scholarly journals Transfer Learning Approach for Classification of Histopathology Whole Slide Images

Sensors ◽  
2021 ◽  
Vol 21 (16) ◽  
pp. 5361
Author(s):  
Shakil Ahmed ◽  
Asadullah Shaikh ◽  
Hani Alshahrani ◽  
Abdullah Alghamdi ◽  
Mesfer Alrizq ◽  
...  

The classification of whole slide images (WSIs) provides physicians with an accurate analysis of diseases and also helps them to treat patients effectively. The classification can be linked to further detailed analysis and diagnosis. Deep learning (DL) has made significant advances in the medical industry, including the use of magnetic resonance imaging (MRI) scans, computerized tomography (CT) scans, and electrocardiograms (ECGs) to detect life-threatening diseases, including heart disease, cancer, and brain tumors. However, more advancement in the field of pathology is needed, but the main hurdle causing the slow progress is the shortage of large-labeled datasets of histopathology images to train the models. The Kimia Path24 dataset was particularly created for the classification and retrieval of histopathology images. It contains 23,916 histopathology patches with 24 tissue texture classes. A transfer learning-based framework is proposed and evaluated on two famous DL models, Inception-V3 and VGG-16. To improve the productivity of Inception-V3 and VGG-16, we used their pre-trained weights and concatenated these with an image vector, which is used as input for the training of the same architecture. Experiments show that the proposed innovation improves the accuracy of both famous models. The patch-to-scan accuracy of VGG-16 is improved from 0.65 to 0.77, and for the Inception-V3, it is improved from 0.74 to 0.79.

Pain Practice ◽  
2021 ◽  
Author(s):  
Marco Reining ◽  
Dirk Winkler ◽  
Joachim Boettcher ◽  
Juergen Meixensberger ◽  
Michael Kretzschmar

Author(s):  
Cheng-Yi Wang ◽  
Wei-Chou Chang ◽  
Hsin-Hung Huang ◽  
Wei-Kuo Chang ◽  
Yu-Lueng Shih ◽  
...  

Objective: Not all endoscopic clips are compatible with magnetic resonance imaging (MRI). The aim of this study is to investigate the safety of MRI-incompatible endoscopic clips in patients undergoing MRI scans. Methods: We retrospectively reviewed the medical records of patients who had received endoscopic clip placement of Olympus Long Clip MRI-incompatible clips and then had undergone MRI scans within two weeks in our hospital between 2014 and 2019. Results: A total of 44,292 patients had undergone an MRI examination at our hospital. Only 15 patients had MRI scans within two weeks after the endoscopic clip placement. Their median age was 65.5 years, and 12 of the 15 patients were men. At the time of the clip placement and MRI scan, four patients were taking anti-coagulation or anti-platelet agents. The indication for endoscopic clip placement of the 15 patients was mucosal/submucosal defect or hemorrhage and colonic perforation. Endoscopic clips were placed in the colon of 14 patients and in the stomach of only one patient for gastric hemorrhage. One patient experienced clip migration and three displayed artifacts in abdominal images. No patient complications of mortality, hemorrhage, or organ perforation occurred. Conclusion: No serious adverse event occurred during MRI scans of patients with MRI-incompatible clips in this study, suggesting that MRI-incompatible clips may be safe to use in MRI scans. However, this does not guarantee the safety of the Long Clip for MRI scans, as further tests are needed to verify that this clip is safe for use during MRI.


Hand ◽  
2022 ◽  
pp. 155894472110643
Author(s):  
Jimmy H. Daruwalla ◽  
Jan Skrok ◽  
Mitchell A. Pet ◽  
Aviram M. Giladi ◽  
James P. Higgins

Background: The medial femoral trochlea (MFT) osteochondral flap is employed for reconstruction of unsalvageable scaphoid proximal pole nonunions. The convex surface of the cartilage-bearing proximal trochlea is used to replace the similarly contoured proximal scaphoid and articulate with the concave scaphoid fossa of the radius. A magnetic resonance imaging (MRI) comparison of the shape of the MFT as it relates to the native proximal scaphoid has not been previously performed. Our study aimed to quantifiably compare the shape of the MFT, proximal scaphoid, and scaphoid fossa. Methods: Using imaging processing software, we measured radius of curvature of the articular segments in MRI scans of 10 healthy subjects’ wrists and knees. Results: Compared with the scaphoid fossa, average ratio of the radius of circumference of the proximal scaphoid was 0.79 and 0.78 in the coronal and sagittal planes, respectively. Compared with the scaphoid fossa, average ratio of the radius of circumference of the MFT was 0.98 and 1.31 in the coronal and sagittal planes, respectively. The radius of curvature of the MFT was larger than the proximal scaphoid, in the coronal and sagittal planes. In the coronal plane, the MFT radius of curvature is nearly identical to the scaphoid fossa, a closer match than the scaphoid itself. In the sagittal plane, the radius of curvature of the MFT was larger than the radius of curvature of the scaphoid fossa. Conclusions: Our data suggest that the radius of curvature, in the sagittal and coronal planes, of the MFT and proximal scaphoid is disparate.


Author(s):  
Deanna C. Easley ◽  
Prahlad G. Menon ◽  
Pamela A. Moalli ◽  
Steven D. Abramowitch

Pelvic floor disorders such as Pelvic Organ Prolapse (POP) negatively impact the health and quality of life of millions of women worldwide. POP is characterized by the descent of the pelvic organs into the vagina due to compromised connective tissue support, resulting in discomfort and urinary/fecal incontinence. Magnetic Resonance Imaging (MRI) has been used to aid in the quantification of these anatomical changes, however the inter- and intra-observer repeatability necessary to make reliable conclusions about changes in anatomical positioning is questioned using current methods. The aim of this study was to quantify the degree of variability produced from inter-observer manual tracings of the vagina from MRI scans using a statistical shape matching approach.


Author(s):  
Robert Laureno

This chapter on “Imaging” examines the relative advantages and disadvantages of computed tomography (CT) and magnetic resonance imaging (MRI) scans. It compares the modalities to each other and to gross neuropathology. For several decades, neurologists have been able to view cross-sectional images of living patients. Analogous to gross neuropathology, cross-sectional imaging displays the brain as an entire organ but does not demonstrate microscopic tissue or cellular pathology. By allowing practitioners to view sections of brain and spinal cord in vivo, imaging has improved neurologic practice and facilitated clinical research. This chapter deals with imaging topics that are important to the neurologist. The timing of scans, the effects of gravity, and the importance of plane of section are considered. Imaging is compared to gross neuropathology, and MRI is compared to CT.


CNS Spectrums ◽  
2002 ◽  
Vol 7 (2) ◽  
pp. 155-158 ◽  
Author(s):  
Matcheri S. Keshavan ◽  
Perambur N. Jayakumar ◽  
Vaibhav A. Diwadkar ◽  
Amitabh Singh

ABSTRACTStudies on schizophrenia (SZ) have documented an increased presence of cavum septi pellucidi (CSP) in individuals suffering from the illness. Moreover, the presence of CSP has been cited in support of the early neurodevelopmental hypothesis in SZ. Our objective was to assess the magnetic resonance imaging (MRI) scans of first-episode patients and healthy controls to evaluate the frequency of CSP. The presence and the size of CSP were visually assessed on the MRI scans of 40 first-episode SZ patients, 19 nonpsychotic child and high-risk adolescent offspring of patients with SZ or schizoaffective disorder, and 59 controls. Our analysis revealed an absence of statistically significant differences in the occurrence of CSP between SZ patients, high-risk subjects, and controls. Even when the analysis was restricted to large CSP, no differences were found. Furthermore, no association between CSP and sex or handedness was observed. The absence of CSP abnormalities in first-episode SZ subjects might indicate that SZ is not characterized by developmentally mediated alterations in CSP. Also, family history of SZ might not increase likelihood for CSP.


2008 ◽  
Vol 90 (1) ◽  
pp. 25-28 ◽  
Author(s):  
S Blyth ◽  
A Blakeborough ◽  
M Peterson ◽  
IC Cameron ◽  
AW Majeed

INTRODUCTION The aim of this study was to evaluate the sensitivity of magnetic resonance imaging (MRI) in the detection of colorectal liver metastases. PATIENTS AND METHODS Pre-operative MRI scanning of the liver was performed by a single radiologist and the size and number of definite liver metastases were recorded. Patients then underwent hepatectomy with routine intra-operative ultrasonography (IOUS) and resected specimens were sent for histopathology. Pathology findings were compared with those of MRI scans to determine the sensitivity of this imaging modality. Exclusions were patients undergoing hepatic resection more than 4 weeks after the MRI scan, those undergoing chemotherapy at the time of the scan, and those with conglomerate unilobar metastases. RESULTS Complete data were available for 84 patients. There was total agreement between MRI, IOUS and histology in 79 patients (101 metastases). MRI missed 5 metastases in 5 patients that were found on IOUS (or palpation of superficial lesions) and subsequently confirmed by histological examination. These measured 5 mm or less (4 patients) and 7 mm (one patient). The sensitivity of MRI in the detection of colorectal liver metastases was thus 94% for all lesions and 100% for lesions 1 cm or larger in diameter. CONCLUSIONS MRI of the liver is a non-invasive technique with an extremely high degree of sensitivity in the detection of colorectal liver metastases and should be considered as the ‘gold standard’ in the pre-operative imaging of these patients.


Sign in / Sign up

Export Citation Format

Share Document