Detection of cervical lesions by multivariate analysis of diffuse reflectance spectra: a clinical study

Cervical Cancer Tissue Analysis Using Photothermal Midinfrared Spectroscopic Imaging

Hyperspectral photothermal mid-infrared spectroscopic imaging (HP-MIRSI) is an emerging technology with promising applications in cervical cancer diagnosis and quantitative, label-free histopathology. This study pioneers the application of HP-MIRSI to the evaluation of clinical cervical cancer tissues, achieving excellent tissue type segmentation accuracy of over 95%. This achievement stems from an integrated approach of optimized data acquisition, computational data reconstruction, and the application of machine learning algorithms. The results are statistically robust, drawing from tissue samples of 98 cervical cancer patients and incorporating over 40 million data points. Traditional cervical cancer diagnosis methods entail biopsy, staining, and visual evaluation by a pathologist. This process is qualitative, subject to variations in staining and subjective interpretations, and requires extensive tissue processing, making it costly and time-consuming. In contrast, our proposed alternative can produce images comparable to those from histological analyses without the need for staining or complex sample preparation. This label-free, quantitative method utilizes biochemical data from HP-MIRSI and employs machine-learning algorithms for the rapid and precise segmentation of cervical tissue subtypes. This approach can potentially transform histopathological analysis by offering a more accurate and label-free alternative to conventional diagnostic processes.

Enhancing Intraoperative Tissue Identification: Investigating a Smart Electrosurgical Knife's Functionality During Electrosurgery

OBJECTIVE

Detecting the cancerous growth margin and achieving a negative margin is one of the challenges that surgeons face during cancer procedures. A smart electrosurgical knife with integrated optical fibers has been designed previously to enable real-time use of diffuse reflectance spectroscopy for intraoperative margin assessment. In this paper, the thermal effect of the electrosurgical knife on tissue sensing is investigated.

METHODS

Porcine tissues and phantoms were used to investigate the performance of the smart electrosurgical knife after electrosurgery. The fat-to-water content ratio (F/W-ratio) served as the discriminative parameter for distinguishing tissues and tissue mimicking phantoms with varying fat content. The F/W-ratio of tissues and phantoms was measured with the smart electrosurgical knife before and after 14 minutes of electrosurgery. Additionally, a layered porcine tissue and phantom were sliced and measured from top to bottom with the smart electrosurgical knife.

RESULTS

Mapping the thermal activity of the electrosurgical knife's electrode during animal tissue electrosurgery revealed temperatures exceeding 400 °C. Electrosurgery for 14 minutes had no impact on the device's accurate detection of the F/W-ratio. The smart electrosurgical knife enables real-time tissue detection and predicts the fat content of the next layer from 4 mm ahead.

CONCLUSION

The design of the smart electrosurgical knife outlined in this paper demonstrates its potential utility for tissue detection during electrosurgery.

SIGNIFICANCE

In the future, the smart electrosurgical knife could be a valuable intraoperative margin assessment tool, aiding surgeons in detecting tumor borders and achieving negative margins.

A combined autofluorescence and diffuse reflectance spectroscopy for mucosa tissue diagnosis: Dual-distance system and data-driven decision

Combined autofluorescence (AF) and diffuse reflectance (DR) spectroscopies have been expected to offer enhanced diagnostic accuracies for noninvasive early detection of mucosa lesions, that is, oral cavity carcinoma and cervical carcinoma. This work reports on a hybrid AF and DR spectroscopic system that is developed for quantification and diagnosis of mucosa abnormalities. The system stability and reliability are firstly assessed by phantom experiments, showing a measurement variation lower than 1% within 20 min. In vitro and in vivo validations are then conducted for tissue identification and lesion differentiation. For enhanced decision, a data-driven diagnosis algorithm is explored in pilot under different experimental configurations. The results conclude a promising accuracy of >96% for the in vivo classification as well as an excellent sensitivity of >88% for the in vitro mucosa lesions detection, and demonstrate sound potential of the system in early detection of mucosa lesions.

Breast Cancer Diagnosis Using Extended-Wavelength-Diffuse Reflectance Spectroscopy (EW-DRS)-Proof of Concept in Ex Vivo Breast Specimens Using Machine Learning

This study aims to investigate the feasibility of using diffuse reflectance spectroscopy (DRS) to distinguish malignant breast tissue from adjacent healthy tissue, and to evaluate if an extended-wavelength range (450-1550 nm) has an advantage over the standard wavelength range (450-900 nm). Multivariate statistics and machine learning algorithms, either linear discriminant analysis (LDA) or support vector machine (SVM) are used to distinguish the two tissue types in breast specimens (total or partial mastectomy) from 23 female patients with primary breast cancer. EW-DRS has a sensitivity of 94% and specificity of 91% as compared to a sensitivity of 40% and specificity of 71% using the standard wavelength range. The results suggest that DRS can discriminate between malignant and healthy breast tissue, with improved outcomes using an extended wavelength. It is also possible to construct a simple analytical model to improve the diagnostic performance of the DRS technique.

Advancing cervical cancer diagnosis and screening with spectroscopy and machine learning

INTRODUCTION

In the UK alone, the incidence of cervical cancer is increasing, hence an urgent need for early and rapid detection of cancer before it develops. Spectroscopy in conjunction with machine learning offers a disruptive technology that promises to be pick up cancer early as compared to the current diagnostic techniques used.

AREAS COVERED

This review article explores the different spectroscopy techniques that have been used for the analysis of cervical cancer. Along with the extensive description of spectroscopic techniques, the various machine learning techniques are also described as well as the applications that have been explored in the diagnosis of cervical cancer. This review delimits the literature specifically associated with cervical cancer studies performed solely with the use of a spectroscopy technique, and machine learning.

EXPERT OPINION

Although there are several methods and techniques to detect cervical cancer, the clinical sector requires to introduce new diagnostic technologies that help improving the quality of life of patient. These innovative technologies involve spectroscopy as a qualitative method and machine learning as a quantitative method. In this article, both the techniques and methodologies that allow and promise to be a new screening tool for the detection of cervical cancer is covered.

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.

Quantification of soft tissue parameters from spatially resolved diffuse reflectance finite element models

Spatially resolved diffuse reflectance spectroscopy (SRDRS) is a non-invasive optical technique that helps in clinical diagnosis of various tissue microcirculation and skin pigmentation disorders based on collected backscattered light from multi-layered tissue. The extraction of the optical properties from the reflectance spectrum using analytical solutions is laborious. Model-based light tissue interaction studies help in quantifying the optical properties. This work presents the use of finite element models of light tissue interaction for this purpose. A bilayer model mimicking human skin was considered and the diffused reflectance spectra at multiple detector points were generated using finite element modelling for varying melanin concentration, epidermal thickness, blood volume fraction, oxygen saturation and scattering components. The reflectance value based on varying optical parameters from multiple detection points lead to the generation of a look-up table (LUT), which is further used for finding the tissue parameters that contribute to the spatially resolved reflectance values. The tissue parameters estimated after inverse modelling showed a high degree of agreement with the expected tissue parameters for a test dataset different from the training dataset.

Wavelength weightings in machine learning for ovine joint tissue differentiation using diffuse reflectance spectroscopy (DRS)

Objective: To investigate the DRS of ovine joint tissue to determine the optimal optical wavelengths for tissue differentiation and relate these wavelengths to the biomolecular composition of tissues. In this study, we combine machine learning with DRS for tissue classification and then look further at the weighting matrix of the classifier to further understand the key differentiating features. Methods: Supervised machine learning was used to analyse DRS data. After normalising the data, dimension reduction was achieved through multiclass Fisher's linear discriminant analysis (Multiclass FLDA) and classified with linear discriminant analysis (LDA). The classifier was first run with all the tissue types and the wavelength range 190 nm - 1081 nm. We analysed the weighting matrix of the classifier and then ran the classifier again, the first time using the ten highest weighted wavelengths and the second using only the single highest. Our method was applied to a dataset containing ovine joint tissue including cartilage, cortical and subchondral bone, fat, ligament, meniscus, and muscle. Results: It achieved a classification accuracy of 100% using the wavelength 190 nm - 1081 nm (2048 attributes) with an accuracy of 90% being present for 10 attributes with the exception of those with comparable compositions such as ligament and meniscus. An accuracy greater than 70% was achieved using a single wavelength, with the same exceptions. Conclusion: Multiclass FLDA combined with LDA is a viable technique for tissue identification from DRS data. The majority of differentiating features existed within the wavelength ranges 370-470 and 800-1010 nm. Focusing on key spectral regions means that a spectrometer with a narrower range can potentially be used, with less computational power needed for subsequent analysis.

Reflectance spectral analysis for novel characterization and clinical assessment of aspirated coronary thrombi in patients with ST elevation myocardial infarction

OBJECTIVE

The visual appearance of coronary thrombi may be clinically informative in ST elevation myocardial infarction (STEMI) patients undergoing primary percutaneous coronary intervention (pPCI). However, subjective assessment is poorly reproducible and cannot provide an objective basis for treatment decisions or patient stratification. We have assessed the feasibility of a novel reflectance spectroscopy technique to systematically characterize coronary artery thrombi retrieved by aspiration during pPCI in patients with STEMI, and the clinical utility for predicting distal microvascular obstruction.

APPROACH

Patients with STEMI treated with pPCI and thrombus aspiration (n = 288) were recruited from the Oxford Acute Myocardial Infarction (OxAMI) Study. Of these, 158 patients underwent cardiac magnetic resonance imaging within 48 h for assessment of microvascular obstruction (MVO). Coronary thrombi were imaged by reflectance spectroscopy across wavelengths 500-800 nm.

MAIN RESULTS

Spectral data were analysed using function fitting and multivariate models. The coefficient 'c _red' determined from the fitting procedure correlated with the visually-assessed colour of thrombi ('red' or 'white') and with MVO. When applied to a reduced data set, consisting of spectra from 20 patients with the largest MVO and from 20 propensity-score-matched patients with no MVO, three multivariate analysis methods were able to discriminate spectra of thrombi from patients without MVO and with the largest MVO.

SIGNIFICANCE

Reflectance spectral analysis of coronary thrombus provides new insights into the pathology of STEMI, with potential clinical implications for emergency patient care. Further studies are warranted for validation as a point-of-care stratification tool in predicting the degree of microvascular injury and clinical outcomes in STEMI.

Label-free, High-Resolution Optical Metabolic Imaging of Human Cervical Precancers Reveals Potential for Intraepithelial Neoplasia Diagnosis

While metabolic changes are considered a cancer hallmark, their assessment has not been incorporated in the detection of early or precancers, when treatment is most effective. Here, we demonstrate that metabolic changes are detected in freshly excised human cervical precancerous tissues using label-free, non-destructive imaging of the entire epithelium. The images rely on two-photon excited fluorescence from two metabolic co-enzymes, NAD(P)H and FAD, and have micron-level resolution, enabling sensitive assessments of the redox ratio and mitochondrial fragmentation, which yield metrics of metabolic function and heterogeneity. Simultaneous characterization of morphological features, such as the depth-dependent variation of the nuclear:cytoplasmic ratio, is demonstrated. Multi-parametric analysis combining several metabolic metrics with morphological ones enhances significantly the diagnostic accuracy of identifying high-grade squamous intraepithelial lesions. Our results motivate the translation of such functional metabolic imaging to in vivo studies, which may enable improved identification of cervical lesions, and other precancers, at the bedside.

Diffuse reflectance spectroscopy-enhanced drill for bone boundary detection

Intramedullary nailing is a routine orthopedic procedure used for treating fractures of femoral or tibial shafts. A critical part of this procedure involves the drilling of pilot holes in both ends of the bone for the placement of the screws that will secure the IM rod to sections of the fractured bone. This step introduces a risk of soft tissue damage because the drill bit, if not stopped in time, can transverse the bone-tissue boundary into the overlying muscle, causing unnecessary injury and prolonging healing time due to periosteum damage. In this respect, detecting the bone-tissue boundary before break-through can reduce the risks and complications associated with intramedullary nailing. Hence, in the present study, a two-wavelength diffuse reflectance spectroscopy technique was integrated into a surgical drill to optically detect bone-tissue boundary and automatically trigger the drill to stop. Furthermore, Monte-Carlo simulations were used to estimate the maximum distance from within the bone at which the bone-tissue boundary could be detected using DRS. The simulation results estimated that the detection distance, termed the "look-ahead-distance" was ∼1.5 mm for 1.3 mm source-detector fiber separation. Experimental measurements with 1.3 mm source-detector fiber separation showed that the look-ahead-distance was in the order of 250 µm in experiments with set drill rate and in the range of 1 mm in experiments where the holes were drilled by hand. Despite this difference, the automated DRS enhanced drill successfully detected the approaching bone tissue boundary when tested on samples of bovine femur and muscle tissue.

Optical techniques for the diagnosis and treatment of lesions induced by the human papillomavirus - A resource letter

Human papillomaviruses (HPV) are the most common sexually-transmitted virus, and carcinogenic HPV strains are reported to be responsible for virtually all cases of cervical cancer and its precursor, the cervical intraepithelial neoplasia (CIN). About 30% of the sexually active population are considered to be affected by HPV. Around 600 million people are estimated to be infected worldwide. Diseases related to HPV cause significant impact from both the personal welfare point of view and public healthcare perspective. This resource letter collects relevant information regarding HPV-induced lesions and discusses both diagnosis and treatment, with particular attention to optical techniques and the challenges involved to the implementation of those approaches.

Intrinsic fluorescence for cervical precancer detection using polarized light based in-house fabricated portable device

An in-house fabricated portable device has been tested to detect cervical precancer through the intrinsic fluorescence from human cervix of the whole uterus in a clinical setting. A previously validated technique based on simultaneously acquired polarized fluorescence and polarized elastic scattering spectra from a turbid medium is used to extract the intrinsic fluorescence. Using a diode laser at 405 nm, intrinsic fluorescence of flavin adenine dinucleotide, which is the dominant fluorophore and other contributing fluorophores in the epithelium of cervical tissue, has been extracted. Different grades of cervical precancer (cervical intraepithelial neoplasia; CIN) have been discriminated using principal component analysis-based Mahalanobis distance and linear discriminant analysis. Normal, CIN I and CIN II samples have been discriminated from one another with high sensitivity and specificity at 95% confidence level. This ex vivo study with cervix of whole uterus samples immediately after hysterectomy in a clinical environment indicates that the in-house fabricated portable device has the potential to be used as a screening tool for in vivo precancer detection using intrinsic fluorescence.

Optical techniques for cervical neoplasia detection

This paper provides an overview of the current research in the field of optical techniques for cervical neoplasia detection and covers a wide range of the existing and emerging technologies. Using colposcopy, a visual inspection of the uterine cervix with a colposcope (a binocular microscope with 3- to 15-fold magnification), has proven to be an efficient approach for the detection of invasive cancer. Nevertheless, the development of a reliable and cost-effective technique for the identification of precancerous lesions, confined to the epithelium (cervical intraepithelial neoplasia) still remains a challenging problem. It is known that even at early stages the neoplastic transformations of cervical tissue induce complex changes and modify both structural and biochemical properties of tissues. The different methods, including spectroscopic (diffuse reflectance spectroscopy, induced fluorescence and autofluorescence spectroscopy, Raman spectroscopy) and imaging techniques (confocal microscopy, optical coherence tomography, Mueller matrix imaging polarimetry, photoacoustic imaging), probe different tissue properties that may serve as optical biomarkers for diagnosis. Both the advantages and drawbacks of these techniques for the diagnosis of cervical precancerous lesions are discussed and compared.