1
|
Erten M, Tuncer I, Barua PD, Yildirim K, Dogan S, Tuncer T, Tan RS, Fujita H, Acharya UR. Automated Urine Cell Image Classification Model Using Chaotic Mixer Deep Feature Extraction. J Digit Imaging 2023; 36:1675-1686. [PMID: 37131063 PMCID: PMC10407001 DOI: 10.1007/s10278-023-00827-8] [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: 02/09/2023] [Revised: 03/28/2023] [Accepted: 03/30/2023] [Indexed: 05/04/2023] Open
Abstract
Microscopic examination of urinary sediments is a common laboratory procedure. Automated image-based classification of urinary sediments can reduce analysis time and costs. Inspired by cryptographic mixing protocols and computer vision, we developed an image classification model that combines a novel Arnold Cat Map (ACM)- and fixed-size patch-based mixer algorithm with transfer learning for deep feature extraction. Our study dataset comprised 6,687 urinary sediment images belonging to seven classes: Cast, Crystal, Epithelia, Epithelial nuclei, Erythrocyte, Leukocyte, and Mycete. The developed model consists of four layers: (1) an ACM-based mixer to generate mixed images from resized 224 × 224 input images using fixed-size 16 × 16 patches; (2) DenseNet201 pre-trained on ImageNet1K to extract 1,920 features from each raw input image, and its six corresponding mixed images were concatenated to form a final feature vector of length 13,440; (3) iterative neighborhood component analysis to select the most discriminative feature vector of optimal length 342, determined using a k-nearest neighbor (kNN)-based loss function calculator; and (4) shallow kNN-based classification with ten-fold cross-validation. Our model achieved 98.52% overall accuracy for seven-class classification, outperforming published models for urinary cell and sediment analysis. We demonstrated the feasibility and accuracy of deep feature engineering using an ACM-based mixer algorithm for image preprocessing combined with pre-trained DenseNet201 for feature extraction. The classification model was both demonstrably accurate and computationally lightweight, making it ready for implementation in real-world image-based urine sediment analysis applications.
Collapse
Affiliation(s)
- Mehmet Erten
- Department of Medical Biochemistry, Malatya Training and Research Hospital, Malatya, Türkiye
| | - Ilknur Tuncer
- Elazig Governorship, Interior Ministry, Elazig, Türkiye
| | - Prabal D. Barua
- Cogninet Australia, Sydney, NSW 2010 Australia
- School of Business (Information System), University of Southern Queensland, Toowoomba, Australia
- Faculty of Engineering and Information Technology, University of Technology Sydney, Sydney, NSW 2007 Australia
- Australian International Institute of Higher Education, Sydney, NSW 2000 Australia
- School of Science and Technology, University of New England, Armidale, Australia
- School of Biosciences, Taylor’s University, Subang Jaya, Malaysia
- School of Computing, SRM Institute of Science and Technology, Chennai, India
- School of Science and Technology, Kumamoto University, Kumamoto, Japan
- Sydney School of Education and Social Work, University of Sydney, Sydney, Australia
| | - Kubra Yildirim
- Department of Digital Forensics Engineering, Technology Faculty, Firat University, Elazig, Türkiye
| | - Sengul Dogan
- Department of Digital Forensics Engineering, Technology Faculty, Firat University, Elazig, Türkiye
| | - Turker Tuncer
- Department of Digital Forensics Engineering, Technology Faculty, Firat University, Elazig, Türkiye
| | - Ru-San Tan
- Department of Cardiology, National Heart Centre Singapore, Singapore, Singapore
- Duke-NUS Medical School, Singapore, Singapore
| | - Hamido Fujita
- Faculty of Information Technology, HUTECH University, Ho Chi Minh City, Vietnam
- Andalusian Research Institute in Data Science and Computational Intelligence, University of Granada, Granada, Spain
- Regional Research Center, Iwate Prefectural University, Iwate, Japan
| | - U. Rajendra Acharya
- School of Mathematics, Physics and Computing, University of Southern Queensland, Springfield, Australia
| |
Collapse
|
2
|
DİNKÇİ S, KİBAR F, DEMİR E, PAYDAS S, ERDOĞAN S, YAMAN A. Frequency of pre- and post-transplant infectious agents in kidney transplant patients. CUKUROVA MEDICAL JOURNAL 2022. [DOI: 10.17826/cumj.1099130] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/07/2022] Open
Abstract
Purpose: Renal transplantation is the most important and successful treatment method for renal failure. In this study, it was aimed to investigate the frequency of Cytomegalovirus (CMV), BK virus (BKV) and bacterial agents in kidney transplant recipient (KTR)s before and in the first six months after transplantation.
Materials and Methods: CMV and BKV were investigated by Real-time PCR in blood samples taken from patients who underwent kidney transplantation at the Organ Transplantation Center of our faculty, one week before the transplantation and in the first, third and sixth months after transplantation. Blood, urine, respiratory tract /wound (if necessary) cultures were performed. Decoy cells were evaluated in urine cytology.
Results: The mean age of KTRs was 32.60±11.71 years, 28 (62.2%) were male. Donor origins were living related donors 39 (86.7%) and cadaveric 6 (13.3%). After transplantation, BKV was detected in 11/38 (28.9%) patients, CMV was found in 25/41 (60.9%) patients, and Decoy cell positivity was detected in 11/31 (35.4%) patients. While the highest rate of Real-time PCR positivities were in the third months and sixth months for BKV and first, month for CMV and gradually decreased towards the sixth month. Escherichia coli, Klebsiella pneumoniae, Candida nonalbicans, Enterococcus faecalis were most commonly grown in urine culture. Staphylococcus hominis, Streptecoccus epidermidis, were grown in blood culture. Acinetobacter baumannii, Klebsiella pneumoniae, Aspergillus fumigatus and Candida albicans grew in the culture of respiratory tract samples.
Conclusion: Bacterial infections developed early in our KTRs. While the highest Real-time PCR positivity rate was in the third and sixth months for BKV, it was the first month for CMV and gradually decreased towards the sixth month. Decoy cell positivity may be also important for diagnosis of BKV infection in KTRs.
Collapse
Affiliation(s)
- Suzan DİNKÇİ
- ÇUKUROVA ÜNİVERSİTESİ, TIP FAKÜLTESİ, DAHİLİ TIP BİLİMLERİ BÖLÜMÜ, İÇ HASTALIKLARI ANABİLİM DALI, ROMATOLOJİ BİLİM DALI
| | - Filiz KİBAR
- ÇUKUROVA ÜNİVERSİTESİ, TIP FAKÜLTESİ, TEMEL TIP BİLİMLERİ BÖLÜMÜ, MİKROBİYOLOJİ ANABİLİM DALI, TIBBİ MİKROBİYOLOJİ BİLİM DALI
| | - Erkan DEMİR
- ÇUKUROVA ÜNİVERSİTESİ, TIP FAKÜLTESİ, CERRAHİ TIP BİLİMLERİ BÖLÜMÜ, ÜROLOJİ ANABİLİM DALI
| | - Saime PAYDAS
- ÇUKUROVA ÜNİVERSİTESİ, TIP FAKÜLTESİ, DAHİLİ TIP BİLİMLERİ BÖLÜMÜ, İÇ HASTALIKLARI ANABİLİM DALI, NEFROLOJİ BİLİM DALI
| | - Seyda ERDOĞAN
- ÇUKUROVA ÜNİVERSİTESİ, TIP FAKÜLTESİ, CERRAHİ TIP BİLİMLERİ BÖLÜMÜ, TIBBİ PATOLOJİ ANABİLİM DALI
| | - Akgün YAMAN
- ÇUKUROVA ÜNİVERSİTESİ, TIP FAKÜLTESİ, TEMEL TIP BİLİMLERİ BÖLÜMÜ, MİKROBİYOLOJİ ANABİLİM DALI, TIBBİ MİKROBİYOLOJİ BİLİM DALI
| |
Collapse
|
3
|
Yan L, Guo H, Han L, Huang H, Shen Y, He J, Liu J. Sternheimer-Malbin Staining to Detect Decoy Cells in Urine of 213 Kidney Transplant Patients. Transplant Proc 2020; 52:823-828. [PMID: 32111385 DOI: 10.1016/j.transproceed.2020.01.044] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/28/2019] [Revised: 10/11/2019] [Accepted: 01/25/2020] [Indexed: 11/18/2022]
Abstract
BACKGROUND Human polyoma virus-associated nephropathy frequently refers to allograft failure after kidney transplant. Thus, the early detection of viral activation is extremely important for these immunocompromised patients. METHODS Previously, urine polyoma virus-infected cells (decoy cells) were indicated as the virus action, usually screened by the routine papanicolaou cytology in renal biopsy, but these methods are complex and the positive rate is low. In this article, the direct microscopy observation method, Wright-Giemsa staining, and Sternheimer-Malbin (SM) staining were all used to screen the decoy cells in urine samples of 213 kidney transplant patients who had used immunosuppressive drugs. RESULTS Among them, decoy cells were detected in 40 cases (18.8%) by the direct observation method, 44 cases (20.7%) by Wright-Giemsa staining and 49 cases (23.0%) by SM staining. Furthermore, the most common polyoma viruses, BK and JC viruses, were also confirmed in 41 (83.7%) cases among these 49 decoy cell-positive samples. Importantly, compared with other decoy cell detection methods, SM staining is fast, easy to operate, and has a high positive rate. CONCLUSION Therefore, SM staining is recommended as a fast and effective method for screening urine decoy cells in kidney transplant patients.
Collapse
Affiliation(s)
- Lizhi Yan
- Department of Clinical Laboratory, Inner Mongolia Baogang Hospital, Baotou, Inner Mongolia, China
| | - Hongbo Guo
- Department of Clinical Laboratory, Inner Mongolia Baogang Hospital, Baotou, Inner Mongolia, China
| | - Lizhong Han
- Department of Urology, Inner Mongolia Baogang Hospital, Baotou, Inner Mongolia, China
| | - Hualiang Huang
- Department of Clinical Laboratory, Inner Mongolia Baogang Hospital, Baotou, Inner Mongolia, China
| | - Yan Shen
- Department of Clinical Laboratory, Zhejiang Provincial People's Hospital, Hangzhou Medical College, Hangzhou, China
| | - Jing He
- Department of Clinical Laboratory, Zhejiang Provincial People's Hospital, Hangzhou Medical College, Hangzhou, China
| | - Jinlin Liu
- Department of Clinical Laboratory, Zhejiang Provincial People's Hospital, Hangzhou Medical College, Hangzhou, China; Key Laboratory of Tumor Molecular Diagnosis and Individualized Medicine of Zhejiang Province, Hangzhou, China.
| |
Collapse
|