1
|
Das BK, Zhao G, Islam S, Re TJ, Comaniciu D, Gibson E, Maier A. Co-ordinate-based positional embedding that captures resolution to enhance transformer's performance in medical image analysis. Sci Rep 2024; 14:9380. [PMID: 38654066 DOI: 10.1038/s41598-024-59813-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/30/2023] [Accepted: 04/15/2024] [Indexed: 04/25/2024] Open
Abstract
Vision transformers (ViTs) have revolutionized computer vision by employing self-attention instead of convolutional neural networks and demonstrated success due to their ability to capture global dependencies and remove spatial biases of locality. In medical imaging, where input data may differ in size and resolution, existing architectures require resampling or resizing during pre-processing, leading to potential spatial resolution loss and information degradation. This study proposes a co-ordinate-based embedding that encodes the geometry of medical images, capturing physical co-ordinate and resolution information without the need for resampling or resizing. The effectiveness of the proposed embedding is demonstrated through experiments with UNETR and SwinUNETR models for infarct segmentation on MRI dataset with AxTrace and AxADC contrasts. The dataset consists of 1142 training, 133 validation and 143 test subjects. Both models with the addition of co-ordinate based positional embedding achieved substantial improvements in mean Dice score by 6.5% and 7.6%. The proposed embedding showcased a statistically significant advantage p-value< 0.0001 over alternative approaches. In conclusion, the proposed co-ordinate-based pixel-wise positional embedding method offers a promising solution for Transformer-based models in medical image analysis. It effectively leverages physical co-ordinate information to enhance performance without compromising spatial resolution and provides a foundation for future advancements in positional embedding techniques for medical applications.
Collapse
Affiliation(s)
- Badhan Kumar Das
- Digital Technology and Innovation, Siemens Healthineers, Erlangen, Germany.
- Pattern Recognition Lab, Department of Computer Science, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany.
| | - Gengyan Zhao
- Digital Technology and Innovation, Siemens Healthineers, Princeton, NJ, USA
| | - Saahil Islam
- Digital Technology and Innovation, Siemens Healthineers, Erlangen, Germany
- Pattern Recognition Lab, Department of Computer Science, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany
| | - Thomas J Re
- Digital Technology and Innovation, Siemens Healthineers, Princeton, NJ, USA
| | - Dorin Comaniciu
- Digital Technology and Innovation, Siemens Healthineers, Princeton, NJ, USA
| | - Eli Gibson
- Digital Technology and Innovation, Siemens Healthineers, Princeton, NJ, USA
| | - Andreas Maier
- Pattern Recognition Lab, Department of Computer Science, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany
| |
Collapse
|
2
|
Cong L, He Y, Wu Y, Li Z, Ding S, Liang W, Xiao X, Zhang H, Wang L. Discovery and validation of molecular patterns and immune characteristics in the peripheral blood of ischemic stroke patients. PeerJ 2024; 12:e17208. [PMID: 38650649 PMCID: PMC11034498 DOI: 10.7717/peerj.17208] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/16/2023] [Accepted: 03/18/2024] [Indexed: 04/25/2024] Open
Abstract
Background Stroke is a disease with high morbidity, disability, and mortality. Immune factors play a crucial role in the occurrence of ischemic stroke (IS), but their exact mechanism is not clear. This study aims to identify possible immunological mechanisms by recognizing immune-related biomarkers and evaluating the infiltration pattern of immune cells. Methods We downloaded datasets of IS patients from GEO, applied R language to discover differentially expressed genes, and elucidated their biological functions using GO, KEGG analysis, and GSEA analysis. The hub genes were then obtained using two machine learning algorithms (least absolute shrinkage and selection operator (LASSO) and support vector machine-recursive feature elimination (SVM-RFE)) and the immune cell infiltration pattern was revealed by CIBERSORT. Gene-drug target networks and mRNA-miRNA-lncRNA regulatory networks were constructed using Cytoscape. Finally, we used RT-qPCR to validate the hub genes and applied logistic regression methods to build diagnostic models validated with ROC curves. Results We screened 188 differentially expressed genes whose functional analysis was enriched to multiple immune-related pathways. Six hub genes (ANTXR2, BAZ2B, C5AR1, PDK4, PPIH, and STK3) were identified using LASSO and SVM-RFE. ANTXR2, BAZ2B, C5AR1, PDK4, and STK3 were positively correlated with neutrophils and gamma delta T cells, and negatively correlated with T follicular helper cells and CD8, while PPIH showed the exact opposite trend. Immune infiltration indicated increased activity of monocytes, macrophages M0, neutrophils, and mast cells, and decreased infiltration of T follicular helper cells and CD8 in the IS group. The ceRNA network consisted of 306 miRNA-mRNA interacting pairs and 285 miRNA-lncRNA interacting pairs. RT-qPCR results indicated that the expression levels of BAZ2B, C5AR1, PDK4, and STK3 were significantly increased in patients with IS. Finally, we developed a diagnostic model based on these four genes. The AUC value of the model was verified to be 0.999 in the training set and 0.940 in the validation set. Conclusion Our research explored the immune-related gene expression modules and provided a specific basis for further study of immunomodulatory therapy of IS.
Collapse
Affiliation(s)
- Lin Cong
- Department of Neurology, The Second Affiliated Hospital of Harbin Medical University, Harbin Medical University, Harbin, China
| | - Yijie He
- Department of Neurology, The Second Affiliated Hospital of Harbin Medical University, Harbin Medical University, Harbin, China
| | - Yun Wu
- Department of Neurology, The Second Affiliated Hospital of Harbin Medical University, Harbin Medical University, Harbin, China
| | - Ze Li
- Department of Neurology, The Second Affiliated Hospital of Harbin Medical University, Harbin Medical University, Harbin, China
| | - Siwen Ding
- Department of Neurology, The Second Affiliated Hospital of Harbin Medical University, Harbin Medical University, Harbin, China
| | - Weiwei Liang
- Department of Neurology, The Second Affiliated Hospital of Harbin Medical University, Harbin Medical University, Harbin, China
| | - Xingjun Xiao
- Department of Neurology, The Second Affiliated Hospital of Harbin Medical University, Harbin Medical University, Harbin, China
| | - Huixue Zhang
- Department of Neurology, The Second Affiliated Hospital of Harbin Medical University, Harbin Medical University, Harbin, China
| | - Lihua Wang
- Department of Neurology, The Second Affiliated Hospital of Harbin Medical University, Harbin Medical University, Harbin, China
| |
Collapse
|
3
|
Tsiftsis D, Manioti EA, Touris G, Kyriakakis E, Tsamopoulos N, Gamvroudi M. Detecting Stroke at the Emergency Department by a Point of Care Device: A Multicenter Feasibility Study. Med Devices (Auckl) 2024; 17:107-112. [PMID: 38476129 PMCID: PMC10928921 DOI: 10.2147/mder.s445075] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/14/2023] [Accepted: 02/28/2024] [Indexed: 03/14/2024] Open
Abstract
Purpose To evaluate if the Strokefinder MD 100 by Medfield Diagnostics AB can be used as a point of care device in overcrowded Emergency Departments (ED). Patients and Methods We used the strokefinder MD 100 by Medfield Diagnostics AB in two Greek National Health System (NHS) Hospitals Emergency Departments. Our research protocol was approved by local scientific and ethics committees. We prospectively enrolled 71 adult patients from two NHS emergency departments in whom stroke was included as a differential diagnosis after triage. The feasibility of using the Strokefinder MD 100 by Medfield Diagnostics AB in various emergency department settings was evaluated through a structured questionnaire. Results The strokefinder MD 100 was used on 71 patients in various settings in the Emergency Department. In every case, the test was completed at the patient bedside without interfering with other ongoing and diagnostic and resuscitation procedures. There was no additional delay to patient care caused by performing the test when compared with current local Emergency Department practice and protocol. In almost 90% of the cases, a clear result was produced by the device. Conclusion The Strokefinder MD 100 can be safely used as a point of care device by all trained healthcare professionals, in the most overcrowded emergency department, in various ED locations. MeSH terms Point of Care Systems, Cerebrovascular Stroke, Proof of Concept Study.
Collapse
Affiliation(s)
- Dimitrios Tsiftsis
- Emergency Department, “Agios Panteleimon” General Hospital, Nikaia, Greece
| | | | - Georgios Touris
- Emergency Department “Konstantopouleio - Patision” General Hospital, Nea Ionia, Greece
| | | | | | - Maria Gamvroudi
- Emergency Department “Konstantopouleio - Patision” General Hospital, Nea Ionia, Greece
| |
Collapse
|
4
|
Chang C, Wang Y, Wang R, Bao X. Considering Context-Specific microRNAs in Ischemic Stroke with Three "W": Where, When, and What. Mol Neurobiol 2024:10.1007/s12035-024-04051-5. [PMID: 38381296 DOI: 10.1007/s12035-024-04051-5] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/22/2022] [Accepted: 02/12/2024] [Indexed: 02/22/2024]
Abstract
MicroRNAs are short non-coding RNA molecules that function as critical regulators of various biological processes through negative regulation of gene expression post-transcriptionally. Recent studies have indicated that microRNAs are potential biomarkers for ischemic stroke. In this review, we first illustrate the pathogenesis of ischemic stroke and demonstrate the biogenesis and transportation of microRNAs from cells. We then discuss several promising microRNA biomarkers in ischemic stroke in a context-specific manner from three dimensions: biofluids selection for microRNA extraction (Where), the timing of sample collection after ischemic stroke onset (When), and the clinical application of the differential-expressed microRNAs during stroke pathophysiology (What). We show that microRNAs have the utilities in ischemic stroke diagnosis, risk stratification, subtype classification, prognosis prediction, and treatment response monitoring. However, there are also obstacles in microRNA biomarker research, and this review will discuss the possible ways to improve microRNA biomarkers. Overall, microRNAs have the potential to assist clinical treatment, and developing microRNA panels for clinical application is worthwhile.
Collapse
Affiliation(s)
- Chuheng Chang
- Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, 100730, China
- M.D. Program, Peking Union Medical College, Beijing, 100730, China
| | - Youyang Wang
- Department of General Practice (General Internal Medicine), Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, 100730, China
| | - Renzhi Wang
- Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, 100730, China
| | - Xinjie Bao
- Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, 100730, China.
| |
Collapse
|
5
|
Zheng PF, Chen LZ, Liu P, Pan HW, Fan WJ, Liu ZY. Identification of immune-related key genes in the peripheral blood of ischaemic stroke patients using a weighted gene coexpression network analysis and machine learning. J Transl Med 2022; 20:361. [PMID: 35962388 PMCID: PMC9373395 DOI: 10.1186/s12967-022-03562-w] [Citation(s) in RCA: 16] [Impact Index Per Article: 8.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/31/2022] [Accepted: 07/30/2022] [Indexed: 11/28/2022] Open
Abstract
Background The immune system plays a vital role in the pathological process of ischaemic stroke. However, the exact immune-related mechanism remains unclear. The current research aimed to identify immune-related key genes associated with ischaemic stroke. Methods CIBERSORT was utilized to reveal the immune cell infiltration pattern in ischaemic stroke patients. Meanwhile, a weighted gene coexpression network analysis (WGCNA) was utilized to identify meaningful modules significantly correlated with ischaemic stroke. The characteristic genes correlated with ischaemic stroke were identified by the following two machine learning methods: the support vector machine-recursive feature elimination (SVM-RFE) algorithm and least absolute shrinkage and selection operator (LASSO) logistic regression. Results The CIBERSORT results suggested that there was a decreased infiltration of naive CD4 T cells, CD8 T cells, resting mast cells and eosinophils and an increased infiltration of neutrophils, M0 macrophages and activated memory CD4 T cells in ischaemic stroke patients. Then, three significant modules (pink, brown and cyan) were identified to be significantly associated with ischaemic stroke. The gene enrichment analysis indicated that 519 genes in the above three modules were mainly involved in several inflammatory or immune-related signalling pathways and biological processes. Eight hub genes (ADM, ANXA3, CARD6, CPQ, SLC22A4, UBE2S, VIM and ZFP36) were revealed to be significantly correlated with ischaemic stroke by the LASSO logistic regression and SVM-RFE algorithm. The external validation combined with a RT‒qPCR analysis revealed that the expression levels of ADM, ANXA3, SLC22A4 and VIM were significantly increased in ischaemic stroke patients and that these key genes were positively associated with neutrophils and M0 macrophages and negatively correlated with CD8 T cells. The mean AUC value of ADM, ANXA3, SLC22A4 and VIM was 0.80, 0.87, 0.91 and 0.88 in the training set, 0.85, 0.77, 0.86 and 0.72 in the testing set and 0.87, 0.83, 0.88 and 0.91 in the validation samples, respectively. Conclusions These results suggest that the ADM, ANXA3, SLC22A4 and VIM genes are reliable serum markers for the diagnosis of ischaemic stroke and that immune cell infiltration plays a crucial role in the occurrence and development of ischaemic stroke. Supplementary Information The online version contains supplementary material available at 10.1186/s12967-022-03562-w.
Collapse
Affiliation(s)
- Peng-Fei Zheng
- Cardiology Department, Hunan Provincial People's Hospital, No.61 West Jiefang Road, Furong District, Changsha, 410000, Hunan, China.,Clinical Research Center for Heart Failure in Hunan Province, No.61 West Jiefang Road, Furong District, Changsha, 410000, Hunan, China.,Institute of Cardiovascular Epidemiology, Hunan Provincial People's Hospital, No.61 West Jiefang Road, Furong District, Changsha, 410000, Hunan, China
| | - Lu-Zhu Chen
- Department of Cardiology, The Central Hospital of ShaoYang, No.36 QianYuan Lane, Daxiang District, Shaoyang, 422000, Hunan, China
| | - Peng Liu
- Department of Cardiology, The Central Hospital of ShaoYang, No.36 QianYuan Lane, Daxiang District, Shaoyang, 422000, Hunan, China
| | - Hong Wei Pan
- Cardiology Department, Hunan Provincial People's Hospital, No.61 West Jiefang Road, Furong District, Changsha, 410000, Hunan, China.,Clinical Research Center for Heart Failure in Hunan Province, No.61 West Jiefang Road, Furong District, Changsha, 410000, Hunan, China.,Institute of Cardiovascular Epidemiology, Hunan Provincial People's Hospital, No.61 West Jiefang Road, Furong District, Changsha, 410000, Hunan, China
| | - Wen-Juan Fan
- Cardiology Department, Hunan Provincial People's Hospital, No.61 West Jiefang Road, Furong District, Changsha, 410000, Hunan, China. .,Clinical Research Center for Heart Failure in Hunan Province, No.61 West Jiefang Road, Furong District, Changsha, 410000, Hunan, China. .,Institute of Cardiovascular Epidemiology, Hunan Provincial People's Hospital, No.61 West Jiefang Road, Furong District, Changsha, 410000, Hunan, China.
| | - Zheng-Yu Liu
- Cardiology Department, Hunan Provincial People's Hospital, No.61 West Jiefang Road, Furong District, Changsha, 410000, Hunan, China. .,Clinical Research Center for Heart Failure in Hunan Province, No.61 West Jiefang Road, Furong District, Changsha, 410000, Hunan, China. .,Institute of Cardiovascular Epidemiology, Hunan Provincial People's Hospital, No.61 West Jiefang Road, Furong District, Changsha, 410000, Hunan, China.
| |
Collapse
|