A wavelet-based optimal texture feature set for classification of brain tumours

Region-wise severity analysis of diabetic plantar foot thermograms

OBJECTIVES

Diabetic foot ulcers (DFU) can be avoided if symptoms of diabetic foot complications are detected early and treated promptly. Early detection requires regular examination, which might be limited for many reasons. To identify affected or potentially affected regions in the diabetic plantar foot, the region-wise severity of the plantar foot must be known.

METHODS

A novel thermal diabetic foot dataset of 104 subjects was developed that is suitable for Indian healthcare conditions. The entire plantar foot thermogram is divided into three parts, i.e., forefoot, midfoot, and hindfoot. The division of plantar foot is based on the prevalence of foot ulcers and the load on the foot. To classify the severity levels, conventional machine learning (CML) techniques like logistic regression, decision tree, KNN, SVM, random forest, etc., and convolutional neural networks (CNN), such as EfficientNetB1, VGG-16, VGG-19, AlexNet, InceptionV3, etc., were applied and compared for robust outcomes.

RESULTS

The study successfully developed a thermal diabetic foot dataset, allowing for effective classification of diabetic foot ulcer severity using the CML and CNN techniques. The comparison of different methods revealed variations in performance, with certain approaches outperforming others.

CONCLUSIONS

The region-based severity analysis offers valuable insights for targeted interventions and preventive measures, contributing to a comprehensive assessment of diabetic foot ulcer severity. Further research and development in these techniques can enhance the detection and management of diabetic foot complications, ultimately improving patient outcomes.

Automated Classification of Brain Tumors from Magnetic Resonance Imaging Using Deep Learning

Brain tumor classification is crucial for medical evaluation in computer-assisted diagnostics (CAD). However, manual diagnosis of brain tumors from magnetic resonance imaging (MRI) can be time-consuming and complex, leading to inaccurate detection and classification. This is mainly because brain tumor identification is a complex procedure that relies on different modules. The advancements in Deep Learning (DL) have assisted in the automated process of medical images and diagnostics for various medical conditions, which benefits the health sector. Convolutional Neural Network (CNN) is one of the most prominent DL methods for visual learning and image classification tasks. This study presents a novel CNN algorithm to classify the brain tumor types of glioma, meningioma, and pituitary. The algorithm was tested on benchmarked data and compared with the existing pre-trained VGG16, VGG19, ResNet50, MobileNetV2, and InceptionV3 algorithms reported in the literature. The experimental results have indicated a high classification accuracy of 98.04%, precision, recall, and f1-score success rate of 98%, respectively. The classification results proved that the most common kinds of brain tumors could be categorized with a high level of accuracy. The presented algorithm has good generalization capability and execution speed that can be helpful in the field of medicine to assist doctors in making prompt and accurate decisions associated with brain tumor diagnosis.

Brain Tumor Diagnosis Using Machine Learning, Convolutional Neural Networks, Capsule Neural Networks and Vision Transformers, Applied to MRI: A Survey

Management of brain tumors is based on clinical and radiological information with presumed grade dictating treatment. Hence, a non-invasive assessment of tumor grade is of paramount importance to choose the best treatment plan. Convolutional Neural Networks (CNNs) represent one of the effective Deep Learning (DL)-based techniques that have been used for brain tumor diagnosis. However, they are unable to handle input modifications effectively. Capsule neural networks (CapsNets) are a novel type of machine learning (ML) architecture that was recently developed to address the drawbacks of CNNs. CapsNets are resistant to rotations and affine translations, which is beneficial when processing medical imaging datasets. Moreover, Vision Transformers (ViT)-based solutions have been very recently proposed to address the issue of long-range dependency in CNNs. This survey provides a comprehensive overview of brain tumor classification and segmentation techniques, with a focus on ML-based, CNN-based, CapsNet-based, and ViT-based techniques. The survey highlights the fundamental contributions of recent studies and the performance of state-of-the-art techniques. Moreover, we present an in-depth discussion of crucial issues and open challenges. We also identify some key limitations and promising future research directions. We envisage that this survey shall serve as a good springboard for further study.

Reviewing Machine Learning and Image Processing Based Decision-Making Systems for Breast Cancer Imaging

Breast cancer (BC) is the leading cause of death among women worldwide. It affects in general women older than 40 years old. Medical images analysis is one of the most promising research areas since it provides facilities for diagnosis and decision-making of several diseases such as BC. This paper conducts a Structured Literature Review (SLR) of the use of Machine Learning (ML) and Image Processing (IP) techniques to deal with BC imaging. A set of 530 papers published between 2000 and August 2019 were selected and analyzed according to ten criteria: year and publication channel, empirical type, research type, medical task, machine learning techniques, datasets used, validation methods, performance measures and image processing techniques which include image pre-processing, segmentation, feature extraction and feature selection. Results showed that diagnosis was the most used medical task and that Deep Learning techniques (DL) were largely used to perform classification. Furthermore, we found out that classification was the most ML objective investigated followed by prediction and clustering. Most of the selected studies used Mammograms as imaging modalities rather than Ultrasound or Magnetic Resonance Imaging with the use of public or private datasets with MIAS as the most frequently investigated public dataset. As for image processing techniques, the majority of the selected studies pre-process their input images by reducing the noise and normalizing the colors, and some of them use segmentation to extract the region of interest with the thresholding method. For feature extraction, we note that researchers extracted the relevant features using classical feature extraction techniques (e.g. Texture features, Shape features, etc.) or DL techniques (e. g. VGG16, VGG19, ResNet, etc.), and finally few papers used feature selection techniques in particular the filter methods.

Image-based state-of-the-art techniques for the identification and classification of brain diseases: a review

Detection and classification methods have a vital and important role in identifying brain diseases. Timely detection and classification of brain diseases enable an accurate identification and effective management of brain impairment. Brain disorders are commonly most spreadable diseases and the diagnosing process is time-consuming and highly expensive. There is an utmost need to develop effective and advantageous methods for brain diseases detection and characterization. Magnetic resonance imaging (MRI), computed tomography (CT), and other various brain imaging scans are used to identify different brain diseases and disorders. Brain imaging scans are the efficient tool to understand the anatomical changes in brain in fast and accurate manner. These different brain imaging scans used with segmentation techniques and along with machine learning and deep learning techniques give maximum accuracy and efficiency. This paper focuses on different conventional approaches, machine learning and deep learning techniques used for the detection, and classification of brain diseases and abnormalities. This paper also summarizes the research gap and problems in the existing techniques used for detection and classification of brain disorders. Comparison and evaluation of different machine learning and deep learning techniques in terms of efficiency and accuracy are also highlighted in this paper. Furthermore, different brain diseases like leukoariaosis, Alzheimer's, Parkinson's, and Wilson's disorder are studied in the scope of machine learning and deep learning techniques.

Glioblastomas and brain metastases differentiation following an MRI texture analysis-based radiomics approach

PURPOSE

To evaluate the potential of 2D texture features extracted from magnetic resonance (MR) images for differentiating brain metastasis (BM) and glioblastomas (GBM) following a radiomics approach.

METHODS

This retrospective study included 50 patients with BM and 50 with GBM who underwent T1-weighted MRI between December 2010 and January 2017. Eighty-eight rotation-invariant texture features were computed for each segmented lesion using six texture analysis methods. These features were also extracted from the four images obtained after applying the discrete wavelet transform (88 features × 4 images). Three feature selection methods and five predictive models were evaluated. A 5-fold cross-validation scheme was used to randomly split the study group into training (80 patients) and testing (20 patients), repeating the process ten times. Classification was evaluated computing the average area under the receiver operating characteristic curve. Sensibility, specificity and accuracy were also computed. The whole process was tested quantizing the images with different gray-level values to evaluate their influence in the final results.

RESULTS

Highest classification accuracy was obtained using the original images quantized with 128 gray-levels and a feature selection method based on the p-value. The best overall performance was achieved using a support vector machine model with a subset of 32 features (AUC = 0.896 ± 0.067, sensitivity of 82% and specificity of 80%). Naïve Bayes and k-nearest neighbors models showed also valuable results (AUC ≈ 0.8) with a lower number of features (<13), thus suggesting that these models may be more generalizable when using external validations.

CONCLUSION

The proposed radiomics MRI approach is able to discriminate between GBM and BM with high accuracy employing a set of 2D texture features, thus helping in the diagnosis of brain lesions in a fast and non-invasive way.

Comprehensive analysis of potential prognostic biomarker in gliomas

Purpose

Gliomas are common intracranial tumors with the characteristic of diffuse and invasive growth. The prognosis is poor, and the recurrence rate and mortality are higher. With the development of big data technology, many methods such as natural language processing, computer vision and image processing have been deeply applied in the medical field. This can help clinicians to provide personalized and precise diagnosis and therapeutic schedule for patients with different type of gliomas to achieve the best therapeutic effect. The purpose of this paper is to summarize and extract useful information from published research results by conducting a secondary analysis of the literature.

Design/methodology/approach

The PubMed and China National Knowledge Infrastructure (CNKI) literature database were used to retrieve published Chinese and English research papers about human gliomas. Comprehensive analysis was applied to conduct this research. The factors affecting survival and prognosis were screened and analyzed respectively in this paper, and different methods for multidimensional data of patients were discussed.

Findings

This paper identified biomarkers and therapeutic modalities associated with prognosis for different grade of gliomas. This paper investigated the relationship among these clinical prognostic factors and different histopathologic tying and grade of gliomas by comprehensive analysis. This paper summarizes the research progress of biomarker in medical imaging and genomics of gliomas to improve prognosis and the current status of treatment in China.

Originality/value

Combined with multimodal data such as genomics data, medical image data and clinical information data, this paper comprehensively analyzed the prognostic factors of glioma and provided guidance and evidence for rational treatment planning and improvement of clinical treatment prognosis.

A Review on a Deep Learning Perspective in Brain Cancer Classification

A World Health Organization (WHO) Feb 2018 report has recently shown that mortality rate due to brain or central nervous system (CNS) cancer is the highest in the Asian continent. It is of critical importance that cancer be detected earlier so that many of these lives can be saved. Cancer grading is an important aspect for targeted therapy. As cancer diagnosis is highly invasive, time consuming and expensive, there is an immediate requirement to develop a non-invasive, cost-effective and efficient tools for brain cancer characterization and grade estimation. Brain scans using magnetic resonance imaging (MRI), computed tomography (CT), as well as other imaging modalities, are fast and safer methods for tumor detection. In this paper, we tried to summarize the pathophysiology of brain cancer, imaging modalities of brain cancer and automatic computer assisted methods for brain cancer characterization in a machine and deep learning paradigm. Another objective of this paper is to find the current issues in existing engineering methods and also project a future paradigm. Further, we have highlighted the relationship between brain cancer and other brain disorders like stroke, Alzheimer’s, Parkinson’s, and Wilson’s disease, leukoriaosis, and other neurological disorders in the context of machine learning and the deep learning paradigm.

MRI Texture Analysis Reveals Bulbar Abnormalities in Friedreich Ataxia

BACKGROUND AND PURPOSE

Texture analysis is an image processing technique that can be used to extract parameters able to describe meaningful features of an image or ROI. Texture analysis based on the gray level co-occurrence matrix gives a second-order statistical description of the image or ROI. In this work, the co-occurrence matrix texture approach was used to extract information from brain MR images of patients with Friedreich ataxia and a control group, to see whether texture parameters were different between these groups. A longitudinal analysis was also performed.

MATERIALS AND METHODS

Twenty patients and 21 healthy controls participated in the study. Both groups had 2 sets of T1-weighted MR images obtained 1 year apart for every subject. ROIs chosen for analysis were the medulla oblongata and pons. Texture parameters were obtained for these ROIs for every subject, for the 2 sets of images. These parameters were compared longitudinally within groups and transversally between groups.

RESULTS

The comparison between patients and the control group showed a significant differences for the medulla oblongata (t test, P < .05, Bonferroni-corrected) but did not show a statistically significant difference for the pons. Longitudinal comparison of images obtained 1 year apart did not show differences for either patients or for controls, in any of the analyzed structures.

CONCLUSIONS

Gray level co-occurrence matrix-based texture analysis showed statistically significant differences for the medulla oblongata of patients with Friedreich ataxia compared with controls. These results highlight the medulla as an important site of damage in Friedreich ataxia.

Feature selection for neural network based defect classification of ceramic components using high frequency ultrasound

The motivation for this research stems from a need for providing a non-destructive testing method capable of detecting and locating any defects and microstructural variations within armour ceramic components before issuing them to the soldiers who rely on them for their survival. The development of an automated ultrasonic inspection based classification system would make possible the checking of each ceramic component and immediately alert the operator about the presence of defects. Generally, in many classification problems a choice of features or dimensionality reduction is significant and simultaneously very difficult, as a substantial computational effort is required to evaluate possible feature subsets. In this research, a combination of artificial neural networks and genetic algorithms are used to optimize the feature subset used in classification of various defects in reaction-sintered silicon carbide ceramic components. Initially wavelet based feature extraction is implemented from the region of interest. An Artificial Neural Network classifier is employed to evaluate the performance of these features. Genetic Algorithm based feature selection is performed. Principal Component Analysis is a popular technique used for feature selection and is compared with the genetic algorithm based technique in terms of classification accuracy and selection of optimal number of features. The experimental results confirm that features identified by Principal Component Analysis lead to improved performance in terms of classification percentage with 96% than Genetic algorithm with 94%.

Automated Feature Extraction in Brain Tumor by Magnetic Resonance Imaging Using Gaussian Mixture Models

This paper presents a novel method for Glioblastoma (GBM) feature extraction based on Gaussian mixture model (GMM) features using MRI. We addressed the task of the new features to identify GBM using T1 and T2 weighted images (T1-WI, T2-WI) and Fluid-Attenuated Inversion Recovery (FLAIR) MR images. A pathologic area was detected using multithresholding segmentation with morphological operations of MR images. Multiclassifier techniques were considered to evaluate the performance of the feature based scheme in terms of its capability to discriminate GBM and normal tissue. GMM features demonstrated the best performance by the comparative study using principal component analysis (PCA) and wavelet based features. For the T1-WI, the accuracy performance was 97.05% (AUC = 92.73%) with 0.00% missed detection and 2.95% false alarm. In the T2-WI, the same accuracy (97.05%, AUC = 91.70%) value was achieved with 2.95% missed detection and 0.00% false alarm. In FLAIR mode the accuracy decreased to 94.11% (AUC = 95.85%) with 0.00% missed detection and 5.89% false alarm. These experimental results are promising to enhance the characteristics of heterogeneity and hence early treatment of GBM.

Quantification of heterogeneity as a biomarker in tumor imaging: a systematic review

BACKGROUND

Many techniques are proposed for the quantification of tumor heterogeneity as an imaging biomarker for differentiation between tumor types, tumor grading, response monitoring and outcome prediction. However, in clinical practice these methods are barely used. This study evaluates the reported performance of the described methods and identifies barriers to their implementation in clinical practice.

METHODOLOGY

The Ovid, Embase, and Cochrane Central databases were searched up to 20 September 2013. Heterogeneity analysis methods were classified into four categories, i.e., non-spatial methods (NSM), spatial grey level methods (SGLM), fractal analysis (FA) methods, and filters and transforms (F&T). The performance of the different methods was compared.

PRINCIPAL FINDINGS

Of the 7351 potentially relevant publications, 209 were included. Of these studies, 58% reported the use of NSM, 49% SGLM, 10% FA, and 28% F&T. Differentiation between tumor types, tumor grading and/or outcome prediction was the goal in 87% of the studies. Overall, the reported area under the curve (AUC) ranged from 0.5 to 1 (median 0.87). No relation was found between the performance and the quantification methods used, or between the performance and the imaging modality. A negative correlation was found between the tumor-feature ratio and the AUC, which is presumably caused by overfitting in small datasets. Cross-validation was reported in 63% of the classification studies. Retrospective analyses were conducted in 57% of the studies without a clear description.

CONCLUSIONS

In a research setting, heterogeneity quantification methods can differentiate between tumor types, grade tumors, and predict outcome and monitor treatment effects. To translate these methods to clinical practice, more prospective studies are required that use external datasets for validation: these datasets should be made available to the community to facilitate the development of new and improved methods.

Multi-scale classification of disease using structural MRI and wavelet transform

Recently, multivariate analysis algorithms have become a popular tool to diagnose neurological diseases based on neuroimaging data. Most studies, however, are biased for one specific scale, namely the scale given by the spatial resolution (i.e. dimension) of the data. In the present study, we propose to use the dual-tree complex wavelet transform to extract information on different spatial scales from structural MRI data and show its relevance for disease classification. Based on the magnitude representation of the complex wavelet coefficients calculated from the MR images, we identified a new class of features taking scale, directionality and potentially local information into account simultaneously. By using a linear support vector machine, these features were shown to discriminate significantly between spatially normalized MR images of 41 patients suffering from multiple sclerosis and 26 healthy controls. Interestingly, the decoding accuracies varied strongly among the different scales and it turned out that scales containing low frequency information were partly superior to scales containing high frequency information. Usually, this type of information is neglected since most decoding studies use only the original scale of the data. In conclusion, our proposed method has not only a high potential to assist in the diagnostic process of multiple sclerosis, but can be applied to other diseases or general decoding problems in structural or functional MRI.

Hippocampal CA3 transcriptome signature correlates with initial precipitating injury in refractory mesial temporal lobe epilepsy

BACKGROUND

Prolonged febrile seizures constitute an initial precipitating injury (IPI) commonly associated with refractory mesial temporal lobe epilepsy (RMTLE). In order to investigate IPI influence on the transcriptional phenotype underlying RMTLE we comparatively analyzed the transcriptomic signatures of CA3 explants surgically obtained from RMTLE patients with (FS) or without (NFS) febrile seizure history. Texture analyses on MRI images of dentate gyrus were conducted in a subset of surgically removed sclerotic hippocampi for identifying IPI-associated histo-radiological alterations.

METHODOLOGY/PRINCIPAL FINDINGS

DNA microarray analysis revealed that CA3 global gene expression differed significantly between FS and NFS subgroups. An integrative functional genomics methodology was used for characterizing the relations between GO biological processes themes and constructing transcriptional interaction networks defining the FS and NFS transcriptomic signatures and its major gene-gene links (hubs). Co-expression network analysis showed that: i) CA3 transcriptomic profiles differ according to the IPI; ii) FS distinctive hubs are mostly linked to glutamatergic signalization while NFS hubs predominantly involve GABAergic pathways and neurotransmission modulation. Both networks have relevant hubs related to nervous system development, what is consistent with cell genesis activity in the hippocampus of RMTLE patients. Moreover, two candidate genes for therapeutic targeting came out from this analysis: SSTR1, a relevant common hub in febrile and afebrile transcriptomes, and CHRM3, due to its putative role in epilepsy susceptibility development. MRI texture analysis allowed an overall accuracy of 90% for pixels correctly classified as belonging to FS or NFS groups. Histological examination revealed that granule cell loss was significantly higher in FS hippocampi.

CONCLUSIONS/SIGNIFICANCE

CA3 transcriptional signatures and dentate gyrus morphology fairly correlate with IPI in RMTLE, indicating that FS-RMTLE represents a distinct phenotype. These findings may shed light on the molecular mechanisms underlying refractory epilepsy phenotypes and contribute to the discovery of novel specific drug targets for therapeutic interventions.

Robust texture features for response monitoring of glioblastoma multiforme on T1-weighted and T2-FLAIR MR images: a preliminary investigation in terms of identification and segmentation

PURPOSE

Image texture has recently attracted much attention in providing quantitative features that are unique to various different tissue types, in particular, in MR images of the brain. Such image features may be useful for tumor response quantification. As a first step, one needs to establish if these features are sensitive to different tissues of clinical relevance. Here, a novel method of texture analysis based on the Hartley transform has been investigated and applied to MR images of glioblastoma multiforme (GBM).

METHODS

Contrast-enhanced T1-weighted gradient-echo and T2-FLAIR spin-echo MR images of 27 GBM patients acquired prior to radiation therapy were available for analysis. Before computing texture features on these images, a novel image transformation was employed in the form of a power map computed from the localized Hartley transform of the image. Haralick statistical texture features were then computed based on the power map. This method was compared to the standard approach of obtaining texture features directly from the image. Twelve different features were computed on different resolution levels. On a regional resolution level, image texture features were identified that are able to correctly classify entire regions within T1-weighted and T2-FLAIR brain MR images of GBM patients into abnormal (containing contrast-enhancing GBM tumor) and brain tissue. Various metrics [area under the ROC curve (AUC), maximum accuracy, and Canberra distance] have been computed to quantify the usefulness of these features. On a local resolution level, it was investigated which of these features are able to provide a voxel-by-voxel enhancement that could be used for assisting the segmentation of the gross tumor volume on T1 images. The "gold standard" for this analysis was a gross tumor volume corresponding to the contrast-enhancing lesion visualized on T1-weighted images as segmented by a radiation oncologist.

RESULTS

The Sum-mean and Variance features demonstrated the best performance overall. For the T1-weighted images, the identification performance of Sum-mean and Variance features computed from the power map was higher (AUC = 0.9959 and AUC = 0.9918, respectively) and with higher Canberra distances as compared to features computed directly from the images (AUC = 0.8930 and AUC = 0.9163, respectively). These results in T2-FLAIR images were even superior. The features computed from the power map showed an unequivocal identification (AUC = 1) with higher Canberra distances, whereas the performance of the features from the original images was slightly lower (AUC = 0.9739 and AUC = 0.9904, respectively). The same features computed on the power map of the T1-weighted images also provided superior enhancement in individual tumor voxels as compared to the features computed on the original images.

CONCLUSIONS

The Sum-mean and Variance features are both useful for identifying and segmenting GBM tumors on localized Hartley transformed MR images.