Architectural analysis of oral cancer, dysplastic, and normal epithelia

Application and Performance of Artificial Intelligence (AI) in Oral Cancer Diagnosis and Prediction Using Histopathological Images: A Systematic Review

Oral cancer (OC) is one of the most common forms of head and neck cancer and continues to have the lowest survival rates worldwide, even with advancements in research and therapy. The prognosis of OC has not significantly improved in recent years, presenting a persistent challenge in the biomedical field. In the field of oncology, artificial intelligence (AI) has seen rapid development, with notable successes being reported in recent times. This systematic review aimed to critically appraise the available evidence regarding the utilization of AI in the diagnosis, classification, and prediction of oral cancer (OC) using histopathological images. An electronic search of several databases, including PubMed, Scopus, Embase, the Cochrane Library, Web of Science, Google Scholar, and the Saudi Digital Library, was conducted for articles published between January 2000 and January 2023. Nineteen articles that met the inclusion criteria were then subjected to critical analysis utilizing QUADAS-2, and the certainty of the evidence was assessed using the GRADE approach. AI models have been widely applied in diagnosing oral cancer, differentiating normal and malignant regions, predicting the survival of OC patients, and grading OC. The AI models used in these studies displayed an accuracy in a range from 89.47% to 100%, sensitivity from 97.76% to 99.26%, and specificity ranging from 92% to 99.42%. The models' abilities to diagnose, classify, and predict the occurrence of OC outperform existing clinical approaches. This demonstrates the potential for AI to deliver a superior level of precision and accuracy, helping pathologists significantly improve their diagnostic outcomes and reduce the probability of errors. Considering these advantages, regulatory bodies and policymakers should expedite the process of approval and marketing of these products for application in clinical scenarios.

Automated detection of premalignant oral lesions on whole slide images using convolutional neural networks

INTRODUCTION

Oral epithelial dysplasia (OED) is a precursor lesion to oral squamous cell carcinoma, a disease with a reported overall survival rate of 56 percent across all stages. Accurate detection of OED is critical as progression to oral cancer can be impeded with complete excision of premalignant lesions. However, previous research has demonstrated that the task of grading of OED, even when performed by highly trained experts, is subject to high rates of reader variability and misdiagnosis. Thus, our study aims to develop a convolutional neural network (CNN) model that can identify regions suspicious for OED whole-slide pathology images.

METHODS

During model development, we optimized key training hyperparameters including loss function on 112 pathologist annotated cases between the training and validation sets. Then, we compared OED segmentation and classification metrics between two well-established CNN architectures for medical imaging, DeepLabv3+ and UNet++. To further assess generalizability, we assessed case-level performance of a held-out test set of 44 whole-slide images.

RESULTS

DeepLabv3+ outperformed UNet++ in overall accuracy, precision, and segmentation metrics in a 4-fold cross validation study. When applied to the held-out test set, our best performing DeepLabv3+ model achieved an overall accuracy and F1-Score of 93.3 percent and 90.9 percent, respectively.

CONCLUSION

The present study trained and implemented a CNN-based deep learning model for identification and segmentation of oral epithelial dysplasia (OED) with reasonable success. Computer assisted detection was shown to be feasible in detecting premalignant/precancerous oral lesions, laying groundwork for eventual clinical implementation.

Role of artificial intelligence in diagnostic oral pathology-A modern approach

Over the decades, new equipment was emerged in medical field, and we have witnessed the importance of medical imaging such as computed tomography, magnetic resonance imaging, ultrasound, mammography and X-ray and their contribution in successful diagnosis and treatment of various diseases. Now, we are in era of artificial intelligence (AI), where machines were modeled after human brain's ability to take inputs and produce outputs from given data. AI has a wide range of uses and applications in health services industry. Factors such as increase in workload, complexity of work and potential fatigue of doctors may compromise diagnostic ability and outcome. AI components in imaging machines would reduce this workload and drive greater efficiency. They also have access to a greater wealth of data than human counterparts and can detect cancer with more accuracy than humans. This study presented an overview of AI, its recent advances in pathology and future prospects.

Automated oral squamous cell carcinoma identification using shape, texture and color features of whole image strips

Despite profound knowledge of the incidence of oral cancers and a large body of research beyond it, it continues to beat diagnosis and treatment management. Post physical observation by clinicians, a biopsy is a gold standard for accurate detection of any abnormalities. Towards the application of artificial intelligence as an aid to diagnosis, automated cell nuclei segmentation is the most essential step for the recognition of the cancer cells. In this study, we have extracted the shape, texture and color features from the histopathological images collected indigenously from regional hospitals. A dataset of 42 whole slide slices was used to automatically segment and generate a cell level dataset of 720 nuclei. Next, different classifiers were applied for classification purposes. 99.4 % accuracy using Decision Tree Classifier, 100 % accuracy using both SVM and Logistic regression and 100 % accuracy using SVM, Logistic regression and Linear Discriminant were acquired for shape, textural and color features respectively. The in-depth analysis showed SVM and Linear Discriminant classifier gave the best result for texture and color features respectively. The achieved result can be effectively converted to software as an assistant diagnostic tool.

Unsupervised morphological segmentation of tissue compartments in histopathological images

Algorithmic segmentation of histologically relevant regions of tissues in digitized histopathological images is a critical step towards computer-assisted diagnosis and analysis. For example, automatic identification of epithelial and stromal tissues in images is important for spatial localisation and guidance in the analysis and characterisation of tumour micro-environment. Current segmentation approaches are based on supervised methods, which require extensive training data from high quality, manually annotated images. This is often difficult and costly to obtain. This paper presents an alternative data-independent framework based on unsupervised segmentation of oropharyngeal cancer tissue micro-arrays (TMAs). An automated segmentation algorithm based on mathematical morphology is first applied to light microscopy images stained with haematoxylin and eosin. This partitions the image into multiple binary ‘virtual-cells’, each enclosing a potential ‘nucleus’ (dark basins in the haematoxylin absorbance image). Colour and morphology measurements obtained from these virtual-cells as well as their enclosed nuclei are input into an advanced unsupervised learning model for the identification of epithelium and stromal tissues. Here we exploit two Consensus Clustering (CC) algorithms for the unsupervised recognition of tissue compartments, that consider the consensual opinion of a group of individual clustering algorithms. Unlike most unsupervised segmentation analyses, which depend on a single clustering method, the CC learning models allow for more robust and stable detection of tissue regions. The proposed framework performance has been evaluated on fifty-five hand-annotated tissue images of oropharyngeal tissues. Qualitative and quantitative results of the proposed segmentation algorithm compare favourably with eight popular tissue segmentation strategies. Furthermore, the unsupervised results obtained here outperform those obtained with individual clustering algorithms.

Nuclear morphometry and ploidy of normal and neoplastic haemocytes in mussels

Haemic neoplasia (HN) in bivalves has been reported in association with mass mortality events in various species of molluscs. The aim of this work was to quantify the nuclear morphometry and DNA content of neoplastic cells of mussels Mytilus galloprovincialis affected by HN using nuclear densitometry in Feulgen-stained preparations. The results were also compared with a population of normal mussel haemocytes. We captured 256 images of 3 different neoplasia stages and 120 images of normal haemocytes; thus, a total of 120,166 nuclei were analysed. We extracted 21 morphological parameters from normal and neoplastic nuclei. Eighteen of these parameters were different (P<0.05). Among those (expressed in pixel units—inter-pixel distance of 0.45 micrometres—as: normal vs. neoplastic) nuclear area (117.1±94.1 vs. 423.1±226.9), perimeter (44.9±14.0 vs. 79.0±21.3) and (IOD) integrated optical density (13.47±34.5 vs. 177.1±150.8) were relevant features to discriminate between normal and neoplastic cells. Those differences allowed identifying two distinctive populations of neoplastic nuclei, occasionally in the same individuals at a given phase of the disease. Moreover, neoplastic haemocytes in less extended lesions showed a ploidy value of 6.2 n along with the presence of a second population of circulating cells with a DNA content of 10.7n. In samples with moderate disease only one peak at 7n was observed. Finally, in more severe conditions, a further ploidy peak of 7.8n was recorded, accompanied by a shallow but broad peak of 31n. This latter extreme value is thought to be due to the presence of giant multinucleated cells where individual nuclei overlap in space and cannot be discerned individually. Computer-based imaging allowed the direct visualization of the cell populations and simultaneous collection of ploidy data as well as morphological features of nuclei.

Morphological classification of odontogenic keratocysts using Bouligand-Minkowski fractal descriptors

The Odontogenic keratocyst (OKC) is a cystic lesion of the jaws, which has high growth and recurrence rates compared to other cysts of the jaws (for instance, radicular cyst, which is the most common jaw cyst type). For this reason OKCs are considered by some to be benign neoplasms. There exist two sub-types of OKCs (sporadic and syndromic) and the ability to discriminate between these sub-types, as well as other jaw cysts, is an important task in terms of disease diagnosis and prognosis. With the development of digital pathology, computational algorithms have become central to addressing this type of problem. Considering that only basic feature-based methods have been investigated in this problem before, we propose to use a different approach (the Bouligand-Minkowski descriptors) to assess the success rates achieved on the classification of a database of histological images of the epithelial lining of these cysts. This does not require the level of abstraction necessary to extract histologically-relevant features and therefore has the potential of being more robust than previous approaches. The descriptors were obtained by mapping pixel intensities into a three dimensional cloud of points in discrete space and applying morphological dilations with spheres of increasing radii. The descriptors were computed from the volume of the dilated set and submitted to a machine learning algorithm to classify the samples into diagnostic groups. This approach was capable of discriminating between OKCs and radicular cysts in 98% of images (100% of cases) and between the two sub-types of OKCs in 68% of images (71% of cases). These results improve over previously reported classification rates reported elsewhere and suggest that Bouligand-Minkowski descriptors are useful features to be used in histopathological images of these cysts.

Quantitative Diagnosis of Tongue Cancer from Histological Images in an Animal Model

We developed a chemically-induced oral cancer animal model and a computer aided method for tongue cancer diagnosis. The animal model allows us to monitor the progress of the lesions over time. Tongue tissue dissected from mice was sent for histological processing. Representative areas of hematoxylin and eosin (H&E) stained tissue from tongue sections were captured for classifying tumor and non-tumor tissue. The image set used in this paper consisted of 214 color images (114 tumor and 100 normal tissue samples). A total of 738 color, texture, morphometry and topology features were extracted from the histological images. The combination of image features from epithelium tissue and its constituent nuclei and cytoplasm has been demonstrated to improve the classification results. With ten iteration nested cross validation, the method achieved an average sensitivity of 96.5% and a specificity of 99% for tongue cancer detection. The next step of this research is to apply this approach to human tissue for computer aided diagnosis of tongue cancer.

Computational analysis of p63(+) nuclei distribution pattern by graph theoretic approach in an oral pre-cancer (sub-mucous fibrosis)

BACKGROUND

Oral submucous fibrosis (OSF) is a pre-cancerous condition with features of chronic, inflammatory and progressive sub-epithelial fibrotic disorder of the buccal mucosa. In this study, malignant potentiality of OSF has been assessed by quantification of immunohistochemical expression of epithelial prime regulator-p63 molecule in correlation to its malignant (oral squamous cell carcinoma [OSCC] and normal counterpart [normal oral mucosa [NOM]). Attributes of spatial extent and distribution of p63(+) expression in the epithelium have been investigated. Further, a correlated assessment of histopathological attributes inferred from H&E staining and their mathematical counterparts (molecular pathology of p63) have been proposed. The suggested analytical framework envisaged standardization of the immunohistochemistry evaluation procedure for the molecular marker, using computer-aided image analysis, toward enhancing its prognostic value.

SUBJECTS AND METHODS

In histopathologically confirmed OSF, OSCC and NOM tissue sections, p63(+) nuclei were localized and segmented by identifying regional maxima in plateau-like intensity spatial profiles of nuclei. The clustered nuclei were localized and segmented by identifying concave points in the morphometry and by marker-controlled watersheds. Voronoi tessellations were constructed around nuclei centroids and mean values of spatial-relation metrics such as tessellation area, tessellation perimeter, roundness factor and disorder of the area were extracted. Morphology and extent of expression are characterized by area, diameter, perimeter, compactness, eccentricity and density, fraction of p63(+) expression and expression distance of p63(+) nuclei.

RESULTS

Correlative framework between histopathological features characterizing malignant potentiality and their quantitative p63 counterparts was developed. Statistical analyses of mathematical trends were evaluated between different biologically relevant combinations: (i) NOM to oral submucous fibrosis without dysplasia (OSFWT) (ii) NOM to oral submucous fibrosis with dysplasia (OSFWD) (iii) OSFWT-OSFWD (iv) OSFWD-OSCC. Significant histopathogical correlates and their corroborative mathematical features, inferred from p63 staining, were also investigated into.

CONCLUSION

Quantitative assessment and correlative analysis identified mathematical features related to hyperplasia, cellular stratification, differentiation and maturation, shape and size, nuclear crowding and nucleocytoplasmic ratio. It is envisaged that this approach for analyzing the p63 expression and its distribution pattern may help to establish it as a quantitative bio-marker to predict the malignant potentiality and progression. The proposed work would be a value addition to the gold standard by incorporating an observer-independent framework for the associated molecular pathology.

Discrete Mereotopology for Spatial Reasoning in Automated Histological Image Analysis

Discrete mereotopology (DM) is a first-order spatial logic that fuses together mereology (the theory of parthood relations) and topology to model discrete space. We show how a set of quasitopological functions defined within DM can be mapped to specific operators defined in mathematical morphology (MM) and easily implemented in scientific image processing programs. These functions provide the means to model topological properties of individual regions and spatial relations between them such as contact, overlap, and the relation of part to whole. DM not only extends the expressive power of image processing applications where mathematical morphology is used, but by functioning as a logic it also supplies the formal basis with which to prove the correctness of implemented algorithms as well as providing the computational basis to mechanically reason about segmented digital images using automated reasoning programs. In particular, we show how DM can supply a model-based and algorithmic context to the otherwise blind pixel-based image processing routines still dominating conventional imaging approaches. A number of worked examples drawn from the histological domain are given, including segmentation of cells in culture, identifying basal cell layers from stratified epithelia sections, and cell sorting in blood smears.

Ka-me: a Voronoi image analyzer

Summary: Ka-me is a Voronoi image analyzer that allows users to analyze any image with a convex polygonal tessellation or any spatial point distribution by fitting Voronoi polygons and their dual, Delaunay triangulations, to the pattern. The analytical tools include a variety of graph theoretic and geometric tools that summarize the distribution of the numbers of edges per face, areas, perimeters, angles of Delaunay triangle edges (anglograms), Gabriel graphs, nearest neighbor graphs, minimal spanning trees, Ulam trees, Pitteway tests, circumcircles and convexhulls, as well as spatial statistics (Clark–Evans Nearest Neighborhood and Variance to Mean Ratio) and export functions for standard relationships (Lewis's Law, Desch's Law and Aboav–Weaire Law).

Availability: Ka-me: a Voronoi image analyzer is available as an executable with documentation and sample applications from the BioQUEST Library (http://bioquest.org/downloads/kame_1.0.rar).

Contact:noppadon.khiripet@nectec.or.th

Recent advances in morphological cell image analysis

This paper summarizes the recent advances in image processing methods for morphological cell analysis. The topic of morphological analysis has received much attention with the increasing demands in both bioinformatics and biomedical applications. Among many factors that affect the diagnosis of a disease, morphological cell analysis and statistics have made great contributions to results and effects for a doctor. Morphological cell analysis finds the cellar shape, cellar regularity, classification, statistics, diagnosis, and so forth. In the last 20 years, about 1000 publications have reported the use of morphological cell analysis in biomedical research. Relevant solutions encompass a rather wide application area, such as cell clumps segmentation, morphological characteristics extraction, 3D reconstruction, abnormal cells identification, and statistical analysis. These reports are summarized in this paper to enable easy referral to suitable methods for practical solutions. Representative contributions and future research trends are also addressed.

Automated oral cancer identification using histopathological images: a hybrid feature extraction paradigm

Oral cancer (OC) is the sixth most common cancer in the world. In India it is the most common malignant neoplasm. Histopathological images have widely been used in the differential diagnosis of normal, oral precancerous (oral sub-mucous fibrosis (OSF)) and cancer lesions. However, this technique is limited by subjective interpretations and less accurate diagnosis. The objective of this work is to improve the classification accuracy based on textural features in the development of a computer assisted screening of OSF. The approach introduced here is to grade the histopathological tissue sections into normal, OSF without Dysplasia (OSFWD) and OSF with Dysplasia (OSFD), which would help the oral onco-pathologists to screen the subjects rapidly. The biopsy sections are stained with H&E. The optical density of the pixels in the light microscopic images is recorded and represented as matrix quantized as integers from 0 to 255 for each fundamental color (Red, Green, Blue), resulting in a M×N×3 matrix of integers. Depending on either normal or OSF condition, the image has various granular structures which are self similar patterns at different scales termed "texture". We have extracted these textural changes using Higher Order Spectra (HOS), Local Binary Pattern (LBP), and Laws Texture Energy (LTE) from the histopathological images (normal, OSFWD and OSFD). These feature vectors were fed to five different classifiers: Decision Tree (DT), Sugeno Fuzzy, Gaussian Mixture Model (GMM), K-Nearest Neighbor (K-NN), Radial Basis Probabilistic Neural Network (RBPNN) to select the best classifier. Our results show that combination of texture and HOS features coupled with Fuzzy classifier resulted in 95.7% accuracy, sensitivity and specificity of 94.5% and 98.8% respectively. Finally, we have proposed a novel integrated index called Oral Malignancy Index (OMI) using the HOS, LBP, LTE features, to diagnose benign or malignant tissues using just one number. We hope that this OMI can help the clinicians in making a faster and more objective detection of benign/malignant oral lesions.

Structural markers for normal oral mucosa and oral sub-mucous fibrosis

This article presents a quantitative approach for the characterization of normal oral mucosa (NOM) in respect to thickness and textural properties of its entire epithelial layer. Histological images of oral mucosa depict that both thickness and tissue architecture at cellular and tissue level undergo change, as mucosa converts from normal to precancerous or cancerous state. In this study the thickness and fractal dimension of the mucosal epithelium of NOM and oral sub-mucous fibrosis (OSF) condition have been computed using 83 normal and 29 OSF images of oral mucosa. The result shows significant delineation between NOM and OSF in respect of both the epithelial thickness (in microm) and fractal dimensions. This quantitative characterization of oral epithelium will be of immense help for oral onco-pathologists and researchers to assess the biological nature of normal and diseased (OSF) mucosa with higher accuracy. Moreover, further differential applications may enable them to find out newer accurate quantitative diagnostic procedures to that of the usual histopathological gold standard for the assessment of malignant potentiality.

Losses of 3p14 and 9p21 as shown by fluorescence in situ hybridization are early events in tumorigenesis of oral squamous cell carcinoma and already occur in simple keratosis

Tumorigenesis of oral squamous cell carcinoma (OSCC) has been postulated to represent a multistep process driven by the accumulation of carcinogen-induced genetic changes. Alterations of the 3p14 fragile site containing the fragile histidine triade gene and of the 9p21 tumor suppressor locus containing methylthioadenosine phosphorylase, p16 and p15 characteristically occur in oral leukoplakia, a known precursor of OSCC, and are at present considered to indicate the transition from simple keratosis (hyperplasia) to dysplasia. The aim of the study was to evaluate the occurrence of losses of 3p14 and 9p21 and to evaluate polysomies 3 and 9 in leukoplakias using highly sensitive fluorescence in situ hybridization (FISH) probes. Examining 67 leukoplakias (24 hyperplasias, 33 dysplasias, 10 in situ carcinomas), control tissues of oral mucosa from infants and adults as well as invasive carcinomas and normal epithelia of tumor patients with locus specific FISH probes targeting 3p14 and 9p21, and centromeric probes for chromosomes 3 and 9 we could demonstrate that losses of these sites appeared very early in the tumorigenesis of OSCC and were already present in the great majority of simple keratoses. Polysomy 3 occurring more frequently than polysomy 9 was characteristic of dysplasia and in situ carcinomas and thus seems to follow losses of 3p14 and 9p21 during oral squamous cell carcinogenesis.

Quantitative in vivo microscopy: the return from the 'omics'

The confluence of recent advances in microscopy instrumentation and image analysis, coupled with the widespread use of GFP-like proteins as reporters of gene expression, has opened the door to high-throughput in vivo studies that can provide the morphological and temporal context to the biochemical pathways regulating cell function. We are now able to quantify the concentration and three-dimensional distribution of multiple spectrally resolved GFP-tagged proteins. Using automatic segmentation and tracking we can then measure the dynamics of the processes in which these elements are involved. In this way, parallel studies are feasible where multiple cell colonies treated with drugs or gene expression repressors can be monitored and analyzed to study the dynamics of relevant biological processes.

Quantitative analysis of the epithelial lining architecture in radicular cysts and odontogenic keratocysts

Background

This paper describes a quantitative analysis of the cyst lining architecture in radicular cysts (of inflammatory aetiology) and odontogenic keratocysts (thought to be developmental or neoplastic) including its 2 counterparts: solitary and associated with the Basal Cell Naevus Syndrome (BCNS).

Methods

Epithelial linings from 150 images (from 9 radicular cysts, 13 solitary keratocysts and 8 BCNS keratocysts) were segmented into theoretical cells using a semi-automated partition based on the intensity of the haematoxylin stain which defined exclusive areas relative to each detected nucleus. Various morphometrical parameters were extracted from these "cells" and epithelial layer membership was computed using a systematic clustering routine.

Results

Statistically significant differences were observed across the 3 cyst types both at the morphological and architectural levels of the lining. Case-wise discrimination between radicular cysts and keratocyst was highly accurate (with an error of just 3.3%). However, the odontogenic keratocyst subtypes could not be reliably separated into the original classes, achieving discrimination rates slightly above random allocations (60%).

Conclusion

The methodology presented is able to provide new measures of epithelial architecture and may help to characterise and compare tissue spatial organisation as well as provide useful procedures for automating certain aspects of histopathological diagnosis.