Spectral classifier design with ensemble classifiers and misclassification-rejection: application to elastic-scattering spectroscopy for detection of colonic neoplasia

Distinguishing tumor from healthy tissue in human liver ex vivo using machine learning and multivariate analysis of diffuse reflectance spectra

The aim of this work was to evaluate the capability of diffuse reflectance spectroscopy to distinguish malignant liver tissues from surrounding tissues and to determine whether an extended wavelength range (450-1550 nm) offers any advantages over using the conventional wavelength range. Furthermore, multivariate analysis combined with a machine learning algorithm, either linear discriminant analysis or the more advanced support vector machine, was used to discriminate between and classify freshly excised human liver specimens from 18 patients. Tumors were distinguished from surrounding liver tissues with a sensitivity of 99%, specificity of 100%, classification rate of 100% and a Matthews correlation coefficient of 100% using the extended wavelength range and a combination of principal component analysis and support vector techniques. The results indicate that this technology may be useful in clinical applications for real-time tissue diagnostics of tumor margins where rapid classification is important.

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.

Development of artificial intelligence technology in diagnosis, treatment, and prognosis of colorectal cancer

Artificial intelligence (AI) technology has made leaps and bounds since its invention. AI technology can be subdivided into many technologies such as machine learning and deep learning. The application scope and prospect of different technologies are also totally different. Currently, AI technologies play a pivotal role in the highly complex and wide-ranging medical field, such as medical image recognition, biotechnology, auxiliary diagnosis, drug research and development, and nutrition. Colorectal cancer (CRC) is a common gastrointestinal cancer that has a high mortality, posing a serious threat to human health. Many CRCs are caused by the malignant transformation of colorectal polyps. Therefore, early diagnosis and treatment are crucial to CRC prognosis. The methods of diagnosing CRC are divided into imaging diagnosis, endoscopy, and pathology diagnosis. Treatment methods are divided into endoscopic treatment, surgical treatment, and drug treatment. AI technology is in the weak era and does not have communication capabilities. Therefore, the current AI technology is mainly used for image recognition and auxiliary analysis without in-depth communication with patients. This article reviews the application of AI in the diagnosis, treatment, and prognosis of CRC and provides the prospects for the broader application of AI in CRC.

Advanced endoscopic imaging for detecting and guiding therapy of early neoplasias of the esophagus

Esophageal cancers, largely adenocarcinoma in Western countries and squamous cell cancer in Asia, present a significant burden of disease and remain one of the most lethal of cancers. Key to improving survival is the development and adoption of new imaging modalities to identify early neoplastic lesions, which may be small, multifocal, subsurface, and difficult to detect by standard endoscopy. Such advanced imaging is particularly relevant with the emergence of ablative techniques that often require multiple endoscopic sessions and may be complicated by bleeding, pain, strictures, and recurrences. Assessing the specific location, depth of involvement, and features correlated with neoplastic progression or incomplete treatment may optimize treatments. While not comprehensive of all endoscopic imaging modalities, we review here some of the recent advances in endoscopic luminal imaging, particularly with surface contrast enhancement using virtual chromoendoscopy, highly magnified subsurface imaging with confocal endomicroscopy, optical coherence tomography, elastic scattering spectroscopy, angle-resolved low-coherence interferometry, and light scattering spectroscopy. While there is no single ideal imaging modality, various multimodal instruments are also being investigated. The future of combining computer-aided assessments, molecular markers, and improved imaging technologies to help localize and ablate early neoplastic lesions shed hope for improved disease outcome.

Spectral response of optical fiber probe with closely spaced fibers

Background

Optical fiber probe spectroscopy can characterize the blood content, hemoglobin oxygen saturation, water content, and scattering properties of a tissue. A narrow probe using closely spaced fibers can access and characterize a local tissue site, but analysis requires the proper light transport theory.

Methods

Monte Carlo simulations of photon transport specified the response of a two-fiber probe as a function of optical properties in a homogeneous tissue. The simulations used the dimensions of a commercial fiber probe (400-micron-diameter fibers separated by 80-microns of cladding) to calculate the response to a range of 20 absorption and 20 reduced scattering values. The 400 simulations yielded an analysis grid (lookup table) to interpolate the probe response to any given pair of absorption and scattering properties.

Results

The probe in contact with tissue is not sensitive to low absorption but sensitive to scattering, as occurs for red to near-infrared spectra. The probe is sensitive to both absorption and scattering for shorter visible spectra (purple-orange). The non-contact probe held above the tissue delivers light to/from a spot on the tissue and fails to collect light that spreads laterally to escape outside the collection spot. Such partial collection can distort the spectra.

Conclusions

Optical fiber spectroscopy using closely spaced fibers requires proper calibration. An analysis subroutine is provided for analysis of a two-fiber probe with the dimensions of a commercial probe (Ocean Insight), but the method can be applied to any probe design. A closely spaced fiber probe can document blood in the shorter visible wavelengths, but has difficulty detecting red and near-infra-red absorption. Hence detection of hydration is difficult. The strength of the closely spaced fiber probe is detecting scattering that depends on tissue structure at the micron to sub-micron scale.

Computer-assisted assessment of colonic polyp histopathology using probe-based confocal laser endomicroscopy

INTRODUCTION

Probe-based confocal laser endomicroscopy (pCLE) is a promising modality for classifying polyp histology in vivo, but decision making in real-time is hampered by high-magnification targeting and by the learning curve for image interpretation. The aim of this study is to test the feasibility of a system combining the use of a low-magnification, wider field-of-view pCLE probe and a computer-assisted diagnosis (CAD) algorithm that automatically classifies colonic polyps.

METHODS

This feasibility study utilized images of polyps from 26 patients who underwent colonoscopy with pCLE. The pCLE images were reviewed offline by two expert and five junior endoscopists blinded to index histopathology. A subset of images was used to train classification software based on the consensus of two GI histopathologists. Images were processed to extract image features as inputs to a linear support vector machine classifier. We compared the CAD algorithm's prediction accuracy against the classification accuracy of the endoscopists.

RESULTS

We utilized 96 neoplastic and 93 non-neoplastic confocal images from 27 neoplastic and 20 non-neoplastic polyps. The CAD algorithm had sensitivity of 95%, specificity of 94%, and accuracy of 94%. The expert endoscopists had sensitivities of 98% and 95%, specificities of 98% and 96%, and accuracies of 98% and 96%, while the junior endoscopists had, on average, a sensitivity of 60%, specificity of 85%, and accuracy of 73%.

CONCLUSION

The CAD algorithm showed comparable performance to offline review by expert endoscopists and improved performance when compared to junior endoscopists and may be useful for assisting clinical decision making in real time.

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.

Developments in optical imaging for gastrointestinal surgery

To improve outcomes for patients with cancer, in terms of both survival and a reduction in the morbidity and mortality that results from surgical resection and treatment, there are two main areas that require improvement. Accurate early diagnosis of the cancer, at a stage where curative and, ideally, minimally invasive treatment is achievable, is desired as well as identification of tumor margins, lymphatic and distant disease, enabling complete, but not unnecessarily extensive, resection. Optical imaging is making progress in achieving these aims. This review discusses the principles of optical imaging, focusing on fluorescence and spectroscopy, and the current research that is underway in GI tract carcinomas.

Detection of antibiotic resistant Escherichia Coli bacteria using infrared microscopy and advanced multivariate analysis

DeterminingE. colibacteria susceptibility by analyzing their FTIR spectra using multivariate analysis.

Methods to become a high performer in characterization of colorectal polyp histology

The recent advent of advanced imaging technologies has brought real time characterization of polyp histology to the forefront. This concept of optical diagnosis of diminutive polyp histology can bring about a huge paradigm shift in the management of these lesions. Instead of resecting and sending all the diminutive polyps to pathology, there is the potential to practice "resect and discard" for those predicted to be adenomas and "do not resect" strategy for the recto-sigmoid polyps predicted to be hyperplastic. However, one of the major steps before the clinical implementation of real-time histology can be a reality, will be training endoscopists with varying levels of experience in novel imaging technologies. The two major methods for training include didactic teaching and the computer based method. After the initial training, it is imperative that the endoscopists practice this skill during performance of routine colonoscopy to auto validate and assess their own competency. Both practice and reinforcement can help endoscopists become high performers in the characterization of polyp histology.

Endoscopic histological assessment of colonic polyps by using elastic scattering spectroscopy

BACKGROUND

Elastic-scattering spectroscopy (ESS) can assess in vivo and in real-time the scattering and absorption properties of tissue related to underlying pathologies.

OBJECTIVES

To evaluate the potential of ESS for differentiating neoplastic from non-neoplastic polyps during colonoscopy.

DESIGN

Pilot study, retrospective data analysis.

SETTING

Academic practice.

PATIENTS

A total of 83 patients undergoing screening/surveillance colonoscopy.

INTERVENTIONS

ESS spectra of 218 polyps (133 non-neoplastic, 85 neoplastic) were acquired during colonoscopy. Spectral data were correlated with the classification of biopsy samples by 3 GI pathologists. High-dimensional methods were used to design diagnostic algorithms.

MAIN OUTCOME MEASUREMENTS

Diagnostic performance of ESS.

RESULTS

Analysis of spectra from polyps of all sizes (N = 218) resulted in a sensitivity of 91.5%, specificity of 92.2%, and accuracy of 91.9% with a high-confidence rate of 90.4%. Restricting analysis to polyps smaller than 1 cm (n = 179) resulted in a sensitivity of 87.0%, specificity of 92.1%, and accuracy of 90.6% with a high-confidence rate of 89.3%. Analysis of polyps 5 mm or smaller (n = 157) resulted in a sensitivity of 86.8%, specificity of 91.2%, and accuracy of 90.1% with a high-confidence rate of 89.8%.

LIMITATIONS

Sample size, retrospective validation used to obtain performance estimates.

CONCLUSION

Results indicate that ESS permits accurate, real-time classification of polyps as neoplastic or non-neoplastic. ESS is a simple, low cost, clinically robust method with minimal impact on procedure flow, especially when integrated into standard endoscopic biopsy tools. Performance on polyps 5 mm or smaller indicates that ESS may, in theory, achieve Preservation and Incorporation of Valuable Endoscopic Innovations performance thresholds. ESS may one day prove to be a useful tool used in endoscopic screening and surveillance of colorectal cancer.

Elastic scattering spectroscopy as an optical marker of inflammatory bowel disease activity and subtypes

BACKGROUND

In 10% to 15% of individuals, inflammatory bowel disease (IBD) is difficult to classify as ulcerative colitis (UC) or Crohn's disease (CD). Previous work has demonstrated that probe-based elastic scattering spectroscopy (ESS) can produce spectra, informed by parameters like tissue ultrastructure and hemoglobin content, capable of differentiating pathologies. This study investigates whether ESS is an in vivo optical biomarker for the presence, activity, and type of IBD in the colon.

METHODS

Pilot study, a retrospective data analysis. ESS spectra of endoscopically normal and inflamed colon were obtained from 48 patients with IBD and 46 non-IBD controls. Measurements from patients with IBD were categorized as CD or UC based on clinical diagnosis. Spectra were analyzed using high-dimensional methods. Leave-one-patient-out cross-validation was used to obtain diagnostic performance estimates.

RESULTS

Patients with IBD were distinguishable from non-IBD controls with a sensitivity of 0.93 and specificity of 0.91 based on readings from endoscopically normal mucosa, and 0.94 and 0.93 from inflamed mucosa. In patients with IBD, histologically normal and inflamed colon were distinguishable with per-class accuracies of 0.83 and 0.89, respectively; histologically normal from inactive inflammation with accuracies of 0.73 and 0.89, respectively; and inactive from active colitis with accuracies of 0.87 and 0.84, respectively. The diagnosis of CD versus UC was made with per-class accuracies of 0.92 and 0.87 in normal and 0.87 and 0.85 in inflamed mucosa, respectively.

CONCLUSIONS

ESS, a simple, low-cost clinically friendly optical biopsy modality, has the potential to enhance the endoscopic assessment of IBD and its activity in real time and may help to distinguish CD from UC.

Integrated Device for in Vivo Fine Needle Aspiration Biopsy and Elastic Scattering Spectroscopy in Preoperative Thyroid Nodules

Thyroid nodules are a frequent clinical finding and the most common endocrine malignancy is thyroid cancer. The standard of care in the management of a patient with a thyroid nodule is to perform a preoperative fine needle aspiration (FNA) biopsy of the suspect nodule under ultrasound imaging guidance. In a significant percentage of the cases, cytological assessment of the biopsy material yields indeterminate results, the consequence of which is diagnostic thyroidectomy. Unfortunately, 75–80% of diagnostic thyroidectomies following indeterminate cytology result in benign designation by post-surgery histopathology, indicating potentially unnecessary surgeries. Clearly, the potential exists for the improvement in patient care and the reduction of overall procedure costs if an improved preoperative diagnostic technique was developed. Elastic scattering spectroscopy (ESS) is an optical biopsy technique that is mediated by optical fiber probes and has been shown to be effective in differentiating benign from malignant thyroid tissue in ex vivo surgical tissue samples. The goal of the current research was to integrate the ESS fiber optic probes into a device that can also collect cells for cytological assessment and, thus, enable concurrent spectroscopic interrogation and biopsy of a suspect nodule with a single needle penetration. The primary challenges to designing the device included miniaturizing the standard ESS fiber optic probe to fit within an FNA needle and maintaining the needle’s aspiration functionality. We demonstrate the value of the fabricated prototype devices by assessing their preliminary performance in an on-going clinical study with >120 patients. The devices have proven to be clinically friendly, collecting both aspirated cells and optical data from the same location in thyroid nodules and with minimal disruption of clinical procedure. In the future, such integrated devices could be used to complement FNA-based cytological results and have the potential to both reduce the number of diagnostic thyroidectomies on benign nodules and improve the surgical approach for patients with thyroid malignancies, thereby, decreasing healthcare costs and improving patient outcomes.

Impact of a computer-based teaching module on characterization of diminutive colon polyps by using narrow-band imaging by non-experts in academic and community practice: a video-based study

BACKGROUND

Experts can accurately characterize the histology of diminutive polyps with narrow-band imaging (NBI). There are limited data on the performance of non-experts.

OBJECTIVE

To assess the impact of a computer-based teaching module on the accuracy of predicting polyp histology with NBI by non-experts (in academics and community practice) by using video clips.

DESIGN

Prospective, observational study.

SETTING

Academic and community practice.

PARTICIPANTS

A total of 15 gastroenterologists participated-5 experts in NBI, 5 non-experts in academic practice, and 5 non-experts in community practice.

INTERVENTION

Participants reviewed a 20-minute, computer-based teaching module outlining the different NBI features for hyperplastic and adenomatous polyps.

MAIN OUTCOME MEASUREMENTS

Performance characteristics in characterizing the histology of diminutive polyps with NBI by using short video clips before (pretest) and after (posttest) reviewing the teaching module.

RESULTS

Non-experts in academic practice showed a significant improvement in the sensitivity (54% vs 79%; P < .001), accuracy (64% vs 81%; P < .001), and proportion of high-confidence diagnoses (49% vs 69%; P < .001) in the posttest. Non-experts in community practice had significantly higher sensitivity (58% vs 75%; P = .004), specificity (76% vs 90%; P = .04), accuracy (64% vs 81%; P < .001), and proportion of high-confidence diagnoses (49% vs 72%; P < .001) in the posttest. Performance of experts in NBI was significantly better than non-experts in both academic and community practice.

LIMITATIONS

Selection bias in selecting good quality videos. Performance not assessed during live colonoscopy.

CONCLUSION

Academic and community gastroenterologists without prior experience in NBI can achieve significant improvements in characterizing diminutive polyp histology after a brief computer-based training. The durability of these results and applicability in everyday practice are uncertain.

Influence of the phase function in generalized diffuse reflectance models: review of current formalisms and novel observations

Diffuse reflectance spectroscopy, which has been demonstrated as a noninvasive diagnostic technique, relies on quantitative models for extracting optical property values from turbid media, such as biological tissues. We review and compare reflectance models that have been published, and we test similar models over a much wider range of measurement parameters than previously published, with specific focus on the effects of the scattering phase function and the source-detector distance. It has previously been shown that the dependence of a forward reflectance model on the scattering phase function can be described more accurately using a variable, γ, which is a more predictive variable for reflectance than the traditional anisotropy factor, g. We show that variations in the reflectance model due to the phase function are strongly dependent on the source-detector separation, and we identify a dimensionless scattering distance at which reflectance is insensitive to the phase function. Further, we evaluate how variations in the phase function and source-detector separation affect the accuracy of inverse property extraction. By simultaneously fitting two or more reflectance spectra, measured at different source-detector separations, we also demonstrate that an estimate of γ can be extracted, in addition to the reduced scattering and absorption coefficients.

Preoperative discrimination of benign from malignant disease in thyroid nodules with indeterminate cytology using elastic light-scattering spectroscopy

Thyroid nodules are common and often require fine needle aspiration biopsy (FNAB) to determine the presence of malignancy to direct therapy. Unfortunately, approximately 15-30% of thyroid nodules evaluated by FNAB are not clearly benign or malignant by cytology alone. These patients require surgery for the purpose of diagnosis alone; most of these nodules ultimately prove to be benign. Elastic light scattering spectroscopy (ESS) that measures the spectral differences between benign and malignant thyroid nodules has shown promise in improving preoperative determination of benign status of thyroid nodules. We describe the results of a large, prospective, blinded study validating the ESS algorithm in patients with thyroid nodules. An ESS system was used to acquire spectra from human thyroid tissue. Spectroscopic results were compared to the histopathology of the biopsy samples. Sensitivity and specificity of the ESS system in the differentiation of benign from malignant thyroid nodules are 74% and 90% respectively, with a negative predictive value of 97%. These data suggest that ESS has the potential for use in real time diagnosis of thyroid nodules as an adjunct to FNAB cytology.