1
|
Driban M, Yan A, Selvam A, Ong J, Vupparaboina KK, Chhablani J. Artificial intelligence in chorioretinal pathology through fundoscopy: a comprehensive review. Int J Retina Vitreous 2024; 10:36. [PMID: 38654344 PMCID: PMC11036694 DOI: 10.1186/s40942-024-00554-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/04/2024] [Accepted: 04/02/2024] [Indexed: 04/25/2024] Open
Abstract
BACKGROUND Applications for artificial intelligence (AI) in ophthalmology are continually evolving. Fundoscopy is one of the oldest ocular imaging techniques but remains a mainstay in posterior segment imaging due to its prevalence, ease of use, and ongoing technological advancement. AI has been leveraged for fundoscopy to accomplish core tasks including segmentation, classification, and prediction. MAIN BODY In this article we provide a review of AI in fundoscopy applied to representative chorioretinal pathologies, including diabetic retinopathy and age-related macular degeneration, among others. We conclude with a discussion of future directions and current limitations. SHORT CONCLUSION As AI evolves, it will become increasingly essential for the modern ophthalmologist to understand its applications and limitations to improve patient outcomes and continue to innovate.
Collapse
Affiliation(s)
- Matthew Driban
- Department of Ophthalmology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
| | - Audrey Yan
- Department of Medicine, West Virginia School of Osteopathic Medicine, Lewisburg, WV, USA
| | - Amrish Selvam
- Department of Ophthalmology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA
| | - Joshua Ong
- Michigan Medicine, University of Michigan, Ann Arbor, USA
| | | | - Jay Chhablani
- Department of Ophthalmology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA.
| |
Collapse
|
2
|
Basheer S, Singh KU, Sharma V, Bhatia S, Pande N, Kumar A. A robust NIfTI image authentication framework to ensure reliable and safe diagnosis. PeerJ Comput Sci 2023; 9:e1323. [PMID: 37346677 PMCID: PMC10280420 DOI: 10.7717/peerj-cs.1323] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/05/2022] [Accepted: 03/10/2023] [Indexed: 06/23/2023]
Abstract
Advancements in digital medical imaging technologies have significantly impacted the healthcare system. It enables the diagnosis of various diseases through the interpretation of medical images. In addition, telemedicine, including teleradiology, has been a crucial impact on remote medical consultation, especially during the COVID-19 pandemic. However, with the increasing reliance on digital medical images comes the risk of digital media attacks that can compromise the authenticity and ownership of these images. Therefore, it is crucial to develop reliable and secure methods to authenticate these images that are in NIfTI image format. The proposed method in this research involves meticulously integrating a watermark into the slice of the NIfTI image. The Slantlet transform allows modification during insertion, while the Hessenberg matrix decomposition is applied to the LL subband, which retains the most energy of the image. The Affine transform scrambles the watermark before embedding it in the slice. The hybrid combination of these functions has outperformed previous methods, with good trade-offs between security, imperceptibility, and robustness. The performance measures used, such as NC, PSNR, SNR, and SSIM, indicate good results, with PSNR ranging from 60 to 61 dB, image quality index, and NC all close to one. Furthermore, the simulation results have been tested against image processing threats, demonstrating the effectiveness of this method in ensuring the authenticity and ownership of NIfTI images. Thus, the proposed method in this research provides a reliable and secure solution for the authentication of NIfTI images, which can have significant implications in the healthcare industry.
Collapse
Affiliation(s)
- Shakila Basheer
- Department of Information Systems, College of Computer and Information Science, Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia
| | - Kamred Udham Singh
- Department of Computer Science and Information Engineering, National Cheng Kung University, Tai-nan, Taiwan, Taiwan
- School of Computing, Graphic Era Hill University, Dehradun, India
| | | | - Surbhi Bhatia
- King Faisal University, Al Hasa, Saudi Arabia
- Department of Data Science, School of Science, Engineering and Environment, University of Salford, Manchester, United Kingdom
| | - Nilesh Pande
- School of Technology Pandit Deendayal Energy University Gandhinagar, Gandhinagar, India
| | | |
Collapse
|
3
|
Fahmi A, Maqbool Z, Amin F, Aslam M. Blockchain knowledge selection under the trapezoidal fermatean fuzzy number. Soft comput 2023; 27:3601-3621. [PMID: 36407892 PMCID: PMC9649411 DOI: 10.1007/s00500-022-07611-w] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 10/24/2022] [Indexed: 11/12/2022]
Abstract
Blockchain knowledge signifies a useful fundamental knowledge to safeguard faith in transboundary transmittals for main banks and financial institutions. In the study of group decision-making, the most important issue is how to coordinate opinions from different blockchains to reach a compromise under uncertainty. To tackle uncertainties surrounding multi-attribute group decision-making (MAGDM) problems in real-life scenes, we introduce a trapezoidal fermatean fuzzy set which generalizes trapezoidal fuzzy sets and fermatean fuzzy sets. The trapezoidal fermatean fuzzy model enables the degrees of membership, abstention, and non-membership to be expressed by linguistic terms. We define the operational laws of trapezoidal fermatean fuzzy numbers, and Einstein aggregation operator based on the trapezoidal fermatean fuzzy number. This makes it more flexible and descriptive to model the attitudes of Blockchain knowledge in MAGDM applications. Since multi-input arguments are interconnected and Blockchain knowledge has a lot of options perception, we also define the TOPSIS technique to facilitate the fusion of trapezoidal fermatean fuzzy information. With the aid of the trapezoidal fermatean fuzzy-TOPSIS technique, the main goal of this research is to present a general MAGDM framework by integrating the step with the complex proportional assessment. A trapezoidal fermatean positive ideal solution always wants the maximum value of the benefit criteria and the minimum value of the cost criteria. On the other hand, the trapezoidal fermatean negative ideal solution always wants the maximum value of the cost criteria and the minimum value of the benefit criteria. An integrated trapezoidal fermatean fuzzy-TOPSIS framework is established. In the proposed decision framework, the trapezoidal fermatean fuzzy-TOPSIS method is utilized to identify the subjective weights of decision attributes, and the trapezoidal fermatean fuzzy-TOPSIS approach is used to rank alternatives. Lastly, a case study concerning blockchain knowledge assessment is presented to demonstrate that the suggested scheme is feasible and effective. Furthermore, sensitivity and comparison analyses are conducted to show the robustness and superiority of the proposed method.
Collapse
Affiliation(s)
- Aliya Fahmi
- Department of Mathematics, The University of Faisalabad, Faisalabad, Pakistan
| | - Zahida Maqbool
- Department of Management Studies, The University of Faisalabad, Faisalabad, Pakistan
| | - Fazli Amin
- Department of Mathematics, Hazara University, Mansehra, Pakistan
| | - Muhammad Aslam
- Department of Mathematics, College of Sciences, King Khalid University, Abha, Saudi Arabia
| |
Collapse
|
4
|
Pulmonary Nodule Detection Based on Multiscale Feature Fusion. COMPUTATIONAL AND MATHEMATICAL METHODS IN MEDICINE 2022; 2022:8903037. [PMID: 36590762 PMCID: PMC9797290 DOI: 10.1155/2022/8903037] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 07/06/2022] [Revised: 12/10/2022] [Accepted: 12/12/2022] [Indexed: 12/24/2022]
Abstract
As cancer with the highest morbidity and mortality in the world, lung cancer is characterized by pulmonary nodules in the early stage. The detection of pulmonary nodules is an important method for the early detection of lung cancer, which can greatly improve the survival rate of lung cancer patients. However, the accuracy of conventional detection methods for lung nodules is low. With the development of medical imaging technology, deep learning plays an increasingly important role in medical image detection, and pulmonary nodules can be accurately detected by CT images. Based on the above, a pulmonary nodule detection method based on deep learning is proposed. In the candidate nodule detection stage, the multiscale features and Faster R-CNN, a general-purpose detection framework based on deep learning, were combined together to improve the detection of small-sized lung nodules. In the false-positive nodule filtration stage, a 3D convolutional neural network based on multiscale fusion is designed to reduce false-positive nodules. The experiment results show that the candidate nodule detection model based on Faster R-CNN integrating multiscale features has achieved a sensitivity of 98.6%, 10% higher than that of the other single-scale model, the proposed method achieved a sensitivity of 90.5% at the level of 4 false-positive nodules per scan, and the CPM score reached 0.829. The results are higher than methods in other works of literature. It can be seen that the detection method of pulmonary nodules based on multiscale fusion has a higher detection rate for small nodules and improves the classification performance of true and false-positive pulmonary nodules. This will help doctors when making a lung cancer diagnosis.
Collapse
|
5
|
Bhatia S, Alojail M, Sengan S, Dadheech P. An efficient modular framework for automatic LIONC classification of MedIMG using unified medical language. Front Public Health 2022; 10:926229. [PMID: 36033768 PMCID: PMC9399779 DOI: 10.3389/fpubh.2022.926229] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/22/2022] [Accepted: 06/27/2022] [Indexed: 01/24/2023] Open
Abstract
Handwritten prescriptions and radiological reports: doctors use handwritten prescriptions and radiological reports to give drugs to patients who have illnesses, injuries, or other problems. Clinical text data, like physician prescription visuals and radiology reports, should be labelled with specific information such as disease type, features, and anatomical location for more effective use. The semantic annotation of vast collections of biological and biomedical texts, like scientific papers, medical reports, and general practitioner observations, has lately been examined by doctors and scientists. By identifying and disambiguating references to biomedical concepts in texts, medical semantics annotators could generate such annotations automatically. For Medical Images (MedIMG), we provide a methodology for learning an effective holistic representation (handwritten word pictures as well as radiology reports). Deep Learning (DL) methods have recently gained much interest for their capacity to achieve expert-level accuracy in automated MedIMG analysis. We discovered that tasks requiring significant responsive fields are ideal for downscaled input images that are qualitatively verified by examining functional, responsive areas and class activating maps for training models. This article focuses on the following contributions: (a) Information Extraction from Narrative MedImages, (b) Automatic categorisation on image resolution with an impact on MedIMG, and (c) Hybrid Model to Predictions of Named Entity Recognition utilising RNN + LSTM + GRM that perform admirably in every trainee for every input purpose. At the same time, supplying understandable scale weight implies that such multi-scale structures are also crucial for extracting information from high-resolution MedIMG. A portion of the reports (30%) are manually evaluated by trained physicians, while the rest were automatically categorised using deep supervised training models based on attention mechanisms and supplied with test reports. MetaMapLite proved recall and precision, but also an F1-score equivalent for primary biomedicine text search techniques and medical text examination on many databases of MedIMG. In addition to implementing as well as getting the requirements for MedIMG, the article explores the quality of medical data by using DL techniques for reaching large-scale labelled clinical data and also the significance of their real-time efforts in the biomedical study that have played an instrumental role in its extramural diffusion and global appeal.
Collapse
Affiliation(s)
- Surbhi Bhatia
- Department of Information Systems, College of Computer Science and Information Technology, King Faisal University, Al Hasa, Saudi Arabia,*Correspondence: Surbhi Bhatia
| | - Mohammed Alojail
- Department of Information Systems, College of Computer Science and Information Technology, King Faisal University, Al Hasa, Saudi Arabia
| | - Sudhakar Sengan
- Department of Computer Science and Engineering, PSN College of Engineering and Technology, Tirunelveli, India
| | - Pankaj Dadheech
- Department of Computer Science and Engineering, Swami Keshvanand Institute of Technology, Management & Gramothan (SKIT), Jaipur, India
| |
Collapse
|
6
|
A CAD System for Alzheimer's Disease Classification Using Neuroimaging MRI 2D Slices. COMPUTATIONAL AND MATHEMATICAL METHODS IN MEDICINE 2022; 2022:8680737. [PMID: 35983528 PMCID: PMC9381208 DOI: 10.1155/2022/8680737] [Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 02/17/2022] [Revised: 04/21/2022] [Accepted: 05/27/2022] [Indexed: 11/22/2022]
Abstract
Developments in medical care have inspired wide interest in the current decade, especially to their services to individuals living prolonged and healthier lives. Alzheimer's disease (AD) is the most chronic neurodegeneration and dementia-causing disorder. Economic expense of treating AD patients is expected to grow. The requirement of developing a computer-aided technique for early AD categorization becomes even more essential. Deep learning (DL) models offer numerous benefits against machine learning tools. Several latest experiments that exploited brain magnetic resonance imaging (MRI) scans and convolutional neural networks (CNN) for AD classification showed promising conclusions. CNN's receptive field aids in the extraction of main recognizable features from these MRI scans. In order to increase classification accuracy, a new adaptive model based on CNN and support vector machines (SVM) is presented in the research, combining both the CNN's capabilities in feature extraction and SVM in classification. The objective of this research is to build a hybrid CNN-SVM model for classifying AD using the MRI ADNI dataset. Experimental results reveal that the hybrid CNN-SVM model outperforms the CNN model alone, with relative improvements of 3.4%, 1.09%, 0.85%, and 2.82% on the testing dataset for AD vs. cognitive normal (CN), CN vs. mild cognitive impairment (MCI), AD vs. MCI, and CN vs. MCI vs. AD, respectively. Finally, the proposed approach has been further experimented on OASIS dataset leading to accuracy of 86.2%.
Collapse
|
7
|
Abstract
Forest fires are a significant threat to the ecological system’s stability. Several attempts have been made to detect forest fires using a variety of approaches, including optical fire sensors, and satellite-based technologies, all of which have been unsuccessful. In today’s world, research on flying ad hoc networks (FANETs) is a thriving field and can be used successfully. This paper describes a unique clustering approach that identifies the presence of a fire zone in a forest and transfers all sensed data to a base station as soon as feasible via wireless communication. The fire department takes the required steps to prevent the spread of the fire. It is proposed in this study that an efficient clustering approach be used to deal with routing and energy challenges to extend the lifetime of an unmanned aerial vehicle (UAV) in case of forest fires. Due to the restricted energy and high mobility, this directly impacts the flying duration and routing of FANET nodes. As a result, it is vital to enhance the lifetime of wireless sensor networks (WSNs) to maintain high system availability. Our proposed algorithm EE-SS regulates the energy usage of nodes while taking into account the features of a disaster region and other factors. For firefighting, sensor nodes are placed throughout the forest zone to collect essential data points for identifying forest fires and dividing them into distinct clusters. All of the sensor nodes in the cluster communicate their packets to the base station continually through the cluster head. When FANET nodes communicate with one another, their transmission range is constantly adjusted to meet their operating requirements. This paper examines the existing clustering techniques for forest fire detection approaches restricted to wireless sensor networks and their limitations. Our newly designed algorithm chooses the most optimum cluster heads (CHs) based on their fitness, reducing the routing overhead and increasing the system’s efficiency. Our proposed method results from simulations are compared with the existing approaches such as LEACH, LEACH-C, PSO-HAS, and SEED. The evaluation is carried out concerning overall energy usage, residual energy, the count of live nodes, the network lifetime, and the time it takes to build a cluster compared to other approaches. As a result, our proposed EE-SS algorithm outperforms all the considered state-of-art algorithms.
Collapse
|
8
|
A Novel Optimization for GPU Mining Using Overclocking and Undervolting. SUSTAINABILITY 2022. [DOI: 10.3390/su14148708] [Citation(s) in RCA: 9] [Impact Index Per Article: 4.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 01/27/2023]
Abstract
Cryptography and associated technologies have existed for a long time. This field is advancing at a remarkable speed. Since the inception of its initial application, blockchain has come a long way. Bitcoin is a cryptocurrency based on blockchain, also known as distributed ledger technology (DLT). The most well-known cryptocurrency for everyday use is Bitcoin, which debuted in 2008. Its success ushered in a digital revolution, and it currently provides security, decentralization, and a reliable data transport and storage mechanism to various industries and companies. Governments and developing enterprises seeking a competitive edge have expressed interest in Bitcoin and other cryptocurrencies due to the rapid growth of this recent technology. For computer experts and individuals looking for a method to supplement their income, cryptocurrency mining has become a big source of anxiety. Mining is a way of resolving mathematical problems based on the processing capacity and speed of the computers employed to solve them in return for the digital currency incentives. Herein, we have illustrated benefits of utilizing GPUs (graphical processing units) for cryptocurrency mining and compare two methods, namely overclocking and undervolting, which are the superior techniques when it comes to GPU optimization. The techniques we have used in this paper will not only help the miners to gain profits while mining cryptocurrency but also solve a major flaw; in order to mitigate the energy and resources that are consumed by the mining hardware, we have designed the mining hardware to simultaneously run longer and consume much less electricity. We have also compared our techniques with other popular techniques that are already in existence with respect to GPU mining.
Collapse
|
9
|
Energy Efficient Clustering Protocol for FANETS Using Moth Flame Optimization. SUSTAINABILITY 2022. [DOI: 10.3390/su14106159] [Citation(s) in RCA: 18] [Impact Index Per Article: 9.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
Abstract
FANET (flying ad-hoc networks) is currently a trending research topic. Unmanned aerial vehicles (UAVs) have two significant challenges: short flight times and inefficient routing due to low battery power and high mobility. Due to these topological restrictions, FANETS routing is considered more complicated than MANETs or VANETs. Clustering approaches based on artificial intelligence (AI) approaches can be used to solve complex routing issues when static and dynamic routings fail. Evolutionary algorithm-based clustering techniques, such as moth flame optimization, and ant colony optimization, can be used to solve these kinds of problems with routes. Moth flame optimization gives excellent coverage while consuming little energy and requiring a minimum number of cluster heads (CHs) for routing. This paper employs a moth flame optimization algorithm for network building and node deployment. Then, we employ a variation of the K-Means Density clustering approach to choosing the cluster head. Choosing the right cluster heads increases the cluster’s lifespan and reduces routing traffic. Moreover, it lowers the number of routing overheads. This step is followed by MRCQ image-based compression techniques to reduce the amount of data that must be transmitted. Finally, the reference point group mobility model is used to send data by the most optimal path. Particle swarm optimization (PSO), ant colony optimization (ACO), and grey wolf optimization (GWO) were put to the test against our proposed EECP-MFO. Several metrics are used to gauge the efficiency of our proposed method, including the number of clusters, cluster construction time, cluster lifespan, consistency of cluster heads, and energy consumption. This paper demonstrates that our proposed algorithm performance is superior to the current state-of-the-art approaches using experimental results.
Collapse
|