Leveraging mid-infrared spectroscopic imaging and deep learning for tissue subtype classification in ovarian cancer.

Mid-infrared spectroscopic imaging (MIRSI) is an emerging class of label-free techniques being leveraged for digital histopathology. Modern histopathologic identification of ovarian cancer involves tissue staining followed by morphological pattern recognition. This process is time-consuming and subjective and requires extensive expertise. This paper presents the first label-free, quantitative, and automated histological recognition of ovarian tissue subtypes using a new MIRSI technique. This optical photothermal infrared (O-PTIR) imaging technique provides a 10× enhancement in spatial resolution relative to prior instruments. It enables sub-cellular spectroscopic investigation of tissue at biochemically important fingerprint wavelengths. We demonstrate that the enhanced resolution of sub-cellular features, combined with spectroscopic information, enables reliable classification of ovarian cell subtypes achieving a classification accuracy of 0.98. Moreover, we present a statistically robust analysis from 78 patient samples with over 60 million data points. We show that sub-cellular resolution from five wavenumbers is sufficient to outperform state-of-the-art diffraction-limited techniques with up to 235 wavenumbers. We also propose two quantitative biomarkers based on the relative quantities of epithelia and stroma that exhibit efficacy in early cancer diagnosis. This paper demonstrates that combining deep learning with intrinsic biochemical MIRSI measurements enables quantitative evaluation of cancerous tissue, improving the rigor and reproducibility of histopathology.

Polarization Sensitive Photothermal Mid-Infrared Spectroscopic Imaging of Human Bone Marrow Tissue.

Collagen quantity and integrity play an important role in understanding diseases such as myelofibrosis (MF). Label-free mid-infrared spectroscopic imaging (MIRSI) has the potential to quantify collagen while minimizing the subjective variance observed with conventional histopathology. Infrared (IR) spectroscopy with polarization sensitivity provides chemical information while also estimating tissue dichroism. This can potentially aid MF grading by revealing the structure and orientation of collagen fibers. Simultaneous measurement of collagen structure and biochemical properties can translate clinically into improved diagnosis and enhance our understanding of disease progression. In this paper, we present the first report of polarization-dependent spectroscopic variations in collagen from human bone marrow samples. We build on prior work with animal models and extend it to human clinical biopsies with a practical method for high-resolution chemical and structural imaging of bone marrow on clinical glass slides. This is done using a new polarization-sensitive photothermal mid-infrared spectroscopic imaging scheme that enables sample and source independent polarization control. This technology provides 0.5 µm spatial resolution, enabling the identification of thin (≈1 µm) collagen fibers that were not separable using Fourier Transform Infrared (FT-IR) imaging in the fingerprint region at diffraction-limited resolution ( ≈ 5 µm). Finally, we propose quantitative metrics to identify fiber orientation from discrete band images (amide I and amide II) measured under three polarizations. Previous studies have used a pair of orthogonal polarization measurements, which is insufficient for clinical samples since human bone biopsies contain collagen fibers with multiple orientations. Here, we address this challenge and demonstrate that three polarization measurements are necessary to resolve orientation ambiguity in clinical bone marrow samples. This is also the first study to demonstrate the ability to spectroscopically identify thin collagen fibers (≈1 µm diameter) and their orientations, which is critical for accurate grading of human bone marrow fibrosis.

Multi-modal image sharpening in fourier transform infrared (FTIR) microscopy.

Mid-infrared Spectroscopic Imaging (MIRSI) provides spatially-resolved molecular specificity by measuring wavelength-dependent mid-infrared absorbance. Infrared microscopes use large numerical aperture objectives to obtain high-resolution images of heterogeneous samples. However, the optical resolution is fundamentally diffraction-limited, and therefore wavelength-dependent. This significantly limits resolution in infrared microscopy, which relies on long wavelengths (2.5 µm to 12.5 µm) for molecular specificity. The resolution is particularly restrictive in biomedical and materials applications, where molecular information is encoded in the fingerprint region (6 µm to 12 µm), limiting the maximum resolving power to between 3 µm and 6 µm. We present an unsupervised curvelet-based image fusion method that overcomes limitations in spatial resolution by augmenting infrared images with label-free visible microscopy. We demonstrate the effectiveness of this approach by fusing images of breast and ovarian tumor biopsies acquired using both infrared and dark-field microscopy. The proposed fusion algorithm generates a hyperspectral dataset that has both high spatial resolution and good molecular contrast. We validate this technique using multiple standard approaches and through comparisons to super-resolved experimentally measured photothermal spectroscopic images. We also propose a novel comparison method based on tissue classification accuracy.

Automated Osteosclerosis Grading of Clinical Biopsies Using Infrared Spectroscopic Imaging.

Osteosclerosis and myefibrosis are complications of myeloproliferative neoplasms. These disorders result in excess growth of trabecular bone and collagen fibers that replace hematopoietic cells, resulting in abnormal bone marrow function. Treatments using imatinib and JAK2 pathway inhibitors can be effective on osteosclerosis and fibrosis; therefore, accurate grading is critical for tracking treatment effectiveness. Current grading standards use a four-class system based on analysis of biopsies stained with three histological stains: hematoxylin and eosin (H&E), Masson's trichrome, and reticulin. However, conventional grading can be subjective and imprecise, impacting the effectiveness of treatment. In this Article, we demonstrate that mid-infrared spectroscopic imaging may serve as a quantitative diagnostic tool for quantitatively tracking disease progression and response to treatment. The proposed approach is label-free and provides automated quantitative analysis of osteosclerosis and collagen fibrosis.

Mitigating fringing in discrete frequency infrared imaging using time-delayed integration.

Infrared (IR) spectroscopic microscopes provide the potential for label-free quantitative molecular imaging of biological samples, which can be used to aid in histology, forensics, and pharmaceutical analysis. Most IR imaging systems use broadband illumination combined with a spectrometer to separate the signal into spectral components. This technique is currently too slow for many biomedical applications such as clinical diagnosis, primarily due to the availability of bright mid-infrared sources and sensitive MCT detectors. There has been a recent push to increase throughput using coherent light sources, such as synchrotron radiation and quantum cascade lasers. While these sources provide a significant increase in intensity, the coherence introduces fringing artifacts in the final image. We demonstrate that applying time-delayed integration in one dimension can dramatically reduce fringing artifacts with minimal alterations to the standard infrared imaging pipeline. The proposed technique also offers the potential for less expensive focal plane array detectors, since linear arrays can be more readily incorporated into the proposed framework.

Selecting optimal features from Fourier transform infrared spectroscopy for discrete-frequency imaging.

Tissue histology utilizing chemical and immunohistochemical labels plays an important role in biomedicine and disease diagnosis. Recent research suggests that mid-infrared (IR) spectroscopic imaging may augment histology by providing quantitative molecular information. One of the major barriers to this approach is long acquisition time using Fourier-transform infrared (FTIR) spectroscopy. Recent advances in discrete frequency sources, particularly quantum cascade lasers (QCLs), may mitigate this problem by allowing selective sampling of the absorption spectrum. However, DFIR imaging only provides a significant advantage when the number of spectral samples is minimized, requiring a priori knowledge of important spectral features. In this paper, we demonstrate the use of a GPU-based genetic algorithm (GA) using linear discriminant analysis (LDA) for DFIR feature selection. Our proposed method relies on pre-acquired broadband FTIR images for feature selection. Based on user-selected criteria for classification accuracy, our algorithm provides a minimal set of features that can be used with DFIR in a time-frame more practical for clinical diagnosis.

SIproc: an open-source biomedical data processing platform for large hyperspectral images.

There has recently been significant interest within the vibrational spectroscopy community to apply quantitative spectroscopic imaging techniques to histology and clinical diagnosis. However, many of the proposed methods require collecting spectroscopic images that have a similar region size and resolution to the corresponding histological images. Since spectroscopic images contain significantly more spectral samples than traditional histology, the resulting data sets can approach hundreds of gigabytes to terabytes in size. This makes them difficult to store and process, and the tools available to researchers for handling large spectroscopic data sets are limited. Fundamental mathematical tools, such as MATLAB, Octave, and SciPy, are extremely powerful but require that the data be stored in fast memory. This memory limitation becomes impractical for even modestly sized histological images, which can be hundreds of gigabytes in size. In this paper, we propose an open-source toolkit designed to perform out-of-core processing of hyperspectral images. By taking advantage of graphical processing unit (GPU) computing combined with adaptive data streaming, our software alleviates common workstation memory limitations while achieving better performance than existing applications.