Figure 2 - uploaded by Balaji Ganeshan
Content may be subject to copyright.
The histogram displays the range and frequency of pixel intensity values within the medium filtered (filter value 1.5) image of a CRC lesion (courtesy of Professor Ashley Groves, Institute of Nuclear Medicine, University College Hospital, London, UK).
Source publication
Heterogeneity is a key feature of malignancy associated with adverse tumour biology. Quantifying heterogeneity could provide a useful non-invasive imaging biomarker. Heterogeneity on computed tomography (CT) can be quantified using texture analysis which extracts spatial information from CT images (unenhanced, contrast-enhanced and derived images s...
Context in source publication
Context 1
... than transform-based and structural methods [2931] . There are different order statistics (first-, second-and higher-order statistics) that basically differ in their approach to describing the image grey-level distribution, often dis- played as a histogram. An image of the histogram of a filtered image (after image transformation) is given in Fig. 2. Numerous statistical and probabilistic parameters can be quantified from an image histogram. First-order statistics are based on the probability distributions of individual grey-level pixel values (mean, entro- pyirregularity, uniformityinhomogeneity), whereas second-order is based on the joint probability distribu- tions of pairs ...
Citations
... However, the predictive capability of conventional contrast-enhanced CT examinations for NAC response is limited, making it challenging for clinicians to determine potential responses in patients with LAGC. Radiomics methods can extract a plethora of hidden features from medical images, providing information about tumor microenvironments such as cell density, hypoxia, and microvessel density, significantly expanding the scope of research in medical imaging [27]. Previous researchers have found that traditional handcrafted radiomics signatures can predict the prognosis and NAC response in LAGC patients [28]. ...
Purpose
Developed and validated a deep learning radiomics nomogram using multi-phase contrast-enhanced computed tomography (CECT) images to predict neoadjuvant chemotherapy (NAC) response in locally advanced gastric cancer (LAGC) patients.
Methods
This multi-center study retrospectively included 322 patients diagnosed with gastric cancer from January 2013 to June 2023 at two hospitals. Handcrafted radiomics technique and the EfficientNet V2 neural network were applied to arterial, portal venous, and delayed phase CT images to extract two-dimensional handcrafted and deep learning features. A nomogram model was built by integrating the handcrafted signature, the deep learning signature, with clinical features. Discriminative ability was assessed using the receiver operating characteristics (ROC) curve and the precision-recall (P-R) curve. Model fitting was evaluated using calibration curves, and clinical utility was assessed through decision curve analysis (DCA).
Results
The nomogram exhibited excellent performance. The area under the ROC curve (AUC) was 0.848 [95% confidence interval (CI), 0.793–0.893)], 0.802 (95% CI 0.688–0.889), and 0.751 (95% CI 0.652–0.833) for the training, internal validation, and external validation sets, respectively. The AUCs of the P-R curves were 0.838 (95% CI 0.756–0.895), 0.541 (95% CI 0.329–0.740), and 0.556 (95% CI 0.376–0.722) for the corresponding sets. The nomogram outperformed the clinical model and handcrafted signature across all sets (all P < 0.05). The nomogram model demonstrated good calibration and provided greater net benefit within the relevant threshold range compared to other models.
Conclusion
This study created a deep learning nomogram using CECT images and clinical data to predict NAC response in LAGC patients undergoing surgical resection, offering personalized treatment insights.
Graphical abstract
... 11 For instance, CAD has been utilized to measure the heterogeneity of lesion, 12 and it has been repeatedly reported that heterogeneity within the mass is correlated with more malignant features. 13 Hounsfield unit (HU) is a measure of linear x-ray attenuation within the tissue, normalized to the linear attenuation coefficient of water, which is universally utilized in mapping CT scan grayscale images. For the first time, Sieghorn et al. proposed that in the absence of calcification, HU measurement can be utilized in differentiating benign pulmonary nodules from malignant ones. ...
Introduction
Chest radiograph and computed tomography (CT) scans can accidentally reveal pulmonary nodules. Malignant and benign pulmonary nodules can be difficult to distinguish without specific imaging features, such as calcification, necrosis, and contrast enhancement. However, these lesions may exhibit different image texture characteristics which cannot be assessed visually. Thus, a computer‐assisted quantitative method like histogram analysis (HA) of Hounsfield unit (HU) values can improve diagnostic accuracy, reducing the need for invasive biopsy.
Methods
In this exploratory control study, nonenhanced chest CT images of 20 patients with benign (10) and cancerous (10) lesion were selected retrospectively. The appearances of benign and malignant lesions were very similar in chest CT images, and only pathology report was used to discriminate them. Free hand region of interest (ROI) was inserted inside the lesion for all slices of each lesion. Mean, minimum, maximum, and standard deviations of HU values were recorded and used to make HA.
Results
HA showed that the most malignant lesions have a mean HU value between 30 and 50, a maximum HU less than 150, and a minimum HU between −30 and 20. Lesions outside these ranges were mostly benign.
Conclusion
Quantitative CT analysis may differentiate malignant from benign lesions without specific malignancy patterns on unenhanced chest CT image.
... The ability of CT to accurately determine tumor thickness and identify nodal metastases holds the potential for refining treatment strategies and optimizing patient management. 2 However, the literature on the correlation between CT findings and histopathological staging in the context of oral cavity malignancies in the Central Indian population is limited. ...
Background Oral cavity malignancies pose a significant health burden, necessitating accurate diagnostic tools for optimal treatment planning. This cross-sectional study investigates the role of Computed Tomography (CT) in evaluating oral cavity malignancies, focusing on tumor thickness and nodal staging. The study is conducted in a tertiary care hospital in Central India, aiming to enhance diagnostic precision in this region. Methods A prospective, cross-sectional design is employed over two years. Patients with clinical or biopsy-proven oral cavity carcinoma are included after obtaining informed consent. CT scans are performed using a standardized protocol, and imaging findings are correlated with histopathological staging. Statistical analyses, including descriptive statistics, bivariate analysis, correlation analysis, and multivariable modeling, are conducted using SPSS version 27.0. Expected Outcome Anticipated outcomes include a comprehensive understanding of the accuracy of CT in evaluating tumor thickness and nodal stage in oral cavity malignancies. The study aims to delineate the extent and depth of soft tissue and bony invasion, assess nodal metastases, and correlate radiologic findings with histopathological results. The expected findings will contribute valuable insights into the utility of CT in the clinical management of oral cavity malignancies, potentially influencing treatment decisions and improving patient outcomes.
... A model for early prediction of acute pancreatitis severity, based on Computed Tomography (CT) radiomics analyzing pancreatic and peripancreatic alterations, was developed in a prior study [2]. In this paper, we extend our investigation beyond the previous screening and modeling of radiomics features of AP to delve into the visual characteristics of the images post-application of Laplacian of Gaussian (LoG), logarithmic, exponential, square, and gradient filters [3][4][5]. ...
... Studying tumour compositions and microenvironments are of particular interest within cancer research, with the understanding that tumour heterogeneity plays a very important role in treatment outcomes, disease progression, metastasis, and/or recurrence [18][19][20][21]. Understanding that genomic heterogeneity could translate to heterogeneous tumour metabolism and eventually anatomy, radiomics analysis presents a hypothetically feasible quantitative signature profiling method. ...
(1) Background: Some cancer patients do not experience tumour shrinkage but are still at risk of experiencing unwanted treatment side effects. Radiomics refers to mining biomedical images to quantify textural characterization. When radiomics features are labelled with treatment response, retrospectively, they can train predictive machine learning (ML) models. (2) Methods: Radiomics features were determined from lymph node (LN) segmentations from treatment-planning CT scans of head and neck (H&N) cancer patients. Binary treatment outcomes (complete response versus partial or no response) and radiomics features for n = 71 patients were used to train support vector machine (SVM) and k-nearest neighbour (k-NN) classifier models with 1–7 features. A deep texture analysis (DTA) methodology was proposed and evaluated for second- and third-layer radiomics features, and models were evaluated based on common metrics (sensitivity (%Sn), specificity (%Sp), accuracy (%Acc), precision (%Prec), and balanced accuracy (%Bal Acc)). (3) Results: Models created with both classifiers were found to be able to predict treatment response, and the results suggest that the inclusion of deeper layer features enhanced model performance. The best model was a seven-feature multivariable k-NN model trained using features from three layers deep of texture features with %Sn = 74%, %Sp = 68%, %Acc = 72%, %Prec = 81%, %Bal Acc = 71% and with an area under the curve (AUC) the receiver operating characteristic (ROC) of 0.700. (4) Conclusions: H&N Cancer patient treatment-planning CT scans and LN segmentations contain phenotypic information regarding treatment response, and the proposed DTA methodology can improve model performance by enhancing feature sets and is worth consideration in future radiomics studies.
... 11,12 Radiomics technology enables the assessment of tumor image heterogeneity acquired during routine clinical practice. 13,14 By extracting and evaluating features from digital images, it facilitates the detection of subtle changes and heterogeneity that may be imperceptible to the naked eye, thereby offering a novel approach for tumor pathological grade diagnosis. 15 The utilization of CT images for the extraction of radiomics features has been demonstrated to differentiate the pathological grade of renal cell carcinoma effectively. ...
Objectives
Development and validation of a computed tomography urography (CTU)‐based machine learning (ML) model for prediction of preoperative pathology grade of upper urinary tract urothelial carcinoma (UTUC).
Methods
A total of 140 patients with UTUC who underwent CTU examination from January 2017 to August 2023 were retrospectively enrolled. Tumor lesions on the unenhanced, medullary, and excretory periods of CTU were used to extract Features, respectively. Feature selection was screened by the Pearson and Spearman correlation analysis, least absolute shrinkage and selection operator algorithm, random forest (RF), support vector machine (SVM), and eXtreme Gradient Boosting (XGBoost). The logistic regression (LR) was used to screen for independent influencing factors of clinical baseline characteristics. Machine learning models based on different feature datasets were constructed and validated using algorithms such as LR, RF, SVM, and XGBoost. By computing the selected features, a radiomics score was generated, and a diverse feature dataset was constructed. Based on the training set, 16 ML models were created, and their performance was evaluated using the validation set for metrics including sensitivity, specificity, accuracy, area under the receiver operating characteristic curve (AUC), and others.
Results
The training set consisted of 98 patients (mean age: 64.5 ± 10.5 years; 30 males), whereas the validation set consisted of 42 patients (mean age: 65.3 ± 9.78 years; 17 males). Hydronephrosis was the best independent influence factor ( p < 0.05). The RF model had the best performance in predicting high‐grade UTUC, with AUC of 0.914 (95% Confidence Interval [95%CI] 0.852–0.977) and 0.903 (95%CI 0.809–0.997) in the training set and validation set, and accuracy of 0.878 and 0.857, respectively.
Conclusions
An ML model based on the RF algorithm exhibits excellent predictive performance, offering a non‐invasive approach for predicting preoperative high‐grade UTUC.
... Manual segmentation involves a radiologist or a trained technician manually outlining the ROI around the tumour and its surrounding tissues [37]. This method can be time-consuming and subject to inter-observer variability but can be useful in cases where the tumour has unclear borders or an irregular shape [38]. Semi-automatic segmentation uses software to assist with the segmentation process [39], reducing the time required and increasing consistency between different observers. ...
Lung cancer is one of the most serious and life-threatening diseases in the world. Imaging modalities like computed tomography (CT) and Positron emission tomography (PET) play a crucial role in cancer diagnosis. Radiomics is an emerging field in medical imaging that uses advanced computational algorithms to extract quantitative features from medical images. Machine learning makes radiomics method of cancer diagnosis easier and more efficient by automating the process of feature selection and classification, which can save time and reduce the risk of human error in the diagnosis. It has the potential to revolutionize cancer detection by providing clinicians with valuable insights into tumour biology that can help in clinical decision-making and improve patient care outcomes. In this review paper, we primarily summarize the workflow of radiomics studies in the context of lung cancer and discussed the practical uses of radiomics in lung cancer, such as malignant tumour identification, classification of histologic subtypes, identification of tumour genotypes, and prediction of treatment response. Additionally, the paper addresses the key challenges associated with the clinical transition of radiomics, the limitations of current approaches, and potential future directions in this field.
... The histogram analysis method provides more and more detailed quantitative information on image heterogeneity by quantifying the grayscale information of lesion images, calculating the characteristic parameters of ROI in the image, evaluating the relationship and distribution of ROI gray intensity, and providing more detailed quantitative information on image heterogeneity, which has the advantages of simple operation process, objective data quantification and strong reproducibility of results [14,15]. So far, many research results have shown that histogram analysis has high medical value for diagnosing benign and malignant tumors, evaluating treatment effects, and predicting prognosis [16][17][18]. ...
... Texture analysis is different from the competent image feature analysis, which quantifies the grayscale information of the image, reflects the microstructure and internal biological indicators of tumor tissue with objective indicators, provides more tumor heterogeneity information that cannot be observed by the naked eye, and [14,15]. Histogram analysis is a useful texture analysis method that the main parameters include the mean, variance, skewness, scale and the 1st, 10th, 50th, 90th and 99th percentiles [19]. ...
Objective
This study sought to evaluate the worth of the general characteristics of enhanced CT images and the histogram parameters of each stage in distinguishing pleomorphic adenoma (PA) and adenolymphoma (AL).
Methods
The imaging features and histogram parameters of preoperative enhanced CT images in 20 patients with PA and 29 patients with AL were analyzed. Tumor morphology and histogram parameters of PA and AL were compared. Area under the curve (AUC), sensitivity, and subject operational feature specificity (ROC) analysis were used to determine the differential diagnostic effect of single-stage or multi-stage parameter combinations.
Results
The difference in CT value and net enhancement value of arterial phase (AP) were significant (p < 0.05); Flat sweep phase (FSP), AP mean, percentiles, 10th, 50th, 90th, 99th and arterial period variance and venous phase (VP) kurtosis in the nine histogram parameters of each period (p < 0.05). An analysis of the ROC curve revealed a maximum area beneath the curve (AUC) in the 90th percentile of FSP for a single-parameter differential diagnosis to be 0.870. The diagnostic efficacy of the mean value of FSP + The 90th percentile of AP + Kurtosis of VP was the best in multi-parameter combination diagnosis, with an AUC of 0.925, and the sensitivity and specificity of 0.900 and 0.850, respectively.
Conclusion
The histogram analysis of enhanced CT images is valuable for the differentiation of PA and AL. Moreover, the combination of single-stage parameters or multi-stage parameters can improve the differential diagnosis efficiency.
... The mixture of pathological components might offset the enhancement differences. 22,23 In addition, the size of PSPs was negatively correlated with the acceleration index. With a similar net enhancement, the shorter TTP of smaller peripheral PSPs resulted in a higher perfusion efficiency and accelerated index, which were more common in malignant lesions. ...
Background:
This study aimed to evaluate the multi-phase CT findings of central and peripheral pulmonary sclerosing pneumocytomas (PSPs) and compared them with Ki-67 to reveal their neoplastic nature.
Patients and methods:
Multi-phase CT and clinical data of 33 PSPs (15 central PSPs and 18 peripheral PSPs) were retrospectively analyzed and compared their multi-phase CT features and Ki-67 levels.
Results:
For quantitative indicators, central PSPs were larger than peripheral PSPs (10.39 ± 3.25 cm3 vs. 4.65 ± 2.61 cm3, P = 0.013), and tumor size was negatively correlated with acceleration index (r = -0.845, P < 0.001). The peak enhancement of central PSPs appeared in the delayed phase, with a longer time to peak enhancement (TTP, 100.81 ± 19.01 s), lower acceleration index (0.63 ± 0.17), progressive enhancement, and higher Ki-67 level. The peak enhancement of peripheral PSPs appeared in the venous phase, with the shorter TTP (62.67 ± 20.96 s, P < 0.001), higher acceleration index (0.99 ± 0.25, P < 0.001), enhancement washout, and lower Ki-67 level. For qualitative indicators, the overlying vessel sign (86.67% vs. 44.44%, P = 0.027), prominent pulmonary artery sign (73.33% vs. 27.78%, P = 0.015), and obstructive inflammation/atelectasis (26.67% vs. 0%, P = 0.033) were more common in central PSPs, while peripheral PSPs were more common with halo sign (38.89% vs. 6.67%, P = 0.046).
Conclusions:
The location of PSP is a possible contributing factor to its diverse imaging-pathological findings. The tumor size, multi-phase enhancement, qualitative signs, and Ki-67 were different between central and peripheral PSPs. Combined tumor size, multi-phase findings, and Ki-67 level are helpful to reveal the nature of the borderline tumor.
... Texture analysis is an emerging technique that can quantitatively and objectively evaluate tumor heterogeneity by analyzing the distribution and relationship between pixels or voxel gray levels on medical imaging (10). It has shown potential clinical value in the preoperative non-invasive diagnosis of tumors and staging, evaluation of the response to treatment, and assessment of outcome in human medicine (11)(12)(13)(14). ...
Objective
This study aimed to investigate the feasibility of computed tomography (CT) texture analysis for distinguishing canine adrenal gland tumors and its usefulness in clinical decision-making.
Materials and methods
The medical records of 25 dogs with primary adrenal masses who underwent contrast CT and a histopathological examination were retrospectively reviewed, of which 12 had adenomas (AAs), 7 had adenocarcinomas (ACCs), and 6 had pheochromocytomas (PHEOs). Conventional CT evaluation of each adrenal gland tumor included the mean, maximum, and minimum attenuation values in Hounsfield units (HU), heterogeneity of the tumor parenchyma, and contrast enhancement (type, pattern, and degree), respectively, in each phase. In CT texture analysis, precontrast and delayed-phase images of 18 adrenal gland tumors, which could be applied for ComBat harmonization were used, and 93 radiomic features (18 first-order and 75 second-order statistics) were extracted. Then, ComBat harmonization was applied to compensate for the batch effect created by the different CT protocols. The area under the receiver operating characteristic curve (AUC) for each significant feature was used to evaluate the diagnostic performance of CT texture analysis.
Results
Among the conventional features, PHEO showed significantly higher mean and maximum precontrast HU values than ACC (p < 0.05). Eight second-order features on the precontrast images showed significant differences between the adrenal gland tumors (p < 0.05). However, none of them were significantly different between AA and PHEO, or between precontrast images and delayed-phase images. This result indicates that ACC exhibited more heterogeneous and complex textures and more variable intensities with lower gray-level values than AA and PHEO. The correlation, maximal correlation coefficient, and gray level non-uniformity normalized were significantly different between AA and ACC, and between ACC and PHEO. These features showed high AUCs in discriminating ACC and PHEO, which were comparable or higher than the precontrast mean and maximum HU (AUC = 0.865 and 0.860, respectively).
Conclusion
Canine primary adrenal gland tumor differentiation can be achieved with CT texture analysis on precontrast images and may have a potential role in clinical decision-making. Further prospective studies with larger populations and cross-validation are warranted.
