unnecessary biopsy of benign ones.AimTo develop and validate a label-free and objective image-guided strategy for the clinical evaluation of suspicious pigmented skin lesions based on multispectral autofluorescence lifetime imaging (maFLIM) dermoscopy.ApproachWe tested the hypothesis that maFLIM-derived autofluorescence global features can be used in machine-learning (ML) models to discriminate malignant from benign pigmented skin lesions. Clinical widefield maFLIM dermoscopy imaging of 41 benign and 19 malignant pigmented skin lesions from 30 patients were acquired prior to tissue biopsy sampling. Three different pools of global image-level maFLIM features were extracted: multispectral intensity, time-domain biexponential, and frequency-domain phasor features. The classification potential of each feature pool to discriminate benign versus malignant pigmented skin lesions was evaluated by training quadratic discriminant analysis (QDA) classification models and applying a leave-one-patient-out cross-validation strategy.ResultsClassification performance estimates obtained after unbiased feature selection were as follows: 68% sensitivity and 80% specificity with the phasor feature pool, 84% sensitivity, and 71% specificity with the biexponential feature pool, and 84% sensitivity and 32% specificity with the intensity feature pool. Ensemble combinations of QDA models trained with phasor and biexponential features yielded sensitivity of 84% and specificity of 90%, outperforming all other models considered.ConclusionsSimple classification ML models based on time-resolved (biexponential and phasor) autofluorescence global features extracted from maFLIM dermoscopy images have the potential to provide objective discrimination of malignant from benign pigmented lesions. ML-assisted maFLIM dermoscopy could potentially assist with the clinical evaluation of suspicious lesions and the identification of those patients benefiting the most from biopsy examination.

Project description:BackgroundPancreatic cancer is a lethal disease with a 5-year survival rate of only 1-5%. The acceleration of intraoperative histological examination would be beneficial for better management of pancreatic cancer, suggesting an improved survival. Nonlinear optical methods based on two-photon excited fluorescence (TPEF) and second harmonic generation (SHG) of intrinsic optical biomarkers show the ability to visualize the morphology of fresh tissues associated with histology, which is promising for real-time intraoperative evaluation of pancreatic cancer.Methodology/principal findingsIn order to investigate whether the nonlinear optical imaging methods have the ability to characterize pancreatic histology at cellular resolution, we studied different types of pancreatic tissues by using label-free TPEF and SHG. Compared with other routine methods for the preparation of specimens, fresh tissues without processing were found to be most suitable for nonlinear optical imaging of pancreatic tissues. The detailed morphology of the normal rat pancreas was observed and related with the standard histological images. Comparatively speaking, the preliminary images of a small number of chemical-induced pancreatic cancer tissues showed visible neoplastic differences in the morphology of cells and extracellular matrix. The subcutaneous pancreatic tumor xenografts were further observed using the nonlinear optical microscopy, showing that most cells are leucocytes at 5 days after implantation, the tumor cells begin to proliferate at 10 days after implantation, and the extracellular collagen fibers become disordered as the xenografts grow.Conclusions/significanceIn this study, nonlinear optical imaging was used to characterize the morphological details of fresh pancreatic tissues for the first time. We demonstrate that it is possible to provide real-time histological evaluation of pancreatic cancer by the nonlinear optical methods, which present an opportunity for the characterization of the progress of spontaneous pancreatic cancer and further application in a non-invasive manner.

Project description:Macrophages are dynamic immune cells that govern both normal tissue function and disease progression. However, standard methods to measure heterogeneity in macrophage function within tissues require tissue excision and fixation, which limits our understanding of diverse macrophage function in vivo. Two-photon microscopy of the endogenous metabolic co-enzymes NAD(P)H and flavin adenine dinucleotide (FAD) (metabolic autofluorescence imaging) enables dynamic imaging of mouse models in vivo. Here, we demonstrate metabolic autofluorescence imaging to assess cell-level macrophage heterogeneity in response to normal and cancerous tissue microenvironments in vivo. NAD(P)H and FAD fluorescence intensities and lifetimes were measured for both tissue-resident macrophages in mouse ear dermis and tumor-associated macrophages in pancreatic flank tumors. Metabolic and spatial organization of macrophages were determined by performing metabolic autofluorescence imaging and single macrophage segmentation in mice engineered for macrophage-specific fluorescent protein expression. Tumor-associated macrophages exhibited decreased optical redox ratio [NAD(P)H divided by FAD intensity] compared to dermal macrophages, indicating that tumor-associated macrophages are more oxidized than dermal macrophages. The mean fluorescence lifetimes of NAD(P)H and FAD were longer in dermal macrophages than in tumor-associated macrophages, which reflects changes in NAD(P)H and FAD protein-binding activities. Dermal macrophages had greater heterogeneity in optical redox ratio, NAD(P)H mean lifetime, and FAD mean lifetime compared to tumor-associated macrophages. Similarly, standard markers of macrophage phenotype (CD206 and CD86) assessed by immunofluorescence revealed greater heterogeneity in dermal macrophages compared to tumor-associated macrophages. Ultimately, metabolic autofluorescence imaging provides a novel tool to assess tissue-specific macrophage behavior and cell-level heterogeneity in vivo in animal models.

Dataset Information

Imaging sub-diffuse optical properties of cancerous and normal skin tissue using machine learning-aided spatial frequency domain imaging.

Publications

Imaging sub-diffuse optical properties of cancerous and normal skin tissue using machine learning-aided spatial frequency domain imaging.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets