Tissue Classification Research Articles

An efficient method for MRI brain tumor tissue segmentation and classification using an optimized support vector machine

An efficient method for MRI brain tumor tissue segmentation and classification using an optimized support vector machine

Tissue science and histology in the system of biomedical scientific and educational disciplines

In this article, the authors analyze the role and significance of histology in biomedical scientific and educational disciplines (history, present condition, and controversial issues). This study discusses the formation and development of histology—the science of tissues—and the emergence of notions about tissues and tissue classification. When describing the microscopic structure of plant organs in 1671, the English botanist and physician N. Grew (1641–1712) used the term “tissue” for the first time. In the mid-19th century, German histologists R.A. Kölliker (1817–1905) and F. Leydig (1821–1908) established the present scientific classification of tissues. These authors classified tissues into four main groups: epithelium, connective tissue and blood, nervous tissue, and muscle tissue. Russian scientists A.A. Zavarzin and N.G. Khlopin made significant contributions to developing tissue classification and evolution problems. It is worth noting that the term “tissue,” which was initially interpreted as purely morphological, has gained physiological content; that is, the idea of “tissue” has become a morphofunctional concept. The idea of tissue stability during the ontogenesis stages of organisms is one of the significant paradigms of histology. Fabric variability is allowed within certain limits within the tissue group to which the fabric belongs. There is no compelling evidence of tissue transition from one tissue group to another. The first histology departments appeared in European higher educational institutions in the middle of the 19th century and Russian higher educational institutions in the late 1960s. As a scientific discipline, histology has not exhausted its capabilities; therefore, excluding histology from the nomenclature of scientific specialization is incorrect.

Histopathology image classification: highlighting the gap between manual analysis and AI automation.

The field of histopathological image analysis has evolved significantly with the advent of digital pathology, leading to the development of automated models capable of classifying tissues and structures within diverse pathological images. Artificial intelligence algorithms, such as convolutional neural networks, have shown remarkable capabilities in pathology image analysis tasks, including tumor identification, metastasis detection, and patient prognosis assessment. However, traditional manual analysis methods have generally shown low accuracy in diagnosing colorectal cancer using histopathological images. This study investigates the use of AI in image classification and image analytics using histopathological images using the histogram of oriented gradients method. The study develops an AI-based architecture for image classification using histopathological images, aiming to achieve high performance with less complexity through specific parameters and layers. In this study, we investigate the complicated state of histopathological image classification, explicitly focusing on categorizing nine distinct tissue types. Our research used open-source multi-centered image datasets that included records of 100.000 non-overlapping images from 86 patients for training and 7180 non-overlapping images from 50 patients for testing. The study compares two distinct approaches, training artificial intelligence-based algorithms and manual machine learning models, to automate tissue classification. This research comprises two primary classification tasks: binary classification, distinguishing between normal and tumor tissues, and multi-classification, encompassing nine tissue types, including adipose, background, debris, stroma, lymphocytes, mucus, smooth muscle, normal colon mucosa, and tumor. Our findings show that artificial intelligence-based systems can achieve 0.91 and 0.97 accuracy in binary and multi-class classifications. In comparison, the histogram of directed gradient features and the Random Forest classifier achieved accuracy rates of 0.75 and 0.44 in binary and multi-class classifications, respectively. Our artificial intelligence-based methods are generalizable, allowing them to be integrated into histopathology diagnostics procedures and improve diagnostic accuracy and efficiency. The CNN model outperforms existing machine learning techniques, demonstrating its potential to improve the precision and effectiveness of histopathology image analysis. This research emphasizes the importance of maintaining data consistency and applying normalization methods during the data preparation stage for analysis. It particularly highlights the potential of artificial intelligence to assess histopathological images.

Saliva Based Diagnostic Prediction of Oral Squamous Cell Carcinoma using FTIR Spectroscopy.

Oral cancer ranks as the sixth most prevalent form of cancer worldwide, presenting a significant public health concern. According to the World Health Organization (WHO), within a 5-year period following diagnosis, the mortality rate among oral cancer patients of all stages stands at 45%. In this study, a total of 60 patients divided into 2 groups were recruited. Group A included 30 histo-pathologically confirmed OSCC patients and Group B included 30 healthy controls. A standardized procedure was followed to collect saliva samples. FTIR spectroscopy was done for all the saliva samples collected from both Group A and B. An IR Prestige-21 (Shimadzu Corp, Japan) spectrometer was used to record IR spectra in the 40-4000cm-1 range SVM classifier was applied in the classification of disease state from normal subjects using FTIR data. The peaks were identified at wave no 1180cm-1, 1230cm-1, 1340cm-1, 1360cm-1, 1420cm-1, 1460cm-1, 1500cm-1, 1540cm-1, 1560cm-1, and 1637cm-1. The observed results of SVM demonstrated the accuracy of 91.66% in the classification of Cancer tissues from the normal subjects with sensitivity of 83.33% while specificity and precision of 100.0%. The development of oral cancer leads to noticeable alterations in the secondary structure of proteins. These findings emphasize the promising use of ATR-FTIR platforms in conjunction with machine learning as a reliable, non-invasive, reagent-free, and highly sensitive method for screening and monitoring individuals with oral cancer.

Genetic algorithm based detection of breast cancer using least square-support vector machine classifier

Breast tumors are a dangerous disease among women worldwide. They are the second leading cause of death among all forms of cancers in women. Their early detection is critical to increasing the survival rate of women. Mammography is a reliable screening technique in the early detection of abnormal breast tissue severity. Radiologist abnormalities in the breast tissue, radiologists employ mammography. However, detecting breast abnormalities through digital diagnostic techniques by a radiologist could be time consuming. Consequently, computerized studying of digital mammography has emerged via the development of CAD systems. Several CAD systems have been developed for breast cancer detection. However, obtaining a satisfactory performance of CAD systems is a challenging task. We propose a CAD architecture for the classification of breast tissues as either benign or malignant using an LS-SVM classifier with various kernels namely linear, quadratic, polynomial, MLP, and RBF kernels. From the experimental outputs, it is clear that GA based LS-SVM classifier with RBF kernel outputs classification accuracy of 94.59% for normal/abnormal case classification is better, when it is compared with all other kernels. It is also stated that GA based LS-SVM classifier with RBF kernel produces a better classification accuracy of 98.26% for benign/malignant case classification when it is compared with other reported works.

Instrument-Tissue Interaction Detection Framework for Surgical Video Understanding.

Instrument-tissue interaction detection task, which helps understand surgical activities, is vital for constructing computer-assisted surgery systems but with many challenges. Firstly, most models represent instrument-tissue interaction in a coarse-grained way which only focuses on classification and lacks the ability to automatically detect instruments and tissues. Secondly, existing works do not fully consider relations between intra- and inter-frame of instruments and tissues. In the paper, we propose to represent instrument-tissue interaction as 〈 instrument class, instrument bounding box, tissue class, tissue bounding box, action class 〉 quintuple and present an Instrument-Tissue Interaction Detection Network (ITIDNet) to detect the quintuple for surgery videos understanding. Specifically, we propose a Snippet Consecutive Feature (SCF) Layer to enhance features by modeling relationships of proposals in the current frame using global context information in the video snippet. We also propose a Spatial Corresponding Attention (SCA) Layer to incorporate features of proposals between adjacent frames through spatial encoding. To reason relationships between instruments and tissues, a Temporal Graph (TG) Layer is proposed with intra-frame connections to exploit relationships between instruments and tissues in the same frame and inter-frame connections to model the temporal information for the same instance. For evaluation, we build a cataract surgery video (PhacoQ) dataset and a cholecystectomy surgery video (CholecQ) dataset. Experimental results demonstrate the promising performance of our model, which outperforms other state-of-the-art models on both datasets.

Fuzzy C Means Clustering Coupled with Firefly Optimization Algorithm for the Segmentation of Neurodisorder Magnetic Resonance Images

Medical image segmentation is a critical task in medical image analysis, and clustering algorithms can be utilized to achieve this goal. This research work focuses on the segmentation of neuro disorder magnetic resonance images using Otsu, K-means, and FCM coupled with the firefly optimization algorithm. Otsu is a classical thresholding algorithm that relies on a single threshold value to segment the images. K-Means is one of the simplest and most widely used clustering algorithms. It aims to partition data into K clusters, where each data point belongs to the cluster with the nearest mean. Fuzzy C-Means is an extension of K-Means, allowing data points to belong to multiple clusters with varying degrees of membership. In medical image segmentation, FCM was used to classify pixels or voxels into different tissue classes with soft boundaries, accounting for partial volume effects. Firefly Optimization helps in improving the convergence speed of the FCM algorithm. Firefly optimization is good at exploring the solution space and finding global optima. The combination of FCM and Firefly Optimization leads to more accurate clustering results. The performance evaluation was done by Renyi entropy and Shannon entropy, FCM coupled with the firefly optimization was found to exhibit superior results when compared with the Otsu and K-means clustering algorithms.

Breast Tissue Classification Method Based on Machine Learning

Abstract: Early detection and treatment of breast cancer are very necessary, and effective classification of breast tissue is helpful for the diagnosis of breast cancer; so, a classification method named FT_GA_GBDT is proposed. First, the correlations between the features and classification labels of breast tissue samples were determined, and features with higher correlation were analyzed statistically and combined by weight. Thus, feature transformation (FT) is realized. The datasets were then enhanced by calculating the mean and root mean square of the feature attributes of each adjacent odd- and even-row sample with both belonging to the same class. Finally, the genetic algorithm (GA) was used to search the optimal parameters of the gradient boosting decision tree (GBDT) model, and the optimal parameters were substituted into the GBDT to classify the breast tissue. In addition, the K-nearest-neighbor (KNN), support-vector-machine (SVM) and GBDT methods were also used to test the breast tissue classification. Results of 6-fold cross validation on three breast tissue datasets showed that the average Precision, Recall, and F1 score obtained by the FT_GA_GBDT method were better than those obtained by the KNN, SVM and GBDT methods. The results further show that the FT algorithm and searching for the optimal hyper-parameters by the GA were helpful in improving the performance of the breast tissue classification model, which is more obvious when the correlations between features and classification labels are generally not high.

Unsupervised Dual Transformer Learning for 3-D Textured Surface Segmentation.

Analysis of the 3-D texture is indispensable for various tasks, such as retrieval, segmentation, classification, and inspection of sculptures, knit fabrics, and biological tissues. A 3-D texture represents a locally repeated surface variation (SV) that is independent of the overall shape of the surface and can be determined using the local neighborhood and its characteristics. Existing methods mostly employ computer vision techniques that analyze a 3-D mesh globally, derive features, and then utilize them for classification or retrieval tasks. While several traditional and learning-based methods have been proposed in the literature, only a few have addressed 3-D texture analysis, and none have considered unsupervised schemes so far. This article proposes an original framework for the unsupervised segmentation of 3-D texture on the mesh manifold. The problem is approached as a binary surface segmentation task, where the mesh surface is partitioned into textured and nontextured regions without prior annotation. The proposed method comprises a mutual transformer-based system consisting of a label generator (LG) and a label cleaner (LC). Both models take geometric image representations of the surface mesh facets and label them as texture or nontexture using an iterative mutual learning scheme. Extensive experiments on three publicly available datasets with diverse texture patterns demonstrate that the proposed framework outperforms standard and state-of-the-art unsupervised techniques and performs reasonably well compared to supervised methods.

Carotid plaque characteristics in the CREST-2 trial

BackgroundThe modest stroke prevention from surgery for asymptomatic carotid disease has prompted a search for predictors that may improve risk stratification beyond luminal stenosis. Plaque disruption and atheroembolization are associated with unique anatomical and histological changes. The Carotid Revascularization and Medical Management for Asymptomatic Carotid Stenosis Trial (CREST-2) collects information on duplex ultrasound (DUS) plaque biomarkers with the goal of evaluating their relationship to the periprocedural and long-term risks of stroke. In this study, we examine the reliability with which carotid plaque features can be measured from DUS images, and report baseline carotid DUS-derived stenosis and plaque features using semiautomated digital image analysis of patients enrolled in CREST-2. MethodsWe studied the first 503 patients in CREST-2. Patients underwent standardized carotid DUS evaluation pre-enrollment. B-Mode images were scaled linearly to normalize brightness. Plaques were outlined manually. Dedicated software automatically measured the longitudinal sectional area (mm2), grayscale median (GSM), Gray-Weale classification, and tissue composition (mm2) of intraplaque hemorrhage (IPH), lipid, fibrous tissue, muscle, and calcium. We present the mean, standard deviation, median, interquartile range, minimum and maximum range, and proportions of carotid peak systolic velocities (PSVs) and plaque morphological features. We tested for autocorrelation among plaque features and computed the proportion of potentially unstable plaques in the cohort. Reliability of the image analysis techniques was tested in 100 patients using Bland-Altman plots and intraclass and interclass correlation coefficients. ResultsMost patients were male (58.4%), older (mean age of 69.3 years), White (87.5%), and had a PSV of ≥230 cm/s (98.6%); the 1.4% with a PSV of <230 cm/s were enrolled based on catheter angiography. Plaques in this study were large; the mean longitudinal sectional area was 62 ± 37 mm2 (range, 6.2–256.5 mm2). The mean GSM was 58 ± 30 (unitless) (range, 0–168) and Gray-Weale classification was 3.5 ± 0.9 (range, 1–5). The mean areas of tissue types were IPH 5.3 ± 8.9 mm2, lipid 9.3 ± 8.6 mm2, fibrous tissue 10 ± 10 mm2, muscle 17 ± 12 mm2, and calcium 1.6 ± 4.1 mm2. The PSV of patients showed poor correlation with plaque features. The proportion of plaques with a GSM of ≤35 was 22.8%, IPH of ≥5 mm2 was 30.0%, and a lipid-rich necrotic core of ≥40% was 3.0% of the cohort. Plaque measurements could be performed with high reliability with good interobserver and intraobserver correlations. ConclusionsSite-generated, core laboratory-interpreted ultrasound examination provides a reliable way of characterizing carotid plaque morphological features across studies performed at many CREST-2 sites. The initial cohort of patients randomized in CREST-2 had heterogeneity of plaque features despite causing high-grade stenosis. Completion of the trial will provide an opportunity to assess whether plaque heterogeneity interacts with response to revascularization and medical management.

Specific of accumulation of manganese in organs and tissues of Hereford cattle.

The elemental status of cattle is one of the important factors, which determine its growth, fertility, fetal development, meat and dairy production, etc. Therefore, the study of content of different elements in cattle organs and tissues and its correlation with cattle characteristics and diet is urgent task. It is also important to develop intravital and low-invasive methods to analyze element content in cattle to regulate its diet during lifetime. In the present work, we have studied the content and distribution of manganese in Hereford cattle from an ecologically clean zone of Western Siberia (Russia). 252 samples were taken from 31 bulls aged 15-18 months. They were collected from various livestock farms in the region and analyzed using atomic absorption spectrophotometry (organs and muscle tissue) and inductively coupled plasma atomic emission spectrometry (hair). The median values of manganese concentration obtained in natural moisture for hair, heart, kidneys, liver, lungs, muscles, spleen, testes, and brain were 25, 0.37, 1.0, 2.6, 0.4, 0.2, 0.4, 0.5, and 0.5 ppm. Accordingly, the concentration of manganese differs significantly in the organs and tissues of animals (H = 188.6, df = 8, p <0.0001). Statistically significant associations of manganese were revealed in pairs: liver-testis, hair-testis, spleen-testis, and heart-brain. The classification of organs and tissues of animals according to the level of content and variability of manganese is carried out. The concentration of manganese in the body is not uniform, most of all it is deposited in the hair and excretory organs of the liver and kidneys. In other organs and muscle tissues, the distribution of manganese is more even and is in the range of 0.2-0.5 ppm. The resulting ranges can be used as a guideline for Hereford cattle bred in Western Siberia.

A comprehensive FTIR micro-spectroscopic analysis and classification of precancerous human oral tissue aided by machine learning

An FTIR micro-spectroscopic approach was able to reveal biochemical alterations in protein spectra (amide I and III areas) in oral precancerous tissues for oral cancer screening.

Analysis of imaging modalities for classification of tumoral vs. normal tissues using an FD-FLIM based MMF endoscopy probe

Current surgical resections for Head and Neck Cancers aim for clear margins to prevent local recurrence. However, up to 20% of cases result in positive margins, with secondary surgery increasing the chances of death after 5 years. We envision a MMF endoscope that collects high resolution images using wavefront shaping to scan a 405 nm beam at the fiber tip and collecting fluorescence intensity and lifetime to map tumor margins and detect residual malignant cells. To address the question whether the information contained in the fluorescence and morphology can be used to classify cancer and normal tissues, we used images acquired with a microscope and artificial neural network. Initial findings show promise to separate cancer from normal tissue when training neural networks on FLIM data. Spatial and temporal resolution and required field of view for effective margin assessment are determined.

Dimension Reduction Using Dual-Featured Auto-Encoder for the Histological Classification of Human Lungs Tissues

Dimension Reduction Using Dual-Featured Auto-Encoder for the Histological Classification of Human Lungs Tissues

Corrections to “Deep-Learning Approach for Tissue Classification Using Acoustic Waves During Ablation With an Er:YAG Laser”

Corrections to “Deep-Learning Approach for Tissue Classification Using Acoustic Waves During Ablation With an Er:YAG Laser”

In vivo corneal confocal microscopy: Technical innovations and frontier applications

At present, the only commercially available system for in vivo Corneal Confocal Microscopy enables unique insights into the corneal morphology, especially corneal nerves. Unfortunately—technology, user‐friendliness and image analysis of the standard platform has remained at the level of 15 years ago. This contribution presents cutting edge insights into new developments based on research within our German Research Foundation supported interdisciplinary research network. This covers a wide range of new developments addressing usability, large field as well as 3D imaging, dedicated contact cap designs, elastic image registration and AI based tissue classification. Furthermore, cutting edge technologies offer 4D imaging to perform dynamic investigations and multiwavelength as well as OCT‐guided Corneal Confocal Microscopy. Finally, key aspects demonstrating the potential of the technique for static and dynamic corneal nerves analysis will be presented.

Transfer Contrastive Learning for Raman Spectroscopy Skin Cancer Tissue Classification.

Using Raman spectroscopy (RS) signals for skin cancer tissue classification has recently drawn significant attention, because of its non-invasive optical technique nature using molecular structures and conformations within biological tissue for diagnosis. In reality, RS signals are noisy and unstable for training machine learning models. The scarcity of tissue samples also makes it challenging to learn reliable deep-learning networks for clinical usages. In this paper, we advocate a Transfer Contrasting Learning Paradigm (TCLP) to address the scarcity and noisy characteristics of the RS for skin cancer tissue classification. To overcome the challenge of limited samples, TCLPleverages transfer learning to pre-train deep learning models using RS data from similar domains (but collected from different RS equipments for other tasks). To tackle the noisy nature of the RS signals, TCLPuses contrastive learning to augment RS signals to learn reliable feature representation to represent RS signals for final classification. Experiments and comparisons, including statistical tests, demonstrate that TCLPoutperforms existing deep learning baselines for RS signal-based skin cancer tissue classification. Both code and data used in our study are available at: https://github.com/yeyimilk/tclp.

Improved bioimpedance spectroscopy tissue classification through data augmentation from generative adversarial networks.

Bioimpedance spectroscopy is a tissue classification technique with many clinical applications. Similarly to other data-driven methods, it requires large amounts of data to accurately distinguish similar classes of tissue. Classifiers trained on small datasets typically suffer from over-fitting and lack the ability to generalise to previously unseen data. However, a large in or ex vivo spectral database is difficult to attain. Data collection is usually limited to studies that occur infrequently, and publicly available data is often not available. A solution to this problem is to artificially increase the training dataset by creating modified, yet accurate, copies of the original dataset. The most common techniques in spectral classification are to add noise to copies of the original data, over-sample it, or randomly interpolate pairs of the original data. However, simply perturbing or interpolating the data does not guarantee that the new dataset captures the key features of the original data needed for accurate classification. This study proposes a novel way to augment bioimpedance spectral data. It uses generative adversarial networks (GAN)-a model in which two neural networks (NN) compete with each other: while one NN artificially manufactures data that could be mistaken for real data, the role of the second NN is to identify which data it receives has been artificially created. The first NN then interactively adapts its output until the second NN can no longer flag artificially created data. The result is a new dataset that truly represents the features of the original data. In this study, three GAN architectures are used, i.e., the vanilla GAN, the deep convolutional GAN, and the Wasserstein GAN. Then, the generated data is used to train five classification methods, and their results are compared to a baseline that only uses the original data. The results from a dataset of 13 different tissue classes show that the deep convolutional GAN is most statistically similar to the original data and improves classification accuracy by 15% when compared to the same model trained only on the original data. The Wasserstein-GAN architecture also provides significant improvements of up to 24% better accuracy.

Opto-Lipidomics of Tissues.

Lipid metabolism and signaling play pivotal functions in biology and disease development. Despite this, currently available optical techniques are limited in their ability to directly visualize the lipidome in tissues. In this study, opto-lipidomics, a new approach to optical molecular tissue imaging is introduced. The capability of vibrational Raman spectroscopy is expanded to identify individual lipids in complex tissue matrices through correlation with desorption electrospray ionization (DESI) - mass spectrometry (MS) imaging in an integrated instrument. A computational pipeline of inter-modality analysis is established to infer lipidomic information from optical vibrational spectra. Opto-lipidomic imaging of transient cerebral ischemia-reperfusion injury in a murine model of ischemic stroke demonstrates the visualization and identification of lipids in disease with high molecular specificity using Raman scattered light. Furthermore, opto-lipidomics in a handheld fiber-optic Raman probe is deployed and demonstrates real-time classification of bulk brain tissues based on specific lipid abundances. Opto-lipidomics opens a host of new opportunities to study lipid biomarkers for diagnostics, prognostics, and novel therapeutic targets.

An artificial intelligence model based on transrectal ultrasound images of biopsy needle tract tissues to differentiate prostate cancer.

We aimed to develop an artificial intelligence (AI) model based on transrectal ultrasonography (TRUS) images of biopsy needle tract (BNT) tissues for predicting prostate cancer (PCa) and to compare the PCa diagnostic performance of the radiologist model and clinical model. A total of 1696 2D prostate TRUS images were involved from 142 patients between July 2021 and May 2022. The ResNet50 network model was utilized to train classification models with different input methods: original image (Whole model), BNT (Needle model), and combined image [Feature Pyramid Networks (FPN) model]. The training set, validation set, and test set were randomly assigned, then randomized 5-fold cross-validation between the training set and validation set was performed. The diagnostic effectiveness of AI models and image combination was accessed by an independent testing set. Then, the optimal AI model and image combination were selected to compare the diagnostic efficacy with that of senior radiologists and the clinical model. In the test set, the area under the curve, specificity, and sensitivity of the FPN model were 0.934, 0.966, and 0.829, respectively; the diagnostic efficacy was improved compared with the Whole and Needle models, with statistically significant differences (P < 0.05), and was better than that of senior radiologists (area under the curve: 0.667). The FPN model detected more PCa compared with senior physicians (82.9% vs. 55.8%), with a 61.3% decrease in the false-positive rate and a 23.2% increase in overall accuracy (0.887 vs. 0.655). The proposed FPN model can offer a new method for prostate tissue classification, improve the diagnostic performance, and may be a helpful tool to guide prostate biopsy.