1
|
Hussain D, Hyeon Gu Y. Exploring the Impact of Noise and Image Quality on Deep Learning Performance in DXA Images. Diagnostics (Basel) 2024; 14:1328. [PMID: 39001219 PMCID: PMC11240833 DOI: 10.3390/diagnostics14131328] [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: 05/10/2024] [Revised: 06/19/2024] [Accepted: 06/20/2024] [Indexed: 07/16/2024] Open
Abstract
BACKGROUND AND OBJECTIVE Segmentation of the femur in Dual-Energy X-ray (DXA) images poses challenges due to reduced contrast, noise, bone shape variations, and inconsistent X-ray beam penetration. In this study, we investigate the relationship between noise and certain deep learning (DL) techniques for semantic segmentation of the femur to enhance segmentation and bone mineral density (BMD) accuracy by incorporating noise reduction methods into DL models. METHODS Convolutional neural network (CNN)-based models were employed to segment femurs in DXA images and evaluate the effects of noise reduction filters on segmentation accuracy and their effect on BMD calculation. Various noise reduction techniques were integrated into DL-based models to enhance image quality before training. We assessed the performance of the fully convolutional neural network (FCNN) in comparison to noise reduction algorithms and manual segmentation methods. RESULTS Our study demonstrated that the FCNN outperformed noise reduction algorithms in enhancing segmentation accuracy and enabling precise calculation of BMD. The FCNN-based segmentation approach achieved a segmentation accuracy of 98.84% and a correlation coefficient of 0.9928 for BMD measurements, indicating its effectiveness in the clinical diagnosis of osteoporosis. CONCLUSIONS In conclusion, integrating noise reduction techniques into DL-based models significantly improves femur segmentation accuracy in DXA images. The FCNN model, in particular, shows promising results in enhancing BMD calculation and clinical diagnosis of osteoporosis. These findings highlight the potential of DL techniques in addressing segmentation challenges and improving diagnostic accuracy in medical imaging.
Collapse
Affiliation(s)
- Dildar Hussain
- Department of Artificial Intelligence and Data Science, Sejong University, Seoul 05006, Republic of Korea
| | - Yeong Hyeon Gu
- Department of Artificial Intelligence and Data Science, Sejong University, Seoul 05006, Republic of Korea
| |
Collapse
|
2
|
Kim K, Cha BK, Jeong HW, Lee Y. Investigation of Deconvolution Method with Adaptive Point Spread Function Based on Scintillator Thickness in Wavelet Domain. Bioengineering (Basel) 2024; 11:330. [PMID: 38671752 PMCID: PMC11048213 DOI: 10.3390/bioengineering11040330] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/07/2024] [Revised: 03/25/2024] [Accepted: 03/26/2024] [Indexed: 04/28/2024] Open
Abstract
In recent years, indirect digital radiography detectors have been actively studied to improve radiographic image performance with low radiation exposure. This study aimed to achieve low-dose radiation imaging with a thick scintillation detector while simultaneously obtaining the resolution of a thin scintillation detector. The proposed method was used to predict the optimal point spread function (PSF) between thin and thick scintillation detectors by considering image quality assessment (IQA). The process of identifying the optimal PSF was performed on each sub-band in the wavelet domain to improve restoration accuracy. In the experiments, the edge preservation index (EPI) values of the non-blind deblurred image with a blurring sigma of σ = 5.13 pixels and the image obtained with optimal parameters from the thick scintillator using the proposed method were approximately 0.62 and 0.76, respectively. The coefficient of variation (COV) values for the two images were approximately 1.02 and 0.63, respectively. The proposed method was validated through simulations and experimental results, and its viability is expected to be verified on various radiological imaging systems.
Collapse
Affiliation(s)
- Kyuseok Kim
- Department of Biomedical Engineering, Eulji University, 553, Sanseong-daero, Sujeong-gu, Seongnam-si 13135, Republic of Korea;
| | - Bo Kyung Cha
- Precision Medical Device Research Center, Korea Electrotechnology Research Institute (KERI), 111 Hanggaul-ro, Sangnok-gu, Ansan-si 15588, Republic of Korea;
| | - Hyun-Woo Jeong
- Department of Biomedical Engineering, Eulji University, 553, Sanseong-daero, Sujeong-gu, Seongnam-si 13135, Republic of Korea;
| | - Youngjin Lee
- Department of Radiological Science, Gachon University, Incheon 21936, Republic of Korea
| |
Collapse
|
3
|
Guo H, Liu H, Zhu H, Li M, Yu H, Zhu Y, Chen X, Xu Y, Gao L, Zhang Q, Shentu Y. Exploring a novel HE image segmentation technique for glioblastoma: A hybrid slime mould and differential evolution approach. Comput Biol Med 2024; 168:107653. [PMID: 37984200 DOI: 10.1016/j.compbiomed.2023.107653] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/03/2023] [Revised: 10/12/2023] [Accepted: 10/31/2023] [Indexed: 11/22/2023]
Abstract
Glioblastoma is a primary brain tumor with high incidence and mortality rates, posing a significant threat to human health. It is crucial to provide necessary diagnostic assistance for its management. Among them, Multi-threshold Image Segmentation (MIS) is considered the most efficient and intuitive method in image processing. In recent years, many scholars have combined different metaheuristic algorithms with MIS to improve the quality of Image Segmentation (IS). Slime Mould Algorithm (SMA) is a metaheuristic approach inspired by the foraging behavior of slime mould populations in nature. In this investigation, we introduce a hybridized variant named BDSMA, aimed at overcoming the inherent limitations of the original algorithm. These limitations encompass inadequate exploitation capacity and a tendency to converge prematurely towards local optima when dealing with complex multidimensional problems. To bolster the algorithm's optimization prowess, we integrate the original algorithm with a robust exploitative operator called Differential Evolution (DE). Additionally, we introduce a strategy for handling solutions that surpass boundaries. The incorporation of an advanced cooperative mixing model accelerates the convergence of BDSMA, refining its precision and preventing it from becoming trapped in local optima. To substantiate the effectiveness of our proposed approach, we conduct a comprehensive series of comparative experiments involving 30 benchmark functions. The results of these experiments demonstrate the superiority of our method in terms of both convergence speed and precision. Moreover, within this study, we propose a MIS technique. This technique is subsequently employed to conduct experiments on IS at both low and high threshold levels. The effectiveness of the BDSMA-based MIS technique is further showcased through its successful application to the medical image of brain glioblastoma. The evaluation of these experimental outcomes, utilizing image quality metrics, conclusively underscores the exceptional efficacy of the algorithm we have put forth.
Collapse
Affiliation(s)
- Hongliang Guo
- College of Information Technology, Jilin Agricultural University, Changchun 130118, China.
| | - Hanbo Liu
- College of Information Technology, Jilin Agricultural University, Changchun 130118, China.
| | - Hong Zhu
- Department of Pathology, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China.
| | - Mingyang Li
- College of Information Technology, Jilin Agricultural University, Changchun 130118, China.
| | - Helong Yu
- College of Information Technology, Jilin Agricultural University, Changchun 130118, China.
| | - Yun Zhu
- Department of Pathology, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China.
| | - Xiaoxiao Chen
- Department of Pathology, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China.
| | - Yujia Xu
- Department of Pathology, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China.
| | - Lianxing Gao
- College of Engineering and Technology, Jilin Agricultural University, Changchun 130118, China.
| | - Qiongying Zhang
- Department of Pathology, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China.
| | - Yangping Shentu
- Department of Pathology, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China.
| |
Collapse
|
4
|
Chen J, Cai Z, Chen H, Chen X, Escorcia-Gutierrez J, Mansour RF, Ragab M. Renal Pathology Images Segmentation Based on Improved Cuckoo Search with Diffusion Mechanism and Adaptive Beta-Hill Climbing. JOURNAL OF BIONIC ENGINEERING 2023; 20:1-36. [PMID: 37361683 PMCID: PMC10154766 DOI: 10.1007/s42235-023-00365-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 10/29/2022] [Revised: 03/03/2023] [Accepted: 03/07/2023] [Indexed: 06/28/2023]
Abstract
Lupus Nephritis (LN) is a significant risk factor for morbidity and mortality in systemic lupus erythematosus, and nephropathology is still the gold standard for diagnosing LN. To assist pathologists in evaluating histopathological images of LN, a 2D Rényi entropy multi-threshold image segmentation method is proposed in this research to apply to LN images. This method is based on an improved Cuckoo Search (CS) algorithm that introduces a Diffusion Mechanism (DM) and an Adaptive β-Hill Climbing (AβHC) strategy called the DMCS algorithm. The DMCS algorithm is tested on 30 benchmark functions of the IEEE CEC2017 dataset. In addition, the DMCS-based multi-threshold image segmentation method is also used to segment renal pathological images. Experimental results show that adding these two strategies improves the DMCS algorithm's ability to find the optimal solution. According to the three image quality evaluation metrics: PSNR, FSIM, and SSIM, the proposed image segmentation method performs well in image segmentation experiments. Our research shows that the DMCS algorithm is a helpful image segmentation method for renal pathological images.
Collapse
Affiliation(s)
- Jiaochen Chen
- College of Computer Science and Artificial Intelligence, Wenzhou University, Wenzhou, 325035 China
| | - Zhennao Cai
- College of Computer Science and Artificial Intelligence, Wenzhou University, Wenzhou, 325035 China
| | - Huiling Chen
- College of Computer Science and Artificial Intelligence, Wenzhou University, Wenzhou, 325035 China
| | - Xiaowei Chen
- Department of Rheumatology and Immunology, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, 325000 China
| | - José Escorcia-Gutierrez
- Department of Computational Science and Electronics, Universidad de la Costa, CUC, 080002 Barranquilla, Colombia
| | - Romany F. Mansour
- Department of Mathematics, Faculty of Science, New Valley University, 72511, El-Kharga, Egypt
| | - Mahmoud Ragab
- Information Technology Department, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah, 21589 Saudi Arabia
- Department of Mathematics, Faculty of Science, Al-Azhar University, Naser City, Cairo, 11884 Egypt
| |
Collapse
|
5
|
Han B, Park M, Kim K, Lee Y. Characterization of Flexible Amorphous Silicon Thin-Film Transistor-Based Detectors with Positive-Intrinsic-Negative Diode in Radiography. Diagnostics (Basel) 2022; 12:diagnostics12092103. [PMID: 36140503 PMCID: PMC9497934 DOI: 10.3390/diagnostics12092103] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/08/2022] [Revised: 08/26/2022] [Accepted: 08/29/2022] [Indexed: 11/16/2022] Open
Abstract
Low-dose exposure and work convenience are required for mobile X-ray systems during the COVID-19 pandemic. We investigated a novel X-ray detector (FXRD-4343FAW, VIEWORKS, Anyang, Korea) composed of a thin-film transistor based on amorphous silicon with a flexible plastic substrate. This detector is composed of a thallium-doped cesium iodide scintillator with a pixel size of 99 μm, pixel matrix of 4316 × 4316, and weight of 2.95 kg. The proposed detector has the advantages of high-noise characteristics and low weight, which provide patients and workers with an advantage in terms of the dose and work efficiency, respectively. We performed a quantitative evaluation and an experiment to demonstrate its viability. The modulation transfer function, noise power spectrum, and detective quantum efficiency were identified using the proposed and comparative detectors, according to the International Electrotechnical Commission protocol. Additionally, the contrast-to-noise ratio and coefficient of variation were investigated using a human-like phantom. Our results indicate that the proposed detector efficiently increases the image performance in terms of noise characteristics. The detailed performance evaluation demonstrated that the outcomes of the use of the proposed detector confirmed the viability of mobile X-ray devices that require low doses. Consequently, the novel FXRD-4343FAW X-ray detector is expected to improve the image quality and work convenience in extended radiography.
Collapse
Affiliation(s)
- Bongju Han
- Quality Assurance Team, Business Division, Vieworks, 41-3, Burim-ro 170beon-gil, Dongan-gu, Anyang-si 14055, Korea
| | - Minji Park
- Department of Radiological Science, College of Health Science, Gachon University, 191, Hambakmoe-ro, Yeonsu-gu, Incheon 21936, Korea
- Department of Health Science, General Graduate School of Gachon University, 191, Hambakmoe-ro, Yeonsu-gu, Incheon 21936, Korea
| | - Kyuseok Kim
- Department of Integrative Medicine, Major in Digital Healthcare, Yonsei University College of Medicine, Unju-ro, Gangman-gu, Seoul 06229, Korea
- Correspondence: (K.K.); (Y.L.); Tel.: +82-02-2019-5447 (K.K.); +82-32-820-4362 (Y.L.)
| | - Youngjin Lee
- Department of Radiological Science, College of Health Science, Gachon University, 191, Hambakmoe-ro, Yeonsu-gu, Incheon 21936, Korea
- Correspondence: (K.K.); (Y.L.); Tel.: +82-02-2019-5447 (K.K.); +82-32-820-4362 (Y.L.)
| |
Collapse
|
6
|
Taresh MM, Zhu N, Ali TAA, Alghaili M, Hameed AS, Mutar ML. KL-MOB: automated COVID-19 recognition using a novel approach based on image enhancement and a modified MobileNet CNN. PeerJ Comput Sci 2021; 7:e694. [PMID: 34616885 PMCID: PMC8459788 DOI: 10.7717/peerj-cs.694] [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: 05/13/2021] [Accepted: 08/05/2021] [Indexed: 06/13/2023]
Abstract
The emergence of the novel coronavirus pneumonia (COVID-19) pandemic at the end of 2019 led to worldwide chaos. However, the world breathed a sigh of relief when a few countries announced the development of a vaccine and gradually began to distribute it. Nevertheless, the emergence of another wave of this pandemic returned us to the starting point. At present, early detection of infected people is the paramount concern of both specialists and health researchers. This paper proposes a method to detect infected patients through chest x-ray images by using the large dataset available online for COVID-19 (COVIDx), which consists of 2128 X-ray images of COVID-19 cases, 8,066 normal cases, and 5,575 cases of pneumonia. A hybrid algorithm is applied to improve image quality before undertaking neural network training. This algorithm combines two different noise-reduction filters in the image, followed by a contrast enhancement algorithm. To detect COVID-19, we propose a novel convolution neural network (CNN) architecture called KL-MOB (COVID-19 detection network based on the MobileNet structure). The performance of KL-MOB is boosted by adding the Kullback-Leibler (KL) divergence loss function when trained from scratch. The KL divergence loss function is adopted for content-based image retrieval and fine-grained classification to improve the quality of image representation. The results are impressive: the overall benchmark accuracy, sensitivity, specificity, and precision are 98.7%, 98.32%, 98.82% and 98.37%, respectively. These promising results should help other researchers develop innovative methods to aid specialists. The tremendous potential of the method proposed herein can also be used to detect COVID-19 quickly and safely in patients throughout the world.
Collapse
Affiliation(s)
| | - Ningbo Zhu
- College of Information Science and Engineering, Hunan University, Changsha, Hunan, China
| | - Talal Ahmed Ali Ali
- College of Information Science and Engineering, Hunan University, Changsha, Hunan, China
| | - Mohammed Alghaili
- College of Information Science and Engineering, Hunan University, Changsha, Hunan, China
| | - Asaad Shakir Hameed
- Department of Mathematics, General Directorate of Thi-Qar Education, Ministry of Education, Thi-Qar, Iraq
| | - Modhi Lafta Mutar
- Department of Mathematics, General Directorate of Thi-Qar Education, Ministry of Education, Thi-Qar, Iraq
| |
Collapse
|
7
|
Kubicek J, Strycek M, Cerny M, Penhaker M, Prokop O, Vilimek D. Quantitative and Comparative Analysis of Effectivity and Robustness for Enhanced and Optimized Non-Local Mean Filter Combining Pixel and Patch Information on MR Images of Musculoskeletal System. SENSORS 2021; 21:s21124161. [PMID: 34204477 PMCID: PMC8233799 DOI: 10.3390/s21124161] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 04/30/2021] [Revised: 06/06/2021] [Accepted: 06/11/2021] [Indexed: 12/27/2022]
Abstract
In the area of musculoskeletal MR images analysis, the image denoising plays an important role in enhancing the spatial image area for further processing. Recent studies have shown that non-local means (NLM) methods appear to be more effective and robust when compared with conventional local statistical filters, including median or average filters, when Rician noise is presented. A significant limitation of NLM is the fact that thy have the tendency to suppress tiny objects, which may represent clinically important information. For this reason, we provide an extensive quantitative and objective analysis of a novel NLM algorithm, taking advantage of pixel and patch similarity information with the optimization procedure for optimal filter parameters selection to demonstrate a higher robustness and effectivity, when comparing with NLM and conventional local means methods, including average and median filters. We provide extensive testing on variable noise generators with dynamical noise intensity to objectively demonstrate the robustness of the method in a noisy environment, which simulates relevant, variable and real conditions. This work also objectively evaluates the potential and benefits of the application of NLM filters in contrast to conventional local-mean filters. The final part of the analysis is focused on the segmentation performance when an NLM filter is applied. This analysis demonstrates a better performance of tissue identification with the application of smoothing procedure under worsening image conditions.
Collapse
Affiliation(s)
- Jan Kubicek
- Department of Cybernetics and Biomedical Engineering, VSB–Technical University of Ostrava, 17. listopadu 15, 70800 Ostrava Poruba, Czech Republic; (M.S.); (M.C.); (M.P.); (D.V.)
- Correspondence:
| | - Michal Strycek
- Department of Cybernetics and Biomedical Engineering, VSB–Technical University of Ostrava, 17. listopadu 15, 70800 Ostrava Poruba, Czech Republic; (M.S.); (M.C.); (M.P.); (D.V.)
| | - Martin Cerny
- Department of Cybernetics and Biomedical Engineering, VSB–Technical University of Ostrava, 17. listopadu 15, 70800 Ostrava Poruba, Czech Republic; (M.S.); (M.C.); (M.P.); (D.V.)
| | - Marek Penhaker
- Department of Cybernetics and Biomedical Engineering, VSB–Technical University of Ostrava, 17. listopadu 15, 70800 Ostrava Poruba, Czech Republic; (M.S.); (M.C.); (M.P.); (D.V.)
| | - Ondrej Prokop
- MEDIN, a.s., Vlachovicka 619, 59231 Nove Mesto na Morave, Czech Republic;
| | - Dominik Vilimek
- Department of Cybernetics and Biomedical Engineering, VSB–Technical University of Ostrava, 17. listopadu 15, 70800 Ostrava Poruba, Czech Republic; (M.S.); (M.C.); (M.P.); (D.V.)
| |
Collapse
|
8
|
Zhao S, Wang P, Heidari AA, Chen H, Turabieh H, Mafarja M, Li C. Multilevel threshold image segmentation with diffusion association slime mould algorithm and Renyi's entropy for chronic obstructive pulmonary disease. Comput Biol Med 2021; 134:104427. [PMID: 34020128 DOI: 10.1016/j.compbiomed.2021.104427] [Citation(s) in RCA: 42] [Impact Index Per Article: 14.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/17/2021] [Revised: 04/16/2021] [Accepted: 04/20/2021] [Indexed: 01/11/2023]
Abstract
Image segmentation is an essential pre-processing step and is an indispensable part of image analysis. This paper proposes Renyi's entropy multi-threshold image segmentation based on an improved Slime Mould Algorithm (DASMA). First, we introduce the diffusion mechanism (DM) into the original SMA to increase the population's diversity so that the variants can better avoid falling into local optima. The association strategy (AS) is then added to help the algorithm find the optimal solution faster. Finally, the proposed algorithm is applied to Renyi's entropy multilevel threshold image segmentation based on non-local means 2D histogram. The proposed method's effectiveness is demonstrated on the Berkeley segmentation dataset and benchmark (BSD) by comparing it with some well-known algorithms. The DASMA-based multilevel threshold segmentation technique is also successfully applied to the CT image segmentation of chronic obstructive pulmonary disease (COPD). The experimental results are evaluated by image quality metrics, which show the proposed algorithm's extraordinary performance. This means that it can help doctors analyze the lesion tissue qualitatively and quantitatively, improve its diagnostic accuracy and make the right treatment plan. The supplementary material and info about this article will be available at https://aliasgharheidari.com.
Collapse
Affiliation(s)
- Songwei Zhao
- College of Computer Science and Artificial Intelligence, Wenzhou University, Wenzhou, Zhejiang, 325035, China.
| | - Pengjun Wang
- College of Electrical and Electronic Engineering, Wenzhou University, Wenzhou, 325035, China.
| | - Ali Asghar Heidari
- College of Computer Science and Artificial Intelligence, Wenzhou University, Wenzhou, Zhejiang, 325035, China; Department of Computer Science, School of Computing, National University of Singapore, Singapore.
| | - Huiling Chen
- College of Computer Science and Artificial Intelligence, Wenzhou University, Wenzhou, Zhejiang, 325035, China.
| | - Hamza Turabieh
- Department of Information Technology, College of Computers and Information Technology, P.O. Box11099, Taif, 21944, Taif University, Taif, Saudi Arabia.
| | - Majdi Mafarja
- Department of Computer Science, Birzeit University, Birzeit 72439, Palestine.
| | - Chengye Li
- Department of Pulmonary and Critical Care Medicine, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, 325000, China.
| |
Collapse
|
9
|
Ladisa M, Lamura A. Diffusion-Driven X-Ray Two-Dimensional Patterns Denoising. MATERIALS (BASEL, SWITZERLAND) 2020; 13:ma13122773. [PMID: 32570931 PMCID: PMC7344928 DOI: 10.3390/ma13122773] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 05/19/2020] [Revised: 06/15/2020] [Accepted: 06/16/2020] [Indexed: 06/11/2023]
Abstract
The use of a mathematical model is proposed in order to denoise X-ray two-dimensional patterns. The method relies on a generalized diffusion equation whose diffusion constant depends on the image gradients. The numerical solution of the diffusion equation provides an efficient reduction of pattern noise as witnessed by the computed peak of signal-to-noise ratio. The use of experimental data with different inherent levels of noise allows us to show the success of the method even in the case, experimentally relevant, when patterns are blurred by Poissonian noise. The corresponding MatLab code for the numerical method is made available.
Collapse
|