Wyniki wyszukiwania - Biblioteka Nauki

1

Assessment of CT for the categorization of hemorrhagic stroke (HS) and cerebral amyloid angiopathy hemorrhage (CAAH): A review

100%

Sudarshan V. K. , Raghavendra U. , Gudigar A. , Ciaccio E. J. , Vijayananthan A. , Sahathevan R. , Acharya U. R.

Biocybernetics and Biomedical Engineering

|

2022

|

tom Vol. 42, no. 3

888--901

EN

The management of intracerebral hemorrhage (ICH) requires prompt diagnostic assessment and recognition. Accurate localization and categorization of ICH-type is crucial. There are two main categories of ICH: 1) hemorrhagic stroke (HS), which occurs in the deeper or subcortical regions of the brain, where the arterial network tapers to fine end-arteries, and, 2) cerebral amyloid angiopathy hemorrhage (CAAH), which occurs at the superficial or cortical-subcortical region of the grey and white matter junction. Computed tomography (CT) and magnetic resonance imaging (MRI) are the most used imaging tools in diagnosing ICH. However, availability, time, and cost often prevent emergent MRI use. Therefore, CT remains the primary tool in the diagnosis of ICH. The assessment of imaging studies is time-dependent, and a radiologist should do a detailed diagnostic evaluation. Human error can occur in a pressured clinical setting, even for highly trained medical professionals. Assisted or automated computer-aided analysis of CT/MRI may help to reduce the assessment time, improve the diagnostic accuracy, better differentiate between types of ICH, and reduce the risk of human errors. This review evaluates CT and MRI’s role in distinguishing between the two varieties of ICH-HS and CAAH. It focuses on how CT could be utilized as the preferred diagnostic tool. In addition, we discuss the role of automation using machine learning (ML) and the role or advantages of ML in the automated assessment of CT for the detection and classification of HS and CAAH. We have included our observations for future research and the requirements for further evaluation.

2

An automated skin melanoma detection system with melanoma-index based on entropy features

84%

Cheong K. H. , Tang K. J. W. , Zhao X. , Koh J. E. W. , Faust O. , Gururajan R. , Ciaccio E. J. , Rajinikanth V. , Acharya U. R.

Biocybernetics and Biomedical Engineering

|

2021

|

tom Vol. 41, no. 3

997--1012

EN

Skin melanoma is a potentially life-threatening cancer. Once it has metastasized, it may cause severe disability and death. Therefore, early diagnosis is important to improve the conditions and outcomes for patients. The disease can be diagnosed based on Digital-Dermoscopy (DD) images. In this study, we propose an original and novel Automated Skin-Melanoma Detection (ASMD) system with Melanoma-Index (MI). The system incorporates image pre-processing, Bi-dimensional Empirical Mode Decomposition (BEMD), image texture enhancement, entropy and energy feature mining, as well as binary classification. The system design has been guided by feature ranking, with Student’s t-test and other statistical methods used for quality assessment. The proposed ASMD was employed to examine 600 benign and 600 DD malignant images from benchmark databases. Our classification performance assessment indicates that the combination of Support Vector Machine (SVM) and Radial Basis Function (RBF) offers a classification accuracy of greater than 97.50%. Motivated by these classification results, we also formulated a clinically relevant MI using the dominant entropy features. Our proposed index can assist dermatologists to track multiple information-bearing features, thereby increasing the confidence with which a diagnosis is given.

3

Novel multiple pooling and local phase quantization stable feature extraction techniques for automated classification of brain infarcts

67%

Dogan S. , Barua P. D. , Baygin M. , Chakraborty S. , Ciaccio E. J. , Tuncer T. , Abd Kadir K. A. , Md Shah M. N. , Azman R. R. , Lee C. C. , Ng K. H. , Acharya U. R.

Biocybernetics and Biomedical Engineering

|

2022

|

tom Vol. 42, no. 3

815--828

EN

This study aims to introduce a hand-crafted machine learning method to classify ischemic and hemorrhagic strokes with satisfactory performance. In the first step of this work, a new CT brain for images dataset was collected for stroke patients. A highly accurate hand-crafted machine learning method is developed and tested for these cases. This model uses preprocessing, feature creation using a novel pooling method (it is named P9), a local phase quantization (LPQ) operator, and a Chi2-based selector responsible for selecting the most significant features. After that, classification is done using the k-nearest neighbor (kNN) classifier with ten-fold cross-validation (CV). The novel aspect of this model is the P9 pooling method. The inspiration for this pooling method was drawn from the deep learning models, where features are extracted with multiple layers using a convolution operator applied to the pooling method. However, pooling decompositions have a routing problem. The P9 pooling function creates nine decomposed models, hence the name. The LPQ feature extractor is applied to images to generate sub-bands for feature generation. The Chi2 selector is then employed to select the most significant features from the created feature vector, and these features are utilized for the classification using the k-nearest neighbor algorithm (kNN). The introduced P9-LPQ feature extraction-based learning model attained over 98% classification accuracy in all cases. The results obtained in this paper show that the proposed method can successfully classify stroke types. For this reason, the developed model can pre-diagnose stroke types in the future.

4

Endoscopy, video capsule endoscopy, and biopsy for automated celiac disease detection: A review

67%

Jahmunah V. , En Wei Koh J. , Sudarshan V. K. , Raghavendra U. , Gudigar A. , Lih Oh S. , Wen Loh H. , Faust O. , Datta Barua P. , Ciaccio E. J. , Rajendra Acharya U.

Biocybernetics and Biomedical Engineering

|

2023

|

tom Vol. 43, no. 1

82--108

EN

Celiac Disease (CD) is a common ailment that affects approximately 1% of the world population. Automated CD detection can help experts during the diagnosis of this condition at an early stage and bring significant benefits to both patients and healthcare providers. For this purpose, scientists have created automatic and semi-automatic CD diagnostic support systems. In this study, we performed information extraction methods that were found useful for efforts to differentiate CD versus non-CD. To focus the review process, only methods for endoscopy, video capsule endoscopy (VCE) and biopsy image analyses were considered. As described herein, we have learned that statistical and non-linear methods are most important for information extraction. These information extraction tools might benefit clinical workflows by reducing intra- and inter-observer variability. However, bias, introduced by resolving design choices during the creation of diagnostic support systems, may limit the general validity of the performance results, impacting the transferability of study outcomes. Therefore, having am overview of information extraction tools. Together with their general and specific limitations, might be assistive in improving the information extraction process. We hope our review results will provide a foundation for the design of next-generation statistical and nonlinear methods that can be used in CD detection systems. We have also compared various review articles and discussed recommendations to improve CD diagnosis. From this review, it is evident that CD diagnosis is slowly moving away from conventional techniques towards advanced deep learning techniques.