Artificial Intelligence-Based Assessment of Colorectal Polyp Histology by Elastic-Scattering Spectroscopy.

BACKGROUND: Colonoscopic screening and surveillance for colorectal cancer could be made safer and more efficient if endoscopists could predict histology without the need to biopsy and perform histopathology on every polyp. Elastic-scattering spectroscopy (ESS), using fiberoptic probes integrated into standard biopsy tools, can assess, both in vivo and in real time, the scattering and absorption properties of tissue related to its underlying pathology. AIMS: The objective of this study was to evaluate prospectively the potential of ESS to predict polyp pathology accurately. METHODS: We obtained ESS measurements from patients undergoing screening/surveillance colonoscopy using an ESS fiberoptic probe integrated into biopsy forceps. The integrated forceps were used for tissue acquisition, following current standards of care, and optical measurement. All measurements were correlated to the index pathology. A machine learning model was then applied to measurements from 367 polyps in 169 patients to prospectively evaluate its predictive performance. RESULTS: The model achieved sensitivity of 0.92, specificity of 0.87, negative predictive value (NPV) of 0.87, and high-confidence rate (HCR) of 0.84 for distinguishing 220 neoplastic polyps from 147 non-neoplastic polyps of all sizes. Among 138 neoplastic and 131 non-neoplastic polyps ≤ 5 mm, the model achieved sensitivity of 0.91, specificity of 0.88, NPV of 0.89, and HCR of 0.83. CONCLUSIONS: Results show that ESS is a viable endoscopic platform for real-time polyp histology, particularly for polyps ≤ 5 mm. ESS is a simple, low-cost, clinically friendly, optical biopsy modality that, when interfaced with minimally obtrusive endoscopic tools, offers an attractive platform for in situ polyp assessment.

Optical Spectroscopy as a Method for Skin Cancer Risk Assessment.

Skin cancer is the most prevalent cancer, and its assessment remains a challenge for physicians. This study reports the application of an optical sensing method, elastic scattering spectroscopy (ESS), coupled with a classifier that was developed with machine learning, to assist in the discrimination of skin lesions that are concerning for malignancy. The method requires no special skin preparation, is non-invasive, easy to administer with minimal training, and allows rapid lesion classification. This novel approach was tested for all common forms of skin cancer. ESS spectra from a total of 1307 lesions were analyzed in a multi-center, non-randomized clinical trial. The classification algorithm was developed on a 950-lesion training dataset, and its diagnostic performance was evaluated against a 357-lesion testing dataset that was independent of the training dataset. The observed sensitivity was 100% (14/14) for melanoma and 94% (105/112) for non-melanoma skin cancer. The overall observed specificity was 36% (84/231). ESS has potential, as an adjunctive assessment tool, to assist physicians to differentiate between common benign and malignant skin lesions.

Towards minimally-invasive, quantitative assessment of chronic kidney disease using optical spectroscopy.

The universal pathologic features implicated in the progression of chronic kidney disease (CKD) are interstitial fibrosis and tubular atrophy (IFTA). Current methods of estimating IFTA are slow, labor-intensive and fraught with variability and sampling error, and are not quantitative. As such, there is pressing clinical need for a less-invasive and faster method that can quantitatively assess the degree of IFTA. We propose a minimally-invasive optical method to assess the macro-architecture of kidney tissue, as an objective, quantitative assessment of IFTA, as an indicator of the degree of kidney disease. The method of elastic-scattering spectroscopy (ESS) measures backscattered light over the spectral range 320-900 nm and is highly sensitive to micromorphological changes in tissues. Using two discrete mouse models of CKD, we observed spectral trends of increased scattering intensity in the near-UV to short-visible region (350-450 nm), relative to longer wavelengths, for fibrotic kidneys compared to normal kidney, with a quasi-linear correlation between the ESS changes and the histopathology-determined degree of IFTA. These results suggest the potential of ESS as an objective, quantitative and faster assessment of IFTA for the management of CKD patients and in the allocation of organs for kidney transplantation.

The color of cancer: Margin guidance for oral cancer resection using elastic scattering spectroscopy.

OBJECTIVES/HYPOTHESIS: To evaluate the usefulness of elastic scattering spectroscopy (ESS) as a diagnostic adjunct to frozen section analysis in patients with diagnosed squamous cell carcinoma of the oral cavity. STUDY DESIGN: Prospective analytic study. METHODS: Subjects for this single institution, institutional review board-approved study were recruited from among patients undergoing surgical resection for squamous cell cancer of the oral cavity. A portable ESS device with a contact fiberoptic probe was used to obtain spectral signals. Four to 10 spectral readings were obtained on each subject from various sites including gross tumor and normal-appearing mucosa in the surgical margin. Each reading was correlated with the histopathologic findings of biopsies taken from the exact location of the spectral readings. A diagnostic algorithm based on multidimensional pattern recognition/machine learning was developed. Sensitivity and specificity, error rate, and area under the curve were used as performance metrics for tests involving classification between disease and nondisease classes. RESULTS: Thirty-four (34) subjects were enrolled in the study. One hundred seventy-six spectral data point/biopsy specimen pairs were available for analysis. ESS distinguished normal from abnormal tissue, with a sensitivity ranging from 84% to 100% and specificity ranging from 71% to 89%, depending on how the cutoff between normal and abnormal tissue was defined (i.e., mild, moderate, or severe dysplasia). There were statistically significant differences in malignancy scores between histologically normal tissue and invasive cancer and between noninflamed tissue and inflamed tissue. CONCLUSIONS: This is the first study to evaluate the effectiveness of ESS in guiding mucosal resection margins in oral cavity cancer. ESS provides fast, real-time assessment of tissue without the need for pathology expertise. ESS appears to be effective in distinguishing between normal mucosa and invasive cancer and between "normal" tissue (histologically normal and mild dysplasia) and "abnormal" tissue (severe dysplasia and carcinoma in situ) that might require further margin resection. Further studies, however, are needed with a larger sample size to validate these findings and to determine the effectiveness of ESS in distinguishing visibly and histologically normal tissue from visibly normal but histologically abnormal tissue. LEVEL OF EVIDENCE: NA Laryngoscope, 127:S1-S9, 2017.

Elastic light-scattering spectroscopy for discrimination of benign from malignant disease in thyroid nodules.

BACKGROUND: Thyroid cancer is the most common endocrine malignancy. The current standard of diagnosis, fine-needle aspiration biopsy, yields approximately 10-25% of indeterminate results leading to twice as many thyroidectomies for further diagnosis. Elastic scattering spectroscopy (ESS) is a new, minimally invasive optical-biopsy technique mediated by fiber-optic probes that is sensitive to cellular and subcellular morphological features. We assessed the diagnostic potential of ESS in the thyroid to differentiate benign from malignant thyroid nodules as determined by histology. METHODS: Under an IRB approved protocol, 36 surgical patients (n = 21 benign thyroid nodules, n = 15 malignant tumors) had collection of ESS data from their fresh ex vivo thyroidectomy specimens. Using surgical pathology as our gold standard, spectral analyses were performed using a training set; these data were used to assess the ESS diagnostic potential using the leave-one-out technique. RESULTS: Our test set was 75% sensitive and 95% specific in differentiating benign from malignant thyroid lesions, with a positive predictive value (PPV) of 0.92 and a negative predictive value (NPV) of 0.83. CONCLUSIONS: The ESS can accurately distinguish benign vs malignant thyroid lesions with high PPV and NPV. With further validation ESS could potentially be used as an in situ real-time diagnostic tool or as an adjunct to conventional cytology.

Design of a system to measure light scattering from individual cells excited by an acoustic wave.

A system to measure light scattering from individual cells excited by an acoustic wave was designed, and tests were performed on live Jurkat cells. Cells passing in a laminar stream within a water bath were excited by a focused ultrasound pulse, while the scattered light from a laser beam was monitored at various scattering angles. The cells were modeled as viscoelastic liquid drops, which return to equilibrium via shape oscillations after an acoustically-induced deformation. The Fast Fourier Transform of the scattered light signal was used to extract information about the highly-damped resonant frequencies of the cells, and the detected frequencies are consistent with theoretical predictions.

Analysis of changes in reflectance measurements on biological tissues subjected to different probe pressures.

Spectral reflectance measurements of biological tissues have been studied for early diagnoses of several pathologies such as cancer. These measurements are often performed with a fiber optic probe in contact with the tissue surface. We report a study in which reflectance measurements are obtained in vivo from mouse thigh muscle while varying the contact pressure of the fiber optic probe. It is determined that the probe pressure is a variable that affects the local optical properties of the tissue. The reflectance spectra are analyzed with an analytical model that extracts the tissue optical properties and facilitates the understanding of underlying physiological changes induced by the probe pressure.

Analytical model of light reflectance for extraction of the optical properties in small volumes of turbid media.

Monte Carlo simulations and experiments in tissue phantoms were used to empirically develop an analytical model that characterizes the reflectance spectrum in a turbid medium. The model extracts the optical properties (scattering and absorption coefficients) of the medium at small source-detector separations, for which the diffusion approximation is not valid. The accuracy of the model and the inversion algorithm were investigated and validated. Four fiber probe configurations were tested for which both the source and the detector fibers were tilted at a predetermined angle, with the fibers parallel to each other. This parallel-fiber geometry facilitates clinical endoscopic applications and ease of fabrication. Accurate extraction of tissue optical properties from in vivo spectral measurements could have potential applications in detecting, noninvasively and in real time, epithelial (pre)cancers.

Optical method for real-time monitoring of drug concentrations facilitates the development of novel methods for drug delivery to brain tissue.

The understanding of drug delivery to organs, such as the brain, has been hampered by the inability to measure tissue drug concentrations in real time. We report an application of an optical spectroscopy technique that monitors in vivo the real-time drug concentrations in small volumes of brain tissue. This method will facilitate development of new protocols for delivery of drugs to treat brain cancers. The delivery of many anticancer drugs to the brain is limited by the presence of the blood-brain barrier (BBB). Mitoxantrone (MTX) is a water-soluble anticancer drug that poorly penetrates the BBB. It is preliminarily determined in an animal model that the brain tissue uptake of chemotherapy agents-in this demonstration, MTX-delivered intra-arterially is enhanced when the BBB is disrupted.

Comparison between ultraviolet-visible and near-infrared elastic scattering spectroscopy of chemically induced melanomas in an animal model.

The work reported compares elastic scattering spectroscopy (ESS) for diagnosis of pigmented skin lesions in two spectral regions: UV-visible and near infrared (NIR). Given the known strong absorption by melanin in the near-UV to mid-visible range of the spectrum, such a comparison can help determine the optimum wavelength range of ESS for diagnosis of pigmented skin lesions. For this purpose, four South American opossums are treated with dimethylbenz(a)anthracene on multiple dorsal sites to induce both malignant melanomas and benign pigmented lesions. Skin lesions are examined in vivo with ESS using both UV-visible and NIR, with wavelength ranges of 330 to 900 nm and 900 to 1700 nm, respectively. Both portable systems use the same fiber optic probe geometry. ESS measurements are made on the lesions, and spectral differences are grouped by diagnosis from standard histopathological procedure. Both ESS datasets show strong spectral trends with the histopathological assignments, and the data suggest a model for the underlying basis of the spectral distinction between benign and malignant pigmented nevi.